In recent years, due to the advancement of machine learning, object detection has become a mainstream task in the computer vision domain. The first phase of object detection is to find the regions where objects can exist. With the improvement of deep learning, traditional approaches such as sliding windows and manual feature selection techniques have been replaced with deep learning techniques. However, object detection algorithms face a problem when performing in low light, challenging weather, and crowded scenes like any other task. Such an environment is termed a challenging environment. This paper exploits pixel-level information to improve detection under challenging situations. To this end, we exploit the recently proposed hybrid task cascade network. This network works collaboratively with detection and segmentation heads at different cascade levels. We evaluate the proposed methods on three complex datasets of ExDark, CURE-TSD, and RESIDE and achieve an mAP of 0.71, 0.52, and 0.43, respectively. Our experimental results assert the efficacy of the proposed approach.

In all areas of medicine, especially in radiology, computers are increasing year by year. Filmless radiology, speech recognition software, electronic application forms, and teleradiology are recent developments that have greatly improved radiologists' performance. This research explores radiology software trends, predictions, and the challenges posed by informatics and historical trend analysis. The rationale behind this research is that information technology (IT) is overgrowing almost every day. We must continuously seek new ways to apply IT to make more use of resources. Consequently, IT becomes increasingly crucial to radiology organizations' innovative thinking, workflow, and business models. This study aimed to analyze all radiology software publications in the Science Citation Index (SCI). From 1991 to July 2021, SCI was used to search for publications systematically. We have also widely used this historical method in radiology software research. The findings and discussions are base on an assessment of trends, predictions, contributions, and challenges in radiology software, and we are exploring radiology software with six evolutionary stages. The gift of this research is that radiology managers realize that the use of new information technologies is closely related to survival in a competitive environment. Radiology companies can review these new technologies to develop more innovative business models and services to improve operational deficiencies.

An extension of the models used to generate fractal traffic flows is presented by means of the formulation of a model that considers the use of one-dimensional chaotic maps. Based on the disaggregation of the temporal series generated by the model, a valid explanation of behavior of the values of Hurst exponent is proposed and the feasibility of their control from the parameters of the proposed model is shown.

Autonomous driving systems are increasingly becoming a necessary trend towards building smart cities of the future. Numerous proposals have been presented in recent years to tackle particular aspects of the working pipeline towards creating a functional end-to-end system, such as object detection, tracking, path planning, sentiment or intent detection. Nevertheless, few efforts have been made to systematically compile all of these systems into a single proposal that effectively considers the real challenges these systems will have on the road, such as real-time computation, hardware capabilities, etc. This paper has reviewed various techniques towards proposing our own end-to-end autonomous vehicle system, considering the latest state on the art on computer vision, DSs, path planning, and parallelization.

Soybean maturity is a trait of critical importance for the development of new soybean cultivars, nevertheless, its characterization based on visual ratings has many challenges. Unmanned aerial vehicles (UAVs) imagery-based high-throughput phenotyping methodologies have been proposed as an alternative to the traditional visual ratings of pod senescence. However, the lack of scalable and accurate methods to extract the desired information from the images remains a significant bottleneck in breeding programs. The objective of this study was to develop an image-based high-throughput phenotyping system for evaluating soybean maturity in breeding programs. Images were acquired twice a week, starting when the earlier lines began maturation until the latest ones were mature. Two complementary convolutional neural networks (CNN) were developed to predict the maturity date. The first using a single date and the second using the five best image dates identified by the first model. The proposed CNN architecture was validated using more than 15,000 ground truth observations from five trials, including data from three growing seasons and two countries. The trained model showed good generalization capability with a root mean squared error lower than two days in four out of five trials. Four methods of estimating prediction uncertainty showed potential at identifying different sources of errors in the maturity date predictions. The architecture used solves limitations of previous research and can be used at scale in commercial breeding programs.

The use of commodity polymers such as polypropylene (PP) is key to open new market segments and applications for the additive manufacturing industry. Technologies such as powder-bed fusion (PBF) can process PP powder; however, much is still to learn concerning process parameters for reliable manufacturing. This study focusses in the process-property relationships of PP using laser-based PBF. The research presents an overview of the intrinsic and the extrinsic characteristic of a commercial PP powder as well as fabrication of tensile specimens with varying process parameters to characterize tensile, elongation at break, and porosity properties. The impact of key process parameters, such as power and scanning speed are systematically modified in a controlled design of experiment. The results were compared to the existing body of knowledge; the outcome is to present a process window and optimal process parameters for industrial use of PP. The computer tomography data revealed a highly porous structure inside specimens ranging between 8.46% and 10.08%, with porosity concentrated in the interlayer planes in the build direction. The results of the design of experiment for this commercial material show a narrow window of 0.122 ≥ Ev ≥ 0.138 J/mm³ led to increased mechanical properties while maintaining geometrical stability.

Non-invasive, electroencephalography (EEG)-based brain-computer interfaces (BCIs) on motor imagery movements translate the subject’s motor intention into control signals through classifying the EEG patterns caused by different imagination tasks, e.g. hand movements. This type of BCI has been widely studied and used as an alternative mode of communication and environmental control for disabled patients, such as those suffering from a brainstem stroke or a spinal cord injury (SCI). Notwithstanding the success of traditional machine learning methods in classifying EEG signals, these methods still rely on hand-crafted features. The extraction of such features is a difficult task due to the high non-stationarity of EEG signals, which is a major cause for the stagnating progress in classification performance. Remarkable advances in deep learning methods allow end-to-end learning without any feature engineering, which could benefit BCI motor imagery applications. We developed three deep learning models: 1) a long short-term memory (LSTM); 2) a proposed spectrogram-based convolutional neural network model (pCNN); and 3) a recurrent convolutional neural network (RCNN), for decoding motor imagery movements directly from raw EEG signals without (manual) feature engineering. Results were evaluated on our own, publicly available, EEG data collected from 20 subjects and on an existing dataset known as 2b EEG dataset from "BCI Competition IV". Overall, better classification performance was achieved with deep learning models compared to state-of-the art machine learning techniques, which could chart a route ahead for developing new robust techniques for EEG signal decoding. We underpin this point by demonstrating the successful real-time control of a robotic arm using our CNN based BCI.

Vibration measurement serves as the basis for various engineering practices such as natural frequency or resonant frequency estimation. As image acquisition devices become cheaper and faster, vibration measurement and frequency estimation through image sequence analysis continue to receive increasing attention. In the conventional photogrammetry and optical methods of frequency measurement, vibration signals are first extracted before implementing the vibration frequency analysis algorithm. In this work, we demonstrated that frequency prediction can be achieved using a single feed-forward convolutional neural network. The proposed method is verified using a vibration signal generator and excitation system, and the result obtained was compared with that of an industrial contact vibrometer in a real application. Our experimental results demonstrate that the proposed method can achieve acceptable prediction accuracy even in unfavorable field conditions.

Based on a local causal model of the dynamics of curved discrete spacetime, a causal model of quantum field theory in curved discrete spacetime is described. At the elementary level, space(-time) is assumed to consists of interconnected space points. Each space point is connected to a small discrete set of neighbor space points. Density distribution of the space points and the lengths of the space point connections depend on the distance from the gravitational sources. This leads to curved spacetime in accordance with general relativity. Dynamics of spacetime (i.e., the emergence of space and the propagation of space changes) dynamically assigns "in-connections" and "out-connections" to the affected space points. Emergence and propagation of quantum fields (including particles) are mapped to the emergence and propagation of space changes by utilizing identical paths of in/out-connections. Compatibility with standard quantum field theory (QFT) requests the adjustment of the QFT techniques (e.g., Feynman diagrams, Feynman rules, creation/annihilation operators), which typically apply to three in/out connections, to n > 3 in/out connections. In addition, QFT computation in position space has to be adapted to a curved discrete space-time.

Based on a local causal model of the dynamics of curved discrete spacetime, a causal model of quantum field theory in curved discrete spacetime is described. At the elementary level, space(-time) is assumed to consists of interconnected space points. Each space point is connected to a small discrete set of neighbor space points. Density distribution of the space points and the lengths of the space point connections depend on the distance from the gravitational sources. This leads to curved spacetime in accordance with general relativity. Dynamics of spacetime (i.e., the emergence of space and the propagation of space changes) dynamically assigns "in-connections" and "out-connections" to the affected space points. Emergence and propagation of quantum fields (including particles) are mapped to the emergence and propagation of space changes by utilizing identical paths of in/out-connections. Compatibility with standard quantum field theory (QFT) requests the adjustment of the QFT techniques (e.g., Feynman diagrams, Feynman rules, creation/annihilation operators), which typically apply to three in/out connections, to n > 3 in/out connections. In addition, QFT computation in position space has to be adapted to a curved discrete space-time.

Crowd accident surveys have shown that regardless of the initial triggering factors, pedestrian fall behaviour is the most critical factor causing and aggravating crowd accidents in public traffic areas (PTAs). Application of pedestrian fall behaviour detection methods in PTAs is significant. Once deployed, it would prevent many pedestrians from losing life in crowed traffic areas accidents. However, most existing methods is still focused on medical assistance for the elderly. Therefore, this paper conducted bibliometric and content analyses, combining fall detection-related keywords to retrieve from internationally recognized literature databases and benchmark pedestrian behaviour datasets. Based on the analysis of the state-of-the-art (SOTA) achievements in fall detection methods, the fall detection methods were classified into different categories according to the research approach. Among them, it undertakes a comprehensive analysis of five predominant methods, namely computer vision, Internet of Things, smartphone, kinematic, and wearable device-based methods. Furthermore, the benchmark datasets including fall scenarios were introduced and compared. Finally, it provides a detailed discussion of existing fall detection methods, and possible future directions are identified considering application requirements in PTAs. This overview may help researchers understand the SOTA fall detection methods and devise new methodologies by improving and synthesizing the highlighted issues in PTAs.

Neuromorphic computing, a brain inspired non-Von Neumann computing system, addresses the challenges posed by the Moore’s law memory wall phenomenon. It has the capability to increasingly enhance performance while maintaining power efficiency. Neuromorphic chip architecture requirements vary depending on the application and optimizing it for large-scale applications remains to be a challenge. Neuromorphic chips are programmed using spiking neural networks which provide them with important properties such as parallelism, asynchronism, and on-device learning. Widely used spiking neuron models include the Hodgkin-Huxley Model, Izhikevich model, integrate-and-fire model and spike response model. Hardware implementation platforms of the chip follow three approaches: analog, digital, or a combination of both. Each platform can be implemented using various memory topologies which interconnects with the learning mechanism. Current neuromorphic computing systems typically use the unsupervised learning spike timing-dependent plasticity algorithms. However, algorithms such as voltage-dependent synaptic plasticity have the potential to enhance performance. This review summarizes the potential neuromorphic chip architecture specifications and highlights which applications they are suitable for.

Cancer diagnosis and classification are pivotal for effective patient management and treatment planning. In this study, we present a comprehensive approach utilizing ensemble deep learning techniques to analyze breast cancer histopathology images. Our datasets are based on two widely employed datasets from different centers for two different tasks: BACH and BreakHis. Within the BACH Dataset, we deployed an ensemble strategy incorporating VGG16 and ResNet50 architec-tures to achieve precise classification of breast cancer histopathology images. Introducing a novel image patching technique, to preprocess a high-resolution image, which facilitates focused analysis of localized regions of interest. The annotated BACH dataset encompasses 400 WSIs across four distinct classes: Normal, Benign, In Situ Carcinoma, and Invasive Carcinoma. In addition, we employed the BreakHis dataset, utilizing VGG16, ResNet34, and ResNet50 models to classify mi-croscopic images into eight distinct categories (four benign and four malignant). For both models we leveraged a five-fold cross-validation approach for rigorous training and testing. Preliminary ex-perimental results indicate a Patch classification accuracy of 95.31% (on BACH dataset) and WSI image classification accuracy of 98.43% (BreakHis). This research significantly contributes to on-going endeavors in harnessing artificial intelligence to advance breast cancer diagnosis, potentially fostering improved patient outcomes and alleviating healthcare burdens.

Skin cancer is one of the widespread diseases that typically develop on the skin due to continuous exposure to sunlight. Although cancer can appear on any part of the human body, skin cancer reports account for over half of all cancer occurrences worldwide. There are substantial obstacles to the precise diagnosis and classification of skin lesions because of the morphological variety and indistinguishable characteristics across skin malignancies. Recently, Deep Learning models have been used in the field of image-based lesion diagnosis, and it has demonstrated diagnostic efficiency on par with that of dermatologists. To increase classification efficiency and accuracy for skin lesions, a cutting-edge multi-layer deep Convolutional Neural Network (CNN) termed SkinLesNet has been built in this study. The ResNetV50 and VGG16 models have been carefully compared to review the performance of the proposed model. The dataset used in this study, PAD-UFES-20, contains 1314 samples in total and includes three common forms of skin lesions. The proposed approach, SkinLesNet, significantly outperforms the well-known compared models in the given conditions

Deep Neutral Networks (DNNs) were initially proposed towards the midst of the 20th century, motivated by the neural structure and mechanism of the human brain. Thanks to major ad-vancements in computational resources, during the past decade DNN-based systems have demonstrated unprecedented performance in terms of accuracy and speed. However, recent work has shown that such models may not be sufficiently robust during the inference process. Furthermore, due to the data-driven learning nature of DNNs, designing interpretable and gen-eralizable networks is a major challenge, especially in critical applications, such as medical Computer Aided Diagnostics (CAD) and other medical imaging tasks, including classification, regression and reconstruction. Within this context, a line of physics-driven approaches for deep learning has recently emerged, aimed at improving the stability and generalization capacity of DNNs for medical imaging applications. In this paper, we review recent work focused on phys-ics-driven or prior-information learning for a variety of imaging modalities and medical applica-tions. We discuss how the inclusion of domain-related knowledge into the learning process and networks’ design supports their stability and generalization capability. In addition, we highlight current and future challenges within this scope.

In autonomous driving systems, high-speed and real-time image processing, along with object recognition, are crucial technologies. This paper builds upon the research achievements in in-dustrial item sorting systems and proposes an object recognition and sorting system for auton-omous driving. In industrial sorting lines, goods sorting robots often need to work at high speeds to efficiently sort large volumes of items. This poses a challenge to the robot's real-time vision and sorting capabilities, making it both practical and economically viable to implement a real-time and low-cost sorting system in a real-world industrial sorting line. Existing sorting systems have limitations such as high cost, high computing resource consumption, and high power consump-tion. These issues lead to the fact that existing sorting systems are typically used only in large industrial plants. In this paper, we design a high-speed, low-cost, low-resource-consumption FPGA (Field-Programmable Gate Array) based item sorting system that achieves similar perfor-mance to current mainstream sorting systems at a lower cost and consumption than existing sorting systems. The recognition part employs a morphological recognition method, which segments the image using a frame difference algorithm and then extracts the color and shape features of the items. To handle sorting, a six-degree-of-freedom robotic arm is introduced in the sorting segment. The improved cubic B-spline interpolation algorithm is employed to plan the motion trajectory and consequently control the robotic arm to execute the corresponding actions. Through a series of experiments, this system achieves an average recognition delay of 25.26ms, ensures smooth operation of the gripping motion trajectory, minimizes resource consumption, and reduces implementation costs.

Computer and Smart Device-assisted Learning (CSDL) has gained increasing attention from educational researchers and practitioners in recent years. However, it remains controversial whether student can benefit from CSDL and what moderator could affect the impact of CSDL. Within the specific context of Korea, where the interest in digital education is steadily increasing, the number of empirical studies exploring the causal effect of CSDL remains relatively scarce. The primary objective of this empirical study was to investigate the impact of CSDL on students' academic achievement in Korea. To achieve this objective, a two-way fixed effect model was employed, utilizing a panel dataset spanning three years derived from the 'Korean Education Longitudinal Study 2013'. The findings revealed a significant positive impact of CSDL on students’ Mathematics achievement. Notably, higher income levels, increased availability of computer resources provided by schools, and the implementation of more individualized education were identified as factors that moderate the effect of CSDL on students' achievement levels in Korean and English subjects. These findings underscore the need for an approach that optimizes the educational benefits of CSDL by considering subject-specific characteristics. Furthermore, this study highlights the importance of allocating educational resources, such as computers and smart devices, and integrating individualized educational activities within the classroom environment.

Uncontrolled and fast cell proliferation is the cause of brain tumors. Early cancer detection is vitally important in order to save many lives. Brain tumors can be divided into several categories depending on the kind, place of origin, pace of development, and stage of progression; as a result, tumor classification is crucial for targeted therapy. The aim of brain tumor segmentation is to accurately delineate the areas of the brain. A specialist with a thorough understanding of brain illnesses must manually identify the proper type of brain tumor. Additionally, processing a lot of images takes time and is tiresome. Therefore, automatic segmentation and classification techniques are required to speed up and enhance the diagnosis of brain tumors. Tumors can be quickly and safely detected by brain scans using imaging modalities, including computed tomography (CT), magnetic resonance imaging (MRI), and others. Machine learning (ML) and artificial intelligence (AI) have shown promise in the development of algorithms that aid in automatic classification and segmentation utilizing various imaging modalities. This review discussed various types of brain tumors, publicly accessible datasets, enhancement methods, segmentation, feature extraction, classification, machine learning techniques, and deep learning, learning through transfer, for the study of brain tumors.

This comprehensive review unfolds a detailed narrative of Artificial Intelligence (AI) making its foray into radiology, a move that is catalysing transformational shifts in the healthcare landscape. It sheds light on the journey of radiology, from the pioneering discovery of X-rays to today’s intricate imaging technologies, infused with machine learning and deep learning in medical image analysis. At the crux of this review lies an in-depth study of AI applications in radiology, elucidating its seminal roles in image segmentation, computer-aided diagnosis, predictive analytics, and workflow optimisation. A spotlight is cast on the profound impact of AI on diagnostic processes, personalised medicine, and clinical workflows, with empirical evidence derived from a series of case studies across multiple medical disciplines. However, the integration of AI in radiology is not devoid of challenges. The review ventures into the labyrinth of obstacles that are inherent to AI-driven radiology — data quality, the ’black box’ enigma, infrastructural and technical complexities, as well as ethical implications. Peering into the future, the review contends that the road ahead for AI in radiology is paved with promising opportunities. It advocates for continuous research, embracing avant-garde imaging technologies, and fostering robust collaborations between radiologists and AI developers. It concludes by firmly cementing the role of AI as a catalyst for change in radiology, a stance that is firmly rooted in sustained innovation, dynamic partnerships, and a steadfast commitment to ethical responsibility.

The prompt and accurate diagnosis of breast lesions, including the distinction between cancer, non-cancer, and suspicious cancer, plays a crucial role in the prognosis of breast cancer. In this paper, we introduce a novel method based on feature extraction and reduction for detection of breast cancer in mammography images. First, we extract features from multiple pre-trained convolutional neural network (CNN) models, and then concatenate them. The most informative features are selected based on their mutual information with the target variable. Subsequently, the selected features can be classified using a machine learning algorithm. We evaluate our approach using four different machine learning algorithms, and our results demonstrate that the neural network-based classifier yields an accuracy as high as 92% for the RSNA dataset which is a new dataset that provides two views and additional features such as age. We compare our proposed algorithm with state-of-the-art methods and demonstrate its superiority, particularly in terms of accuracy and sensitivity. For the MIAS dataset, we achieve an accuracy as high as 94.5%, and for the DDSM dataset, an accuracy of 96% is attained. These results highlight the effectiveness of our method in accurately diagnosing breast lesions and surpassing existing approaches.

Geometrical assessments of human skull have been done based on the anatomical landmarks. Automatic detection of the landmarks, if developed, will be a great help not only medically, but also anthropologically. An automated system with multi-phased deep learning networks to predict three-dimensional coordinate values of cranio-facial landmarks, was developed. From a publicly available database, computed tomography images of craniofacial area were obtained. They were digitally reconstructed into three-dimensional models. Sixteen anatomical landmarks were plotted on each of the models and coordinate values of them were recorded. Three-phased regression deep learning networks were trained respectively with 90 training datasets. For evaluation, 30 testing datasets were employed. Three-dimensional error for the first phase, testing 30 data, was 11.60 pixels in average. (1 pixel = 500 / 512 mm) For the second phase, it was significantly improved to 4.66 pixels. For the third phase, it was significantly progressed to 2.88. This was comparable to the gaps between the landmarks, plotted by two experienced practitioners. Our proposing method of multi-phased prediction, coarse detection first and narrowing down the detection area, may be a possible solution, within the physical limitation of memory and computation.

Deepfakes are manipulated or altered images, or video, that are created using deep learning models with high levels of photorealism. The two popular methods of producing a deepfake are based on either convolutional neural networks (CNN), or autoencoders. Deepfakes created using CNN comparatively show higher qualities of realism, yet oftentimes leave artifacts and distortions in the generated media that can be detected using machine learning and deep learning algorithms. In recent years, there has been an influx of periocular image and video data because of the increase usage of face masks. By wearing masks, much of what is used for facial recognition is hidden, leaving only the periocular region visible to an observer. This loss of vital information leads to easier misidentification of media, allowing deepfakes to less likely be identified as fake. In this work, feature extraction methods, such as Scale-Invariant Feature Transform (SIFT), Histogram of Oriented Gradients (HOG), and CNN, are used to train an ensemble deep learning model to detect deepfakes in videos on a frame-by-frame level based on the periocular region. Our proposed model is able to distinguish original and manipulated images with accuracies around 98.9 percent, which is an improvement to previous works by combining SIFT and HOG for deepfake detection in convolutional neural networks.

Sustainability is the core concern of every business; The exploration of avenues to maintain sustainability while staying competitive with high level of productivity remains a vital endeavor. Production flexibility is a key area that can enhance the sustainability of manufacturing industries as it can ensure product availability, scalability, agility/fault tolerance as well as disaster recovery potentials. Technological advancements have provided avenues where companies can enhance virtually all aspects of their operations for efficiency, effectiveness, and productivity. This paper uses both quantitative and qualitative research approach to identify the capability requirements for smart and effective production management and subsequent analysis is done using Multi-Criteria Decision-Making methodology to identify and rank various industry 4.0 technologies and concepts that can provide these smart capabilities in in manufacturing industries to aid the businesses to achieve sustainability with production flexibility. The paper identifies over 12 smart capabilities and 9 Industry 4.0 Technologies which are applicable to production management. It also compares results from the analytics of historical I4.0 implementation as discussed in literatures with the current state as deduces from survey feedbacks from various manufacturing industries.

In this paper, we investigate the relationship people have with their smart devices. We use the concept of agency to capture aspects of users’ sense of mastery as they relate to their device. This study gives preliminary evidence of the existence of two independent dimensions of agency for modeling the interaction between humans and smart devices: (i) user agency and (ii) device agency. These constructs emerged from an exploratory factorial analysis conducted on a survey data collected from 587 participants. In addition, we investigate the correlation between user agency and device agency with background variables of the respondents. Finally, we argue that mapping the users’ dynamics with their device into user agency and device agency fosters a better understanding of the needs of the users and helps in designing interfaces tailored for the specific capabilities and expectations of the users.

Cognitive deterioration caused by illness or aging often occurs before symptoms arise, and their timely diagnosis is crucial to reducing its medical, personal, and societal impacts. Brain-Computer Interfaces (BCIs) stimulate and analyze key cerebral rhythms, enabling reliable cognitive assessment that can accelerate diagnosis. The BCI system presented analyzes Steady-State Visually Evoked Potentials (SSVEPs) elicited in subjects of varying age to detect cognitive aging, predict its magnitude, and identify its relationship with SSVEP features (band power and frequency detection accuracy), which were hypothesized to indicate cognitive decline due to aging. The BCI system was tested with subjects of varying age to assess its ability to detect aging-induced cognitive deterioration. Rectangular stimuli flickering at theta, alpha, and beta frequencies were presented to subjects, and frontal and occipital EEG responses were recorded. These were processed to calculate detection accuracy for each subject and calculate SSVEP band power. A neural network was trained using the features to predict cognitive age. The results showed potential cognitive deterioration through age-related variations in SSVEP features. Frequency detection accuracy declined after age group 20–40 and band power, throughout all age groups. SSVEPs generated at theta and alpha frequencies, especially 7.5 Hz, were the best indicators of cognitive deterioration. Here, frequency detection accuracy consistently declined after age group 20-40 from an average of 96.64% to 69.23%. The presented system can be used as an effective diagnosis tool for age related cognitive decline.

Bell type inequalities are proven using oversimplified probabilistic models and/or counterfactual definiteness (CFD). If setting-dependent variables describing measuring instruments are correctly introduced none of these inequalities may be proven. In spite of this a belief in a mysterious quantum nonlocality is not fading. Computer simulations of Bell tests allow studying different scenarios how the experimental data might have been created. They allow also to generate outcomes of various counterfactual experiments such as repeated or simultaneous measurements performed in different settings on the same ‘’ photon-pair” etc. They allow reinforcing or relaxing CFD- compliance and /or to study the impact of various “photon identification procedures” mimicking those used in real experiments. Using a specific setting- dependent identification procedure data samples consistent with quantum predictions may be generated. It reflects an active role of instruments during the measurement process. Each setting dependent data samples are consistent with specific setting –dependent probabilistic models which may not be deduced using non-contextual local realistic or stochastic hidden variables. In this paper we discuss the results of these simulations. Since the data samples are generated in a locally causal way, these simulations provide additional strong arguments for closing the door on quantum nonlocality

Recent research has drawn attention to the ambiguity surrounding the definition and learnability of Out Of Distribution (OOD) recognition. The term "Out of Model Scope" (OMS) detection provides a clearer perspective, although the original problem remains unsolved. The ability to detect OMS inputs is particularly beneficial in safety-critical applications such as autonomous driving or medicine. By detecting OMS situations, we not only enhance the system's robustness but also prevent it from operating in unknown and unsafe scenarios. In this paper, we introduce a novel approach for OMS detection that combines three sources of information: 1) the original input, 2) the latent feature space extracted by an encoder, and 3) the synthetic data generated from the encoder's latent feature space. We demonstrate the effectiveness of combining original images and synthetically generated images against adversary attacks in the computer vision domain. We achieve results comparable to other state-of-the-art methods and demonstrate that any Encoder can be integrated into our pipeline in a plug-and-train fashion. Our experiments further evaluate which combination of the Encoder's features works best in order to discover OMS samples and highlight the importance of a compact feature space for the training of the Generator.

This work investigates into the enhancement of iris recognition systems through a two-module approach focusing on low level image preprocessing techniques and advanced feature extraction. The first module is dedicated to the preprocessing of iris images, leveraging the Canny algorithm for edge detection followed by the circle-based Hough transform for precise iris extraction. This process, integral to our methodology, ensures the quality and consistency necessary for the subsequent Binary Statistical Image Features (BSIF) analysis. The second module employs the BSIF technique, incorporating domain-specific filters trained on iris-specific data, for robust biometric identification. By combining these advanced image preprocessing techniques, the proposed method addresses key challenges in iris recognition, such as occlusions, varying pigmentation, and textural diversity. Experimental results on the Human-inspired Domain-specific Binarized Image Features (HDBIF) Dataset composed of 1892 iris images demonstrate the effectiveness of the developped approach, with potential future improvements including adaptive algorithms and machine learning integrations for enhanced performance in diverse and unpredictable real-world applications. The novel contributions of this paper lie in the provision of a reproducible research framework, making the iris-domain-specific BSIF filters, training patches, testing database, and source codes publicly available for further application and study.

This proposal advocates using energy states as fundamental units for quantum computer programming. Each programming unit is conceptualized as an energy state, manifested as an essence-unit. The essence-unit, defined as the minimal form encapsulating complete uniqueness or specificity, serves as the cornerstone of this paradigm. This approach allows recording various forms of matter on quantum computers in essence-units representing their energy states. Recording energy states is achieved through the creation of four distinct coherent potentials facilitated by quantum dots or crystals. It is crucial to note that these energy states, embodied in essence-units, cannot be subdivided. The study examines the intricate relationship between similarity, fractals, and uniqueness in quantum dot operations, elucidating their profound implications for information transfer efficiency. Normalized entropy quantifies the charge localization in quantum dot impurities in systems exhibiting distortions. Utilizing N-level recording and entropy-fractal dimension equivalence enables this paper to elucidate the potential of quantum dots in reducing transmission time and modeling complex systems. The proposed methodology represents a significant departure from the current state of quantum computing, offering unprecedented potential for overcoming previously intractable challenges.

This study presents a novel semi-supervised deep learning method utilizing 3D TOF-MRA images for the detection of brain aneurysms, employing landmark-based techniques that mimic radiologist practices to enhance detection accuracy and efficiency.

This study delves into the realm of heuristic evaluation within usability testing, focusing on refining and applying a tailored algorithm that simplifies the implementation of the Analytic Hierarchy Process for heuristic prioritization. Addressing the prevalent challenge of disparate evaluation methodologies and the absence of a standardized process in usability testing, our research introduces a novel method that leverages the Analytic Hierarchy Process alongside a bespoke algorithm. This algorithm utilizes transitive properties for pairwise comparisons, significantly reducing the evaluative workload. This innovative approach facilitates the estimation of heuristic relevance irrespective of the number of items per heuristic or the item scale and streamlines the overall evaluation process. In addition to rigorous simulation testing of the tailored algorithm, we applied our method in a practical scenario, engaging seven usability experts in evaluating a web interface. This empirical application underscored our method’s capacity to diminish the requisite comparison count and spotlight areas of improvement for critically weighted yet underperforming heuristics. The outcomes of this study underscore the efficacy of our approach in enhancing the efficiency and organization of heuristic evaluations, paving the way for more structured and effective usability testing methodologies in future research endeavors.

This papers objective is to describe a method of in-office 3D designing and 3D printing of a tooth-borne molar distalizer and create a library to easily reproduce the needed distalizers in a private orthodontic clinic. Research objectives were to design and 3D print molar distalizers, clinically use in orthodontic treatment settings, assess the strength and frequency of debonding and or breakage. The print resin used was Dental LT Clear V2 (RS-F2-DLCL-02) from formlabs. The 3D printer used was a Formlabs 3B+printer. 16 patients were treated with these 3D printed distalizers. Patients selected were between 11 Years and 49 Years old, Class II occlusion with no skeletal Class II values. The skeletal cephalometric values of the six patients were within the range of SNA = 81± 3º , SNB = 78 ± 3º , ANB = 3 ± 2º. The mean duration of the 3D printed appliance was 14.58 ± 4.31 weeks. The aim reached, was to position molars and canines in a dental Class I position. The combined failure rate was 0.94. A library of distalizers has been made of sizes between 16mm and 29mm, they are easy to print and easy to use in office.

Abstract: In recent years, low frequency (LF) electroencephalogram (EEG) signals have been decoded to obtain kinematic trajectories with the aim of achieving closed loop natural control. However, decoded trajectories suffer variability and low correlation against the measured movement. This paper suggests that if EEG features of motor intents are directly associated with fixed points along a given range of motion, then the natural kinematic parameters can be quantified. A novel classification method is proposed to evaluate this hypothesis along with equations for obtaining the movement parameters (velocity, acceleration). Furthermore, this implies that the force of an end-effector and work done in moving the end-effector may be computed, leading to natural control of EEG-based robotic arms and neuroprostheses. Finally, expectations are that combining the proposed method with existing classification methods will increase the accuracy and scalability of brain-computer interface (BCI) applications.

The cell concentration measurement on a culture assay using bioimpedance is a very useful tool, but it is complex to translate impedance to cell concentration values. The purpose of this work is to find a method to obtain in real time the cell concentration values for a given cell-culture assay using an oscillator as the measurement circuit. From a basic cell-electrode model, enhanced models of the cell culture immersed in a saline solution (culture medium) can be derived. These models can be used in a fitting routine to real time estimation of the cell concentration in a cell culture, using the oscillation frequency and amplitude delivered by measurement circuits proposed by the authors. Based on real experimental data (frequency and amplitude of oscillations) obtained by connecting the cell culture to an oscillator as a load, the fitting routine is simulated, and real time data of cell concentration is achieved. These results are compared with concentration data found by traditional optical methods for counting. In addition, the error obtained is divided and analyzed in two parts: in the first part of the experiment (when the few cells are adapting to the culture medium) and the second part of the experiment (when the cells grow exponentially until they completely cover the well).Low error values are obtained in the growth phase of the cell culture (the relevant phase), therefore the results obtained are considered to be promising, proving that the fitting routine is valid, and that the cell concentration can be measured in real time using an oscillator.

Lung cancer is the leading cause of cancer mortality worldwide, and it is urgently necessary to diagnose it as early as possible. Usually, the diagnostic process begins with a radiological examination which, when a possible tumour is present, is followed by a biopsy to extract tissue samples from the patient's lungs. Therefore, the purpose of this study is the development of an artificial intelligence algorithm that will analyse the Whole Slide Image (WSI) generated by the digitisation of the glass slides obtained from the extracted samples and detect if there is a tumour. The developed learning algorithms as well as the tested neural networks (NNs) were trained on a dataset composed of previously annotated WSI tiles, classified as Tumour or Non-Tumour. From these, four developed convolutional neural networks stood out and were selected to be compared with each other and with the tested NNs. When the best result of each of the developed architectures was compared to the highest result of the tested NNs, it was possible to denote that version 4 of CancerDetecNN achieved an average accuracy of 89.749 \% and an average loss of 0.220. Furthermore, the results for the four selected versions are in agreement with the results reported in the literature, however, the limited size of the given dataset must be considered. Given the results obtained, the fourth version has the potential to optimise the lung cancer diagnosis process.

Objective: Radiomics and deep transfer learning are two popular technologies used to develop computer-aided detection and diadnosis (CAD) schemes of medical images. This study aims to investigate and compare advantages and potential limitations of applying these two technologies in developing CAD schemes. Methods: A relatively large and diverse retrospective dataset including 3,000 digital mammograms is assembled in which 1,496 images depict malignant lesions and 1,054 images depict benign lesions. Two CAD schemes are developed to classify breast lesions. The first scheme is developed using four steps namely, applying an adaptive multi-layer topographic region growing algorithm to segment lesions, computing initial radiomics features, applying a principal component algorithm to generate an optimal feature vector, and building a support vector machine classifier. The second CAD scheme is built based a pre-trained residual net architecture (ResNet50) as a transfer learning model to classify breast lesions. Both CAD schemes are trained and tested using a 10-fold cross-validation method. Several score fusion methods are also investigated to classify breast lesions. CAD classification performance is evaluated by the area under ROC curve (AUC). Results: ResNet50 model-based CAD scheme yields AUC = 0.85±0.02, which is significantly higher than radiomics feature-based CAD scheme with AUC = 0.77±0.02 (p < 0.01). Additionally, fusion of classification scores generated by two CAD schemes does not further improve classification performance. Conclusion: This study indicates that using deep transfer learning is more efficient to develop CAD schemes and enables to yield higher lesion classification performance than CAD schemes developed using radiomics-based technology.

This study observes, interprets, and analyses the knowledge production in the research field of Human–Robot Interaction (HRI). It intends to foreground the hidden assumptions that are often taken for granted when roboticist design and conduct their research. By doing so, this study demonstrates how these assumptions influence the result of their research. Based on data collected through sociological field observation, this study argues that the current practise in HRI research is highly anthropocentric. In short, the robots are designed to be like human instead of for human. Therefore, the human–robot relationship embodies the existing power relations between human beings. These relations generate inequality, hierarchy, and dominance, which are the opposite of the common imagination of the robotised future among roboticists. For the purpose of enabling the robotised future closer to their ideal, this study suggests that HRI researchers to go beyond the conventional methodology, to allow a human–robot relationship that realises reciprocity.

This work presents an approach for detecting mathematical formulas in scanned document images. The proposed approach is end-to-end trainable. Since many OCR engines cannot reliably work with the formulas, it is essential to isolate them to obtain the clean text for information extraction from the document. Our proposed pipeline comprises a hybrid task cascade network with deformable convolutions and a Resnext101 backbone. Both of these modifications help in better detection. We evaluate the proposed approaches on the ICDAR-2017 POD and Marmot datasets and achieve an overall accuracy of 96% for the ICDAR-2017 POD dataset. We achieve an overall reduction of error of 13%. Furthermore, the results on Marmot datasets are improved for the isolated and embedded formulas. We achieved an accuracy of 98.78% for the isolated formula and 90.21% overall accuracy for embedded formulas. Consequently, it results in an error reduction rate of 43% for isolated and 17.9% for embedded formulas.

This paper proposes several LabVIEW applications to accomplish the data acquisition, processing, features extraction and real-time classification of the electroencephalographic (EEG) signal detected by the embedded sensor of the NeuroSky Mindwave Mobile headset. The LabVIEW applications are aimed at the implementation of a Brain-Computer Interface system, which is necessary to people with neuromotor disabilities. It is analyzed a novel approach regarding the preparation and automatic generation of the EEG dataset by identifying the most relevant multiple mixtures between selected EEG rhythms (both time and frequency domains of raw signal, delta, theta, alpha, beta, gamma) and extracted statistical features (mean, median, standard deviation, route mean square, Kurtosis coefficient and others). The acquired raw EEG signal is processed and segmented into temporal sequences corresponding to the detection of the multiple voluntary eye-blinks EEG patterns. The main LabVIEW application accomplished the optimal real-time artificial neural networks techniques for the classification of the EEG temporal sequences corresponding to the four states: 0 - No Eye-Blink Detected; 1 - One Eye-Blink Detected; 2 – Two Eye-Blinks Detected and 3 – Three Eye-Blinks Detected. Nevertheless, the application can be used to classify other EEG patterns corresponding to different cognitive tasks, since the whole functionality and working principle could estimate the labels associated with various classes.

Object detection is one of the most critical tasks in the field of Computer vision. This task comprises identifying and localizing an object in the image. Architectural floor plans represent the layout of buildings and apartments. The floor plans consist of walls, windows, stairs, and other furniture objects. While recognizing floor plan objects is straightforward for humans, automatically processing floor plans and recognizing objects is a challenging problem. In this work, we investigate the performance of the recently introduced Cascade Mask R-CNN network to solve object detection in floor plan images. Furthermore, we experimentally establish that deformable convolution works better than conventional convolutions in the proposed framework. Identifying objects in floor plan images is also challenging due to the variety of floor plans and different objects. We faced a problem in training our network because of the lack of publicly available datasets. Currently, available public datasets do not have enough images to train deep neural networks efficiently. We introduce SFPI, a novel synthetic floor plan dataset consisting of 10000 images to address this issue. Our proposed method conveniently surpasses the previous state-of-the-art results on the SESYD dataset and sets impressive baseline results on the proposed SFPI dataset. The dataset can be downloaded from SFPI Dataset Link. We believe that the novel dataset enables the researcher to enhance the research in this domain further.

This paper presents a novel architecture for detecting mathematical formulas in document images, which is an important step for reliable information extraction in several domains. Recently, Cascade Mask R-CNN networks have been introduced to solve object detection in computer vision. In this paper, we suggest a couple of modifications to the existing Cascade Mask R-CNN architecture: First, the proposed network uses deformable convolutions instead of conventional convolutions in the backbone network to spot areas of interest better. Second, it uses a dual backbone of ResNeXt-101, having composite connections at the parallel stages. Finally, our proposed network is end-to-end trainable. We evaluate the proposed approach on the ICDAR-2017 POD and Marmot datasets. The proposed approach demonstrates state-of-the-art performance on ICDAR-2017 POD at a higher IoU threshold with an f1-score of 0.917, reducing the relative error by 7.8%. Moreover, we accomplished correct detection accuracy of 81.3% on embedded formulas on the Marmot dataset, which results in a relative error reduction of 30%.

The objective of this study was aimed to study the sensory processes of the “human-computer interaction” model when classifying visual images with an incomplete set of signs based on the analysis of early, middle, late and slow components of event-related potentials (ERPs). 26 healthy subjects (men) aged 20-22 years were investigated. ERPs in 19 monopolar sites according to the 10/20 system were recorded. Discriminant and factor analysis were applied. The component N450 is the most specialized indicator of the perception of unrecognizable (oddball) visual images. The amplitude of the ultra-late components N750 and N900 is also higher under conditions of presentation of the oddball image, regardless of the location of the registration points. The results of the study are discussed in the light of the paradigm of the P300 wave application in brain-computer interface systems, as well as with the peculiarities in brain pathology. Promising directions for the development of studies of the “Brain Computer Interface” (BCI) P300 systems are to increase the throughput of information flows. To extend the application of the P300 ERPs to multiple modalities, the underlying physiological mechanisms and responses of the brain for a particular sensory system and mental function must be carefully examined.

Manual interpretation of these huge amounts of image volumes are susceptible to inter-reader variability and human error. Thus, accurate automated CAD scheme is highly desirable in clinical pathological diagnosis. In this research, plethora of machine learning paradigms (e.g. feature extraction, dimensionality reduction and supervised classification methods) were explored, evaluated, compared and analyzed to identify the optimal pathway for brain MR images (normal vs neoplastic) binary classification task. External validation dataset was used to test the generalizability of the optimal predictive models implemented. Relevant and informative features were selected to construct cross-validated decision tree and eventually simple rule set was built based on the decision tree. The experimental results show that almost all pattern recognition paradigms achieve high accuracy with careful selection of number of attributes. LDA+ELM with 55 features are the optimal pipelines which achieve perfect classification when training and test data are of same source; and achieving (accuracy=97.5%, AUC=0.989, sensitivity=95% and specificity=100%) under balanced test dataset; (accuracy=99.5%, AUC=0.988, sensitivity=95% and specificity=100%). Cross-validated decision tree model also shows comparable performance: accuracy=98.8%, AUC=99.1%, sensitivity=99.6% and specificity=98.2%. Three highly relevant and robust attributes are visualized and selected for construction of decision tree models and finally a rule sets are read directly off the decision tree. This rule sets can potentially serve as fast and accurate classification algorithm.

Augment reality (AR) is crucial for immersive human-computer interaction (HCI) and vision of artificial intelligence (AI). Labeled data drove object recognition in AR. However, manual annotating data is expensive and labor-intensive, and furthermore, scanty labeled data limits the application of AR. Aiming at solving the problem of insufficient training data in AR object recognition, an automated vision data synthesis method called BAGAN is proposed in this paper based on the 3D modeling and GAN algorithm. Our approach has been validated to have better performance than other methods through image recognition task on natural image database ObjectNet3D. This study can shorten the algorithm development time of AR and expand the application scope of AR, which is of great significance to immersive interactive systems.

In this paper we introduce a computational model for the simulation of hypervelocity impact (HVI) phenomena which is based on the Discrete Element Method (DEM). Our paper constitutes the first application of DEM to the modeling and simulating of impact events for velocities beyond 5 kms⁻¹. We present here the results of a systematic numerical study on HVI of solids. For modeling the solids, we use discrete spherical particles that interact with each other via potentials. In our numerical investigations we are particularly interested in the dynamics of material fragmentation upon impact. We model a typical HVI experiment configuration where a sphere strikes a thin plate and investigate the properties of the resulting debris cloud. We provide a quantitative computational analysis of the resulting debris cloud caused by impact and a comprehensive parameter study by varying key parameters of our model. We compare our findings from the simulations with recent HVI experiments performed at our institute. Our findings are that the DEM method leads to very stable, energy–conserving simulations of HVI scenarios that map the experimental setup where a sphere strikes a thin plate at hypervelocity speed. Our chosen interaction model works particularly well in the velocity range where the local stresses caused by impact shock waves markedly exceed the ultimate material strength.

Interaction between multiple phases especially liquid-gas is one of the most common phenomena in industrial systems. Computer Vision enables modern systems to derive an automated and robust image-processing technique to address this issue. However, diverse experimental conditions still make it a challenging task to overcome. In Deep Learning, instance segmentation has created a niche in various computer vision applications as it aims to provide different instance IDs to objects belonging to the same class. In this paper, we propose a novel approach employing the YOLO (You Look Only Once) an end-to-end deep learning architecture for the detection and segmentation of bubbles in fluid flows. YOLOv5 offers significant improvements in both speed and accuracy compared to its predecessors like YOLOv3. Therefore, we train the YOLOv5 network on a dataset consisting of annotated images representing instances of bubbles across varied fluid flow environments. This paper provides both qualitative and quantitative evaluation of three YOLOv5 networks (small, medium, and extra-large) in detecting and segmenting bubbles with high accuracy and efficiency. Based on the metrics all three YOLOv5 models achieved high precision above 0.98 for bounding box detection and 0.97 for segmentation mask, and high recall value greater than 0.97 for box and 0.96 for mask quality was achieved. The achieved results indicate that the models accurately localize bubbles within the image frame, and distinguish bubble pixels from the background. Also, mean average precision (mAP50-95) consistently exceeds 0.84 across all models for the bounding box and 0.73 for the segmentation mask. The ascertained results emphasize the potential of YOLOv5 architecture as a powerful tool for automated bubble analysis in real-world applications. Our research contributes to advancing the field of deep neural networks in fluid systems and provides a fresh approach for future developments in automated bubble analysis.

Brain-Computer Interface (BCI) systems can assist physically challenged people to interact with their surroundings and improve the quality of their lives. Decoding human thoughts is a powerful technique that can assist paralyzed people who have lost their speech production ability. Speaking is a combined process involving synchronizing the brain and the oral articulators. This paper proposed a high-accuracy brain wave pattern recognition based on inner speech using a novel feature extraction method. Only eight EEG electrodes were used in this study, and they were set on selected spots on the scalp. Support Vector Machine (SVM) was employed to decode the recorded EEG dataset into four internally spoken words which are: Up, Down, Left, and Right. The proposed approach achieved overall classification accuracy that ranges between 96.20% to 97.5%. In addition, more performance evaluation metrics were estimated to test the reliability of classifying the EEG-based inner speech data, and we obtained 97.61%, 97.50%, and 97.73% for F1-score, recall, and precision respectively. Furthermore, the Area Under Curve of the Receiver Operating Characteristic (AUC-ROC) proved the strength of the proposed approach for classifying the specified inner speech commands by achieving a macro-average amount of 99.32%. The inner speech classification method using electroencephalography proposed in this work can clinically help improve communication for patients with problems including speech disorder, mutism, cognitive development, executive function, and psychopathology.

OpenAI has managed to turn 100 million heads in two months towards their new Language model tool (LM) ChatGPT. The third generation Generative Pretrained Transformer (GPT-3) has the capacity to tackle simple to complex and sophisticated problems, while providing reasoning behind its generated answers. ChatGPT can be used to increase productivity and improve efficiency and face challenging problems. Programming mobile applications is a challenging task that requires professional software engineers, skills and abilities to be developed. The following paper takes a case study approach to assess how novice app developers can use ChatGPT to generate Java scripts that will be used in Android studio to create a functional application. The results after, many iterations and ongoing conversations with ChatGPT managed to create an application for the anticipated function. Important insights have been drawn from the case study that could set the ground for any novice user seeking to create applications using Java scripting and Android Studio.

Tracking and early identification of suspected cases are essential to control and prevent potential COVID-19 outbreaks. One of the most popular techniques used to track this disease is the use of Infrared cameras to identify individuals with elevated body temperatures. However, they are limited by their inability to be implemented in open public settings such as public parks or even outdoor recreational centers. This limits their ability to effectively track possible COVID-19 patients as open public recreational places such as parks, concert venues and other public venues are hotspots for the spreading of the virus. Other technological solutions such as thermal scanners require an individual to perform the actual testing as they are not individual standalone technologies. This method of testing can potentially cause the transmission of the virus between the tester and the individual getting tested. As can be seen, an alternative solution is essential to solving this issue. In this study, we aim to present the system, design and potential scope of a non-invasive system that can diagnose and identify potential COVID-19 patients using thermal and optical images of the individual using drone technology. The proposed system (COVIDRONE) combines multi-modal machine intelligence, computer vision and real-time monitoring to enable scalable monitoring. The system will also involve the use of machine learning algorithms for better and more accurate diagnosis. We envisage that development of such technologies may help in developing technological solutions to combat infectious disease threats in the future pandemics.

Fall prediction using machine learning has become one of the most fruitful and socially relevant applications of computer vision in gerontological research. Since its inception in the early 2000s, this subfield has proliferated into a robust body of research underpinned by various machine learning algorithms (including neural networks, support vector machines, and decision trees) as well as statistical modeling approaches (Markov chains, Gaussian mixture models, and hidden Markov models). Furthermore, some advancements have been translated into commercial and clinical practice, with companies in various stages of development capitalizing on the aging population to develop new commercially available products. Yet despite the marvel of modern machine learning-enabled fall prediction, little research has been conducted to shed light on the security and privacy concerns that such systems pose for older adults. The present study employs an interdisciplinary lens in examining privacy issues associated with machine learning fall prediction and exploring the implications of these models in elderly care and the Internet of Medical Things (IoMT). Ultimately, a justice-informed set of best practices rooted in social geroscience is suggested to help fall prediction researchers and companies continue to advance the field while preserving elderly privacy and autonomy.

Cells receive external stimuli to incur structural and functional damages. On scanning acoustic microscopy (SAM), speed-of-sound (SOS), attenuation-of-sound (AOS), and thickness values are plotted on the screen to create cellular images, which are related to stiffness, viscosity, and cell size, respectively. The obtained digital data compared using statistical analysis. We aimed to investigate the effects of anticancer drugs, acidic fluids, and heat effects on the cells by using SAM. Anticancer drug cisplatin induced cancer cell apoptosis/necrosis and regeneration in culture, causing elevated SOS, reduced AOS, and thickness. During a more prolonged incubation, the SAM values fluctuated differently between the cisplatin-treated and untreated cells. The tannic and acetic acid and microwave stimuli induced SOS and AOS elevations. These stimuli altered the cell size, number, differentiation, viscosity, and stiffness, which corresponded well to the fluctuation of the SOS and AOS values after incubation. Different anticancer drugs interacted with cancer cells to induce the characteristic alterations of the SAM values. These structural and mechanical alterations induced in cells was difficult to observe on light microscopy. Cellular damages were statistically compared between different stimuli and time-lapse cellular changes were observed using a SAM analysis. SAM is a useful modality to evaluate cellular damage.

Lung cancer is a deadly disease if not diagnosed in its early stages. However, early detection of lung cancer is a challenging task due to the shape and size of its nodules. Radiologists need support from automated tools for precise opinion. Automated detection of the affected lungs nodule is difficult because of the shape similarity among healthy tissues. Over the years, several expert systems have been developed that help radiologists to diagnose lung cancer. In this article, we propose a framework to precisely detect lungs cancer by classifying it between benign and malignant nodules. The framework is tested using the subset of the publicly available at the Lung Image Database Consortium image collection (LIDC-IDRI). Multiple techniques including filtering and noise removing are applied for pre-processing. Subsequently, the OTSU and the semantic segmentation are used to accurately detect the unhealthy lungs nodules. In total, 13 nodules features were extracted using Principal Components Analysis (PCA) algorithm. Four optimal features are selected based on the classification performance. In the classification phase, 9 different classifiers are used along with two types of validation schemes i.e. train test holdout validation with 70-30 data split and 10 fold cross-validation. Our experiments show that the proposed system provides 99.23\% accuracy using logic boost classifier.

Low learning rates in Convolutional Neural Networks (CNNs) for image segmentation tasks can lead to convergence issues, unstable models, large oscillations, risk of divergence, sensitivity to the initial weights, and biases of the model. All these challenges make CNNs computationally expensive and require considerable training data. In contrast, achieving a high learning rate has been a significant challenge for CNNs. In response to the abovementioned challenge, this work aims to increase the learning rate of the image segmentation process. Images of any object in the universe taken from any camera can broadly be divided into spectral and geometric properties. Both these properties are prominently visible in satellite images. Therefore, if a method can classify satellite images, it can be applied to classify any image using the same technique. In this paper, Evolutionary Convergent Functions (ECF) are proposed. They convert features' spectral and geometric properties in a satellite image into mathematical equations using a decision tree and then converge with a neural network. Different high-resolution data have been chosen to extract additional features from it. This transformation process, anchored in decision tree methodology, converges with neural networks to yield unparalleled results, all while eliminating the need for computationally intensive convolutions. The proposed method extends beyond conventional boundaries by using varied high-resolution datasets. Each dataset is carefully selected with distinctive features. This research explores the untapped potential of ECF to advance the field of automated image classification, aiming to broaden the scope of current methodologies. The synergy between spectral and geometric properties emerges as a powerful combination, endowing our methods with the ability to extract nuanced and context-rich information for image analysis. The results show a high accuracy, i.e., above 90%, in almost all objects of different shapes and spectral signatures. Additionally, in terms of prediction time, ECF is faster than U-Net, a state-of-the-art method that is evidence of ECF efficiency and speed.

There have been multiple technological advancements that promise to gradually enable devices to measure and record signals with high resolution and accuracy in the domain of Brain Computer Interfaces (BCI). Multi-modal BCIs have been able to gain significant traction given the potential to enhance signal processing by integrating different recording modalities. In this review, we explore the integration of multiple neuroimaging and neurophysiological modalities including Electroencephalography (EEG), Magnetoencephalography (MEG), Functional Magnetic Resonance Imaging (fMRI), Electrocorticography (ECoG), and Single-Unit Activity (SUA). This multimodal approach leverages the high temporal resolution of EEG and MEG with the spatial precision of fMRI, the invasive yet precise nature of ECoG, and the single-neuron specificity provided by SUA. The paper highlights the advantages of integrating multiple modalities, such as increased accuracy and reliability, and discusses the challenges and limitations of multimodal integration. Furthermore, we explain the data acquisition approaches for each of these modalities. We also demonstrate various software programs that help in extracting, cleaning, and refining the data. We conclude this paper with discussion on the available literature highlighting recent advances, challenges, and future directions for each of these modalities.

Technology-based approaches for upper limb (UL) motor rehabilitation after stroke are designed mostly for severely affected patients to increase their recovery chances. However, the available randomised controlled trials (RCTs) focused on efficacy of technology-based interventions often include patients with a wide range of motor impairment. This scoping review aims at overviewing the actual severity of stroke patients enrolled in RCTs which claim to specifically address UL severe motor impairment. The literature search was conducted on Scopus and PubMed databases and included articles from 2008 to May 2024, specifically RCTs investigating the impact of technology-based interventions on UL motor functional recovery after stroke. Forty-eight studies were selected. They showed that upon patients’ enrolment the values of the UL Fugl-Meyer Assessment and Action Research Arm Test covered the whole range of both scales, thus revealing the non-selective inclusion of severe impaired patients. Heterogeneity in terms of numerosity, characteristics of enrolled patients, trial design, implementation, and reporting were present across studies. No clear difference in the severity of the included patients according to the intervention type was found. Patient stratification upon enrolment is crucial to best direct resources to those patients who will benefit the most from a given technology-assisted approach (personalised rehabilitation).

Few prior works study self-driving cars by deep learning with IoT collaboration. SDC-Net, which is an end-to-end multitask self-driving car camera cocoon IoT-based system, is one of the research that tackles this direction. However, by design SDC-Net is not able to identify the accident locations, it only classifies if it is a crash scene or not. In this work, we introduce an enhanced design for the SDC-Net system by 1) replacing the classification network with a detection one, 2) adapting our benchmark dataset labels built on the CARLA simulator to include the vehicles bounding boxes while keeping the same training, validation, and testing samples, 3) modifying the shared information via IoT to include the accident location. We keep the same path planning and automatic emergency braking network, the digital automation platform, and the input representations to formulate the comparative study. SDC-Net++ system is proposed to 1) output the relevant control actions, especially in case of accidents: accelerate, decelerate, maneuver, and brake, and 2) share the most critical information to the connected vehicles via IoT, especially the accident locations. A comparative study is also conducted between SDC-Net and SDC-Net++ with the same input representations: front camera only, panorama and bird-eye views, and with single-task networks: crash avoidance only, and multitask networks. Multitask network with a BEV input representation outperforms the nearest representation in precision, recall, f1-score, and accuracy by more than 15.134%, 12.046%, 13.593%, and 5%, respectively. SDC-Net++ multitask network with BEV outperforms SDC-Net multitask with BEV in precision, recall, f1-score, accuracy, and average MSE by more than 2.201%, 2.8%, 2.505%, 2%, and 18.677% respectively.

A single paragraph of about 200 words maximum. For research articles, abstracts should give a pertinent overview of the work. We strongly encourage authors to use the following style of structured abstracts, but without headings: (1) Background: Place the question addressed in a broad context and highlight the purpose of the study; (2) Methods: briefly describe the main methods or treatments applied; (3) Results: summarize the article’s main findings; (4) Conclusions: indicate the main conclusions or interpretations. The abstract should be an objective representation of the article and it must not contain results that are not presented and substantiated in the main text and should not exaggerate the main conclusions.

Emotions play a vital role in understanding human behavior and interpersonal relationships. The ability to recognize emotions through Electroencephalogram (EEG) signals offers an alternative to traditional methods, such as questionnaires, enabling the identification of emotional states in a non-intrusive manner. Automatic emotion recognition holds great potential, eliminating the need for clinical examinations or physical visits, thereby contributing significantly to the advancement of Brain-Computer Interface (BCI) technology. However, one of the key challenges lies in effectively selecting and extracting relevant features from the EEG signal to establish meaningful distinctions between different emotional states. The process of feature selection is often time-consuming and demanding. In this research, we propose a groundbreaking approach for automatically identifying three emotional states (positive, negative, and neutral) by leveraging auditory stimulation of EEG signals. Our novel method directly applies the raw EEG signal to a Convolutional Neural Network-Long Short Term Memory (CNN-LSTM) architecture, bypassing the conventional feature extraction and selection steps. This unconventional approach offers a significant departure from existing literature. Our proposed network architecture comprises ten convolutional layers, followed by three LSTM layers and two fully connected layers. Through extensive simulations and evaluations on 12 active channels, our algorithm demonstrates exceptional performance, achieving an accuracy of 97.42\% and 95.23\% for the binary classification of negative and positive emotions, as well as a Cohen's Kappa coefficient of 0.96 and 0.93 for the three-class classification (negative, neutral, and positive), respectively. These promising results highlight the efficacy of our novel methodology and its potential implications in advancing emotion recognition using EEG signals.

Document deskewing is a fundamental problem in document image processing. While existing methods have limitations, such as Hough Line Transformation that can deskew images upside down, and Deep Learning models that require huge amounts of human labour and computational resources and still fail to deskew while taking care of orientation, OCR-based methods also struggle to read text when it is tilted. In this paper, we propose a novel, simple, cost-effective deep learning method for fixing the skew and orientation of documents. Our approach reduces the search space for the machine learning model to predict whether an image is upside down or not, avoiding the huge search space of predicting an angle between 0 and 360. We finetuned a MobileNetV2 model, which was pre-trained on imagenet, using only 200 images and achieve good results. This method is useful for automation-based tasks, such as data extraction using OCR technology, and can greatly reduce manual labour.

A machine learning method for classifying Lung UltraSound is here proposed to pro- vide a point of care tool for supporting a safe, fast and accurate diagnosis, that can also be useful during a pandemic like as SARS-CoV-2. Given the advantages (e.g. safety, rapidity, portability, cost-effectiveness) provided by the ultrasound technology over other methods (e.g. X-ray, computer tomography, magnetic resonance imaging), our method was validated on the largest LUS public dataset. Focusing on both accuracy and efficiency, our solution is based on an efficient adaptive ensembling of two EfficientNet-b0 models reaching 100% of accuracy, which, to our knowledge, outperforms the previous state-of-the-art. The complexity of this solution keeps the number of parameters in the same order as an EfficientNet-b0 by adopting specific design choices that are adaptive ensembling with a combination layer, ensembling performed on the deep features, minimal ensemble only two weak models. Moreover, a visual analysis of the saliency maps on sample images of all the classes of the dataset reveals where the focus is on an inaccurate weak model versus an accurate model.

Diffuse interstitial lung diseases (DILD) are a heterogeneous group of over 200 entities, some with dramatical evolution and poor prognostic. Because of their overlapping clinical, physiopathological and imagistic nature, successful management requires early detection and proper progression evaluation. This paper tests a complex networks (CN) algorithm for imagistic aided diagnosis fitness for the possibility of achieving relevant and novel DILD management data. 65 DILD and 31 normal high resolution computer tomography (HRCT) scans were selected and analyzed with the CN model. The algorithm is showcased in two case reports and then statistical analysis on the entire lot shows that a CN algorithm quantifies progression evaluation with a very fine accuracy, surpassing functional parameters’ variations. The CN algorithm can also be successfully used for early detection, mainly on the ground glass opacity Hounsfield Units band of the scan.

Mental stress state recognition using electroencephalogram (EEG) signals for real-life applications needs a conventional wearable device. This requires an efficient number of EEG channels and an optimal feature set. The main objective of the study is to identify an optimal feature subset that can best discriminate mental stress states while enhancing the overall performance. Thus, multi-domain feature extraction methods were employed, namely, time domain, frequency domain, time-frequency domain, and network connectivity features, to form a large feature vector space. To avoid the computational complexity of high dimensional space, a hybrid feature selection (FS) method of minimum Redundancy Maximum Relevance with Particle Swarm Optimization and Support Vector Machine (mRMR-PSO-SVM) is proposed to remove noise, redundant, and irrelevant features and keep the optimal feature subset. The performance of the proposed method is evaluated and verified using four datasets, namely EDMSS, DEAP, SEED, and EDPMSC. To further consolidate, the effectiveness of the proposed method is compared with that of the state-of-the-art heuristic methods. The proposed model has significantly reduced the features vector space by an average of 70% in comparison to the state-of-the-art methods while significantly increasing overall detection performance.

Virtual Reality (VR) is increasingly used in neuropsychological assessments due to its ability to simulate real-world environments. This study aimed to develop and evaluate the Trail Making Test in VR (TMT-VR) and investigate the effects of different interaction modes and gaming skills on cognitive performance. A total of 71 young female and male adults (aged 18-35) with high and low gaming skills participated in this study. Participants completed the TMT-VR using three interaction modes: eye-tracking, head movement, and controller. Performance metrics included task completion time and accuracy. User experience, usability, and acceptability of TMT-VR were also examined. Results showed that both eye-tracking and head movement modes significantly outperformed the controller in terms of task completion time and accuracy. No significant differences were found between eye-tracking and head movement modes. Gaming skills did not significantly influence task performance using any interaction mode. Participants reported high usability, acceptability, and user experience for the TMT-VR. Eye-tracking facilitated higher accuracy, while head movement higher speed. The TMT-VR offers high usability and user experience and is well accepted by the participants. The findings suggest that VR-based assessments can effectively measure cognitive performance without being influenced by prior gaming skills, indicating potential applicability for diverse populations.

This paper presents the development of a hardware-in-the-loop ground testbed featuring active gravity compensation via software-in-the-loop integration, specially designed to support research in autonomous robotic removal of space debris. The testbed is designed to replicate six degrees of freedom (6DOF) motion maneuvering to accurately simulate the dynamic behaviors of free-floating robotic manipulators and free-tumbling space debris under microgravity conditions. The testbed incorporates two industrial 6DOF robotic manipulators, a 3-finger robotic gripper, and a suite of sensors, including cameras, force/torque sensors, and tactile tensors. Such a setup provides a robust platform for testing and validating technologies related to autonomous tracking, capture, and post-capture stabilization within the context of active space debris removal missions. Preliminary experimental results have demonstrated advancements in motion control, computer vision, and sensor fusion. This facility is positioned to become an essential resource for the development and validation of robotic manipulators in space, offering substantial improvements in the effectiveness and reliability of autonomous capture operations in space missions.

This study investigates the use of artificial intelligence (AI) and ethnobotanical knowledge to identify natural compounds with potent antiviral activity against HPV-16. Through molecular docking simulations, we evaluated the binding affinities of various essential oil components, in-cluding Apigenin, against the key antigenic determinants of HPV-16 (E6, E7 oncoproteins, and L1 major capsid protein). Apigenin exhibited strong binding affinities, particularly to E6 Oncopro-tein. To further enhance these findings, we employed an AI-driven reverse engineering approach to predict a natural compound with superior efficacy. The AI identified Luteolin, a flavonoid with additional hydroxyl groups, as a promising candidate. Comparative docking studies demon-strated that Luteolin possesses significantly stronger binding affinities to all three HPV-16 targets, surpassing the efficacy of Apigenin. This work highlights the potential of AI in accelerating the discovery and optimization of natural antiviral agents, providing a powerful tool for pharmaco-logical and ethnopharmacological screenings. Our findings suggest that AI-guided identification of natural compounds can lead to the development of more effective antiviral therapies, warranting further experimental validation and potential synthesis in laboratory settings.

Image generation optimization is an important research direction in the field of deep learning, which aims to improve the performance of image generation models and the quality of generated images. In recent years, researchers have made significant progress in image generation optimization with the development of deep generative models such as generative adversarial networks (GANs) and variational autoencoders (VAEs). These models are able to generate high-quality, realistic images by learning the distribution of image data. In this study, a deep learning-based image generation optimization model was adopted, which combined the advantages of GAN and VAE. The model architecture consists of a generator and a discriminator, where the generator is responsible for generating the image and the discriminator is used to judge the authenticity of the image. In addition, the model also introduces attention mechanism and self-supervised learning strategy to further improve the quality and diversity of generated images. In the training process, a large-scale image dataset is used, and a variety of optimization algorithms are used to improve the stability and efficiency of the model. By evaluating various indicators of the generative model, including image quality, generation speed and model convergence, it was found that the introduced attention mechanism and self-supervised learning strategy significantly improved the performance of the model.

Celiac disease is a gluten-sensitive immune-mediated enteropathy of the small intestine that occurs in genetically predisposed individuals. Abnormal immune response results in mucosal inflammation, villous atrophy, and crypt hyperplasia. This study was a proof-of-concept exercise that used a convolutional neural network to classify hematoxylin and eosin (H&E) histological images of celiac disease, normal small intestine control, and non-specified duodenal inflammation; 7294, 11,642, and 5966 images, respectively. The trained network classified celiac disease images with high performance (accuracy 99.7%, precision 99.6%, recall 99.3%, F1-score 99.5%, and specificity 99.8%). Interestingly, when the same network (already trained for the 3 class images), analyzed duodenal adenocarcinoma (3723 images), the new images were classified as duodenal inflammation in 63.65%, small intestine control in 34.73%, and celiac disease in 1.61% of the cases. Finally, when the network was retrained using the 4 histological subtypes of images, all performance parameters were above 99% for celiac disease. In conclusion, the convolutional neural network (CNN)-based deep neural system was able to classify medical histological images with high performance. Narrow artificial intelligence (AI) is designed to perform tasks that typically require human intelligence, but it operates within limited constraints and is task specific.

Visualization of volumetric data holds great importance, particularly in medical and biomedical applications. Various imaging techniques, such as FMRI and 3D microscopy, are employed to generate volumetric data, used in medical practice, research, and teaching. Commonly utilized tools like 3D Slicer, Fiji, and MATLAB ® aid in rendering and analyzing 3D images. However, these tools may lack comprehensive rendering functionality and face challenges in handling computational demands as data sizes grow. To address these limitations, this work introduces a GPU-supported renderer with a MATLAB ® interface. This solution gives the user flexible control over rendering parameters and optimizes data transfer through a memory management system. By leveraging the computational power of NVIDIA GPUs, the renderer enables complex and high-quality renderings, enhancing speed and efficiency. It therefore facilitates the analysis and visualization of volumetric data within an integrated environment, namely MATLAB ®, streamlining their workflows. This advancement provides valuable opportunities for researchers and medical professionals to explore and comprehend volumetric data effectively.

The human mandible's cancellous bone, which is characterized by its unique porosity and directional sensitivity to external forces, is crucial for sustaining biting stress. Traditional CAD models fail to fully represent bone's anisotropic structure and thus depend on simple isotropic assumptions. For our research, we use the latest versions of nTopology and Creo Parametric software to make biomimetic Voronoi lattice models that accurately reflect the complex geometry and mechanical properties of trabecular bone. The porosity of human cancellous bone is accurately modeled in this work using biomimetic Voronoi lattice models. The porosities range from 70% to 95%, which can be achieved by changing the pore sizes to 1.0 mm, 1.5 mm, 2.0 mm, and 2.5 mm. Finite element analysis (FEA) was used to examine the displacements, stresses, and strains acting on dental implants with a buttress thread, abutment, retaining screw, and biting load surface. The results show that the Voronoi model accurately depicts the complex anatomy of the trabecular bone in the human jaw, compared to standard solid block models. The ideal pore size for biomimetic Voronoi lattice trabecular bone models is 2 mm, taking into account both the von Mises stress distribution over the dental implant, screw retention, cortical bone, cancellous bone, and micromotions. This pore size displayed balanced performance by successfully matching natural bone's mechanical characteristics. Advanced finite element analysis (FEA) improves biomechanical understanding of how bones and implants interact by creating more accurate models of biological problems and dynamic loading situations. This makes biomechanical engineering better.

How crucial is big data in contemporary molecular design? In this publication we elucidate fundamental concepts and terminology in this field, critically addressing overlooked issues. We thoroughly examine the size, accessibility, quality, and structural aspects of big data alongside the primary methodologies employed for their analysis. Within chemical compounds, properties and descriptors represent two distinct data types, forming the basis for categorizing molecular big data. The primary objective of chemistry is property production, which means we are searching for novel drugs or materials rather than chemical compounds, and big data is the central issue of this philosophy. The increasing availability of data in computer-aided technology propels advancements in artificial intelligence (AI), machine learning (ML), and deep learning (DL). Accordingly, a broad chemical audience must comprehend these methods to understand data-centered chemistry. Thus, we aim to systemize big data issues through a simple illustrative framework with fundamental descriptor categories: coding, computer-generated descriptors, and property correlates. Although we employ computer-generated descriptors as big data for predictions, the measured data are irreplaceable for achieving high-quality and reliable outcomes and controlling molecular effects. The scarcity of property data remains a significant hurdle limiting comprehensive studies on the structure-property relationships within big data. Accordingly, guided by pragmatics not-so-big data is an option for drug design. We presented also a brief review of the recent big data literature.

In the dynamic landscape of drug discovery, Computer-rAided Drug Design (CADD) emerges as a transformative force, bridging the realms of biology and technology. This paper overviews CADD's historical evolution, categorization into structure-based and ligand-based approaches, and its crucial role in rationalizing and expediting drug discovery. As CADD advances, incorporating diverse biological data and ensuring data privacy become paramount. Challenges persist, demanding the optimization of algorithms and robust ethical frameworks. Integrating Machine Learning and Artificial Intelligence amplifies CADD's predictive capabilities, yet ethical considerations and scalability challenges linger. Collaborative efforts and global initiatives, exemplified by platforms like Open Source Malaria, underscore the democratization of drug discovery. The convergence of CADD with personalized medicine offers tailored therapeutic solutions, though ethical dilemmas and accessibility concerns must be navigated. Emerging technologies like quantum computing, immersive technologies, and green chemistry promise to redefine the future of CADD. The trajectory of CADD, marked by rapid advancements, anticipates challenges in ensuring accuracy, addressing biases in AI, and incorporating sustainability metrics. The paper concludes by highlighting the need for proactive measures in navigating the ethical, technological, and educational frontiers of CADD to shape a healthier, brighter future in drug discovery.

Thanks to the continued development of experimental structural biology and the half-a-century old Protein Data Bank, 2021 saw a big step forward in the development of protein structure prediction with deep learning algorithms. Recently, DeepMinds AlphaFold has determined the structures of ∼ 200 million proteins from 1 million species. The speed of this progress raise the question of what becomes possible for computational drug discovery and design when we have a systems-wide account of the structures and motions of most proteins. Therefore, this article puts forward the concept of a general intermolecular binding affinity calculator (GIBAC): Kd = f(molA, molB, envPara), towards the acceleration of traditional computer-aided drug design (CADD) and artificial intelligence-integrated drug discovery (AIDD), for both small molecules and biologics such as therapeutic proteins.

Pure orbital blowout fractures occur within the confines of the internal orbital wall. Restoration of orbital form and volume is paramount to prevent functional and esthetic impairment. The anatomical peculiarity of the orbit has encouraged surgeons to develop implants with customized features to restore its architecture. This has resulted in worldwide clinical demand for patient-specific implants (PSIs) designed to fit precisely in the patient's unique anatomy. Fused filament fabrication (FFF) three-dimensional (3D) printing technology has enabled the fabrication of implant-grade polymers such as Polyetheretherketone (PEEK), paving the way for a more sophisticated generation of biomaterials. This study evaluates the FFF 3D printed PEEK orbital mesh customized implants with a metric considering the relevant design, biomechanical, and morphological parameters. The performance of the implants is studied as a function of varying thicknesses and porous design constructs through a finite element (FE) based computational model and a decision matrix based statistical approach. The maximum stress values achieved in our results predict the high durability of the implants, and the maximum deformation values were under one-tenth of a millimeter (mm) domain in all the implant profile configurations. The circular patterned implant (0.9 mm) had the best performance score. The study demonstrates that compounding multi-design computational analysis with 3D printing can be beneficial for the optimal restoration of the orbital floor.

Medical brain image analysis is a necessary step in the Computers Assisted /Aided Diagnosis (CAD) systems. Advancements in both hardware and software in the past few years have led to improved segmentation and classification of various diseases. In the present work, we review the published literature on systems and algorithms that allow for classification, identification, and detection of White Matter Hyperintensities (WMHs) of brain MRI images specifically in cases of ischemic stroke and demyelinating diseases. For the selection criteria, we used the bibliometric networks. Out of a total of 140 documents we selected 38 articles that deal with the main objectives of this study. Based on the analysis and discussion of the revised documents, there is constant growth in the research and proposal of new models of deep learning to achieve the highest accuracy and reliability of the segmentation of ischemic and demyelinating lesions. Models with indicators (Dice Score, DSC: 0.99) were found, however with little practical application due to the uses of small datasets and lack of reproducibility. Therefore, the main conclusion is to establish multidisciplinary research groups to overcome the gap between CAD developments and their complete utilization in the clinical environment.

Objectives: Studies evaluating machine learning (ML) algorithms on cross-lingual RadLex® mappings for developing context-sensitive radiological reporting tools are lacking. Therefore, we investigated whether ML-based approaches can be utilized to assist radiologists in providing key imaging biomarkers – such as The Alberta stroke programme early CT score (APECTS). Material and Methods: A stratified random sample (age, gender, year) of CT reports (n=206) with suspected ischemic stroke was generated out of 3997 reports signed off between 2015-2019. Three independent, blinded readers assessed these reports and manually annotated clinico-radiologically relevant key features. The primary outcome was whether ASPECTS should have been provided (yes/no: 154/52). For all reports, both the findings and impressions underwent cross-lingual (German to English) RadLex®-mappings using natural language processing. Well-established ML-algorithms including classification trees, random forests, elastic net, support vector machines (SVMs) and boosted trees were evaluated in a 5 x 5-fold nested cross-validation framework. Further, a linear classifier (fastText) was directly fitted on the German reports. Ensemble learning was used to provide robust importance rankings of these ML-algorithms. Performance was evaluated using derivates of the confusion matrix and metrics of calibration including AUC, brier score and log loss as well as visually by calibration plots. Results: On this imbalanced classification task SVMs showed the highest accuracies both on human-extracted- (87%) and fully automated RadLex® features (findings: 82.5%; impressions: 85.4%). FastText without pre-trained language model showed the highest accuracy (89.3%) and AUC (92%) on the impressions. Ensemble learner revealed that boosted trees, fastText and SVMs are the most important ML-classifiers. Boosted trees fitted on the findings showed the best overall calibration curve. Conclusions: Contextual ML-based assistance suggesting ASPECTS while reporting neuroradiological emergencies is feasible, even if ML-models are restricted to be developed on limited and highly imbalanced data sets.

Alzheimer’s is a degenerative brain condition brought on by intricate brain changes after cell injury. The prevalence of Alzheimer’s disease is highest in those over 65. Presently, AD is detected by manual examination of a patient’s Magnetic Resonance Imaging (MRI) scan and neuropsychological assessments. Due to its subpar accuracy and performance, solutions based on artificial intelligence (AI) have not been made more widely available. This paper proposes a deep Convolutional Neural Network (CNN) with the pre-trained MobileNet-V2 architecture as the model’s base for feature extraction and reducing the parameter size and training time. The proposed model accurately classifies various early stages of AD. The proposed approach visualizes the class activation maps in the form of a heat map on brain MRI scans via the Grad-CAM algorithm that lifts the black-box nature of the deep model. To handle the imbalance problem of the AD data set acquired from Kaggle, the oversampling algorithm SMOTE interpolates new images to even the class samples. The proposed multi-class classification deep model is named Bio-AD and evaluated using different quality metrics. The Bio-AD model has the following values for evaluation metrics: 97.92%, 99.87%, 97.87%, 98%, 98%, and 0.0646 for accuracy, Area Under the Curve (AUC), F1-score, Precision, Recall, and loss respectively. The proposed Bio-AD model is compared with DenseNet-169, VGG-19, and InceptionResNet-V2 using different quality metrics and outperformed in comparison with all other models using the AD data set.

In this paper, we present a novel approach for managing the file system in Linux using a voice assistant. Our system allows users to perform file system operations such as creating directories, renaming files, and deleting files by issuing voice commands. We develop a voice assistant using Python libraries and integrate it with the file system in Linux. The voice assistant is capable of understanding natural language and executing commands based on the user's voice inputs. We conduct experiments to evaluate the performance of the system and demonstrate that our approach is effective and efficient in managing the file system using voice commands. Our system can enhance the accessibility and usability of the file system in Linux for individuals with disabilities or those who prefer a hands-free approach to file management.

This paper presents a methodology for the integration of Building Information Modeling (BIM) and Life Cycle Cost Analysis (LCCA) to assess the economic implications of designing aging-in-place (AIP) homes at the conceptual stage. With the global increase in the aging population, there is an increased demand for housing solutions tailored to the needs of elderly individuals. Focusing on the importance of the early phase of design, this study aims to improve the process of making efficient decisions by providing a comprehensive assessment of the life cycle costs associated with AIP homes. The study introduces a semi-automated model for the economic evaluation of AIP homes, enabling the estimation of costs throughout the houses’ entire life cycle, from design and construction to operation, maintenance, and eventual renovation or disposal. The said model facilitates the exploration of the long-term economic feasibility of design’s related decisions with an emphasis on the importance of considering the life cycle costs early during the design process to optimize the functionality and economic viability. By investing in accessible and universal design features upfront, the initial costs for modifications can lead to long-term savings by reducing the need for extensive retrofits. The model can easily do comparison between different design alternatives in terms of their lifecycle costs, allowing designers to assess the financial impact of using important features in their design such as wider doorways, accessible bathrooms, and elevators. Overall, this study provides valuable insights for designers and homeowners about the economic aspects of designing AIP homes as a support for efficient decision-making during the early stages of the design process.

In this paper, we critically examine the burgeoning role of advanced computational methodologies in deciphering the complex tapestry of farm animal behaviors and emotions. Leveraging digital imaging and artificial intelligence, we unearth nuanced behavioral patterns and micro-expressions, offering predictive insights into animal emotional states. Sound vocalization analysis, often overlooked, emerges as a pivotal tool, decoding intricate communicative nuances and emotional undertones. Cognitive tests, including mirror and bias assessments, challenge long-standing perceptions, revealing surprising depths of animal self-awareness and cognitive sophistication. However, the paper also underscores the imperative of integrating these tools with a profound understanding of animal psyche, ensuring technology serves as an enhancer, not a replacement, of traditional observational methods. This research not only highlights the transformative potential of cognitive computing in animal welfare but also calls for a judicious application, ensuring technology augments, not undermines, the intrinsic value of human-animal interactions and understanding.

In recent years, the analysis of macro- and micro-expression has drawn the attention of researchers since they provide visual cues of an individual's emotions for a broad range of potential applications such as lie detection and criminal detection. In this paper, we address the challenge of spotting facial macro- and micro-expression from videos and present compelling results by using a deep learning approach to analyze the optical flow features. Different from other deep learning approaches that are mainly based on Convolutional Neural Networks (CNNs), we propose a Transformer-based deep learning approach that predicts a score indicating the probability of a frame being within an expression interval. In contrast to other Transformer-based models that achieve high performance by being pre-trained on large datasets, our deep learning model, called SL-Swin, which incorporates Shifted Patch Tokenization and Locality Self-Attention into the backbone network Swin Transformer, effectively spots macro- and micro-expressions by being trained from scratch on small-size expression datasets. Our evaluation outcomes surpass the MEGC 2022 spotting baseline result, obtaining an overall F1-score of 0.1366. Additionally, our approach also performs well in the MEGC 2021 spotting task with an overall F1-score of 0.1824 and 0.1357 on CAS(ME)^2 and SAMM Long Videos, respectively. The code is publicly available on GitHub (https://github.com/eddiehe99/pytorch-expression-spotting).

Chatbots are intelligent conversational computer systems designed to mimic human conversation to enable automated online guidance and support. The increased benefits of chatbots led to their wide adoption by many industries in order to provide virtual assistance to customers. Chatbots utilise methods and algorithms from two Artificial Intelligence domains: Natural Language Processing and Machine Learning. However, there are many challenges and limitations in their application. In this survey we review recent advances on chatbots, where Artificial Intelligence and Natural Language processing are used. We highlight the main challenges and limitations of current work and make recommendations for future research investigation

Recent progress in deep learning has led to accurate and efficient generic object detection networks. Training of highly reliable models depends on large datasets with highly textured and rich images. However, in real-world scenarios, the performance of the generic object detection system decreases when (i) occlusions hide the objects, (ii) objects are present in low-light images, or (iii) they are merged with background information. In this paper, we refer to all these situations as challenging environments. With the recent rapid development in generic object detection algorithms, notable progress has been observed in the field of object detection in challenging environments. However, there is no consolidated reference to cover state-of-the-art in this domain. To the best of our knowledge, this paper presents the first comprehensive overview, covering recent approaches that have tackled the problem of object detection in challenging environments. Furthermore, we present the quantitative and qualitative performance analysis of these approaches and discuss the currently available challenging datasets. Moreover, this paper investigates the performance of current state-of-the-art generic object detection algorithms by benchmarking results on the three well-known challenging datasets. Finally, we highlight several current shortcomings and outline future directions.

Incorporating substantial sustainable development issues into teaching and learning is the ultimate task of Education for Sustainable Development (ESD). The purpose of our study is to identify the confused students who have failed to master the skill(s) given by the tutors as a homework using Intelligent Tutoring System (ITS). We have focused ASSISTments, an ITS in this study and scrutinized the skill-builder data using machine learning techniques and methods. We used seven candidate models that include: Naïve Bayes (NB), Generalized Linear Model (GLM), Logistic Regression (LR), Deep Learning (DL), Decision Tree (DT), Random Forest (RF), and Gradient Boosted Trees (XGBoost). We trained, validated and tested learning algorithms, performed stratified cross-validation and measured the performance of the models through various performance metrics i.e., ROC (Receiver Operating Characteristic), Accuracy, Precision, Recall, F-Measure, Sensitivity & Specificity. We found GLM, DT & RF are high accuracies achieving classifiers. However, other perceptions such as detection of unexplored features that might be related to the forecasting of outputs can also boost the accuracy of the prediction model. Through machine learning methods, we identified the group of students which were confused attempting the homework exercise and can help students foster their knowledge, and talent to play a vital role in environmental development.

The Release of government dataset for public use can potentially strengthen the relationship between the government and its constituents. However, research shows that there are several challenges for open data effectiveness. This paper reviews current determinants and issues associated with the open government data (OGD) procedures. The review concentrates on two ends of the spectrum: First, from the perspective of the preparation by the government, focusing on the organization of traditional governmental datasets and how the recording of the data is administered. Second, from the perspective of the users, focusing on the way in which the data is released to the general public and on human-computer interaction (HCI) issues between end-user and data-consumption interfaces. Following a thorough analysis of these two opposing challenges, the paper proposes approaches to mitigate them. This review and subsequent recommendations contribute and expand current understanding of open government data effectiveness and can lead to public policy changes, development of new procedures and strategies, and ultimately improvements at both ends of the federal open data endeavor.

Experiential learning, defined as the result, source, or context of knowledge obtained from direct experience, is increasingly bolstered by technology and human-computer interaction (HCI) principles and protocols. This article will describe how technology-based human-computer interaction (HCI) has been used in experiments to enhance direct knowledge experience. In this paper, immersive technologies (such as Virtual Reality, Augmented Reality, and haptics) will be described based on cognitive learning theory, empirical research and practical experience gained from implementing cutting-edge technology and maximizing HCI principles in constructive learning. Additionally, this article will describe how AI can be used to create individualized learning environments according to a student's needs in the future. While cost, technical issues and specialised trainings are the obstacle to deliver personalized learning experiences especially to economically disadvantaged students, technological change opportunities are projected, and these technologies will take educational outcomes to the next level.

The rapid technological progress has ushered in a new era of human-computer interaction, where the distinction between the physical and virtual realms is becoming increasingly blurred. This research paper explores the profound and multifaceted intersection of Human-Data Interaction (HDI) and Data Virtualization (DV), examining how emerging technologies can significantly enhance the exploration, comprehension, and utilization of complex, multidimensional data sets. Informed by the insights gleaned from prior research in this domain , the present study delves into the potential of DV techniques to improve HDI, with a particular focus on three experimental investigations conducted within the realms of education, healthcare, and retail. The findings reveal the benefits and potential challenges associated with the implementation of DV in these diverse contexts, offering valuable guidance for the design and development of future HDI systems. Drawing upon a diverse array of authoritative sources, this paper presents a holistic, forward-looking perspective on the future of HDI, underscoring the critical role that DV will play in shaping the next generation of human-computer interfaces and facilitating a deeper, more intuitive understanding of the digital world. Furthermore, the paper presents a preliminary framework for integrating HDI principles into standard design practices. This framework outlines key considerations and guidelines to help designers and developers incorporate HDI techniques more effectively into the development of data-driven applications and interfaces.The proposed framework outlines key considerations for enhancing data accessibility and comprehension, empowering users to exercise greater control over their data, and cultivating transparent dialogues between data providers and end-users. By establishing this conceptual foundation, the paper aims to facilitate the seamless integration of HDI principles into standard design practices, ultimately leading to more intuitive, user-centric, and ethically-grounded approaches to data interaction and utilization.

Cognitive impairment is a condition which affects aging patient populations and is becoming more prevalent as the geriatric population expands. It is categorized by the extent of cognitive decline, memory loss, functional abilities, and the capability to perform daily tasks. Based on the degree of impairment, it is classified as mild cognitive impairment or dementia. Considering the debilitating nature of cognitive decline and its impact on quality of life, prevention is key. In addition, given the gradual progression of the underlying causes of cognitive decline, early diagnosis and appropriate management has the potential to offer clinical advantages and better patient outcomes. Management options discussed vary from pharmaceutical regimens to neurosurgical interventions, with differing outcomes based on etiology of cognitive impairment.

Improper sitting positions are known as the primary reason for back pain and the emergence of musculoskeletal disorders (MSD) among individuals who spend prolonged time working with computer screens, keyboards, and mice. At the same time, it is well understood that automated technological tools can play an important role in the process of unhealthy habit alteration, so plenty of research efforts are focused on research and technology development (RTD) activities that aim to provide support to the prevention of back pain or the development of MSD. Here, we report on creating a new resource in support of RTD activities aiming at the automated detection of improper sitting positions. It consists of multimodal multisensory recordings of 100 persons made with a video recorder, camera, and wrist-attached sensors that capture physiological signals (PPG, EDA, Skin temperature), as well as motion sensors (three-axis accelerometer). Our multimodal multisensory dataset (MMD-MSD) opens new opportunities for modeling the body stance (sitting posture and movements), physiological state (stress level, attention, emotional arousal, valence), and performance (success rate on the Stroop test) of people working with a computer. Finally, we demonstrate two use cases: improper neck posture detection from pictures and cognitive load detection from physiological signals.

Due to Motor Neurone Diseases, a large population remains disabled worldwide, negatively impacting their independence and quality of life. This typically involves a weakness in the hand and forearm muscles, making it difficult to perform fine motor tasks such as writing, buttoning a shirt, or gripping objects. This project presents a vision-enabled rehabilitation hand exoskeleton to assist disabled persons in their hand movements. The design goal was to create an accessible tool to help with a simple interface requiring no training. This prototype is built on a commercially available glove where a camera and embedded processor were integrated to help open and close the hand, using air pressure, thus grabbing an object. An accelerometer is also implemented to detect the characteristic hand gesture to release the object when desired. This passive vision-based control differs from active EMG-based designs as it does not require individualized training. Continuing the research will reduce the cost, weight, and power consumption to facilitate mass implementation.

Some of the barriers preventing Virtual Reality (VR) from being widely adopted are the cost and unfamiliarity of VR systems. Here we propose that in many cases the specialized controllers shipped with most VR head-mounted displays can be replaced by a regular smartphone, cutting the cost of the system, and allowing users to interact in VR using a device they are already familiar with. To achieve this, we developed SmartVR Pointer, an approach that uses smartphones as a replacement for the specialized controllers for two essential operations in VR: selection and navigation by teleporting. In SmartVR Pointer a camera mounted on the head-mounted display (HMD) is tilted downwards so that it points to where the user will naturally be holding their phone in front of them. SmartVR Pointer supports three selection modalities: tracker-based,gaze-based, and combined/hybrid. In the tracker-based SmartVR pointer selection we use image-based tracking to track a QR code displayed on the phone screen and them map the phone’s position to a pointer shown within the field of view of the camera in the Virtual Environment. In the gaze-based selection modality the user places the pointer using their gaze and taps on the phone for selection. The combined technique is a hybrid between head-gaze-based interaction in VR and smartphone-based Augmented Reality. It allows the user to control a VR pointer that behaves like a mouse pointer by moving their smartphone to select objects within the virtual environment, and to interact with the selected objects using the smartphone’s touch screen. The touchscreen is used for selection and dragging. SmartVR Pointer is simple and requires no calibration and no complex hardware assembly or disassembly. We demonstrate successful interactive applications of SmartVR Pointer in a VR environment with a demo where the user navigates in the virtual environment using teleportation points on the floor and then solves a Tetris-style key-and-lock challenge.

The paper presents a proposal for an in-house mathematical model of a liquid flat-plate solar collector. The proposed model is a one-dimensional distributed parameter model enabling simulations of the collector operation under arbitrarily variable boundary conditions. The model is based on the solution of energy balance equations for all components of the collector. The formulated differential equations are solved iteratively using an implicit difference scheme. In order to obtain a stable numerical solution, it is necessary to use appropriate steps of the time and spatial division. These were found by comparing the results obtained from the model with the results of the analytical solution available in the literature for the transient state, which constitutes the novelty of the present study. The accuracy of the results obtained from the model was verified experimentally by comparing the measured and the calculated history of the fluid temperature at the outlet of the collector. The calculation of the collector time constant is proposed in the paper as an example of the model practical application. The results of the time constant calculation were compared with the values obtained experimentally on the test stand. This is another novelty of the presented research. The analysed collector instantaneous efficiency was then calculated for selected outdoor conditions. The presented mathematical model can also be used to verify the correctness of the collector operation. Comparing, on an ongoing basis, the measured and the calculated values of the fluid temperature at the collector outlet, conclusions can be drawn about the process of solar glass fouling or glycol gelling. The model's simplicity and the low computational demands enable such comparisons in an online mode.

This paper presents a methodology to detect the erosion and progradation effects in the Atrato river delta, located in the Gulf of Urabá in Colombia using SAR (Synthetic Aperture Radar) data. Other authors have studied this area by using optical images, encountering difficulties given the persistent presence of clouds in this region. SAR imagery is rescaled, despeckled, composed, classified, registered and processed to obtain a heatmap that shows the overall changes during eight years (2016–2023) in one single image. Since this technology uses microwave active signals, the presence of clouds and other environmental phenomena was not a problem, and all the images from the selected dates could be processed and analyzed. Also, detection of the dynamic processes was obtained for every period considered in this study with respect to the previous one in time, calculating areas and detecting the regions subject to the erosion and progradation processes. The multitemporal analysis performed found that the newly created mouth is the most active area for these processes, coinciding with other studies. A comparison of these findings with the Oceanic Niño Index (ONI) showed a relative delayed coupling to the erosion process and a coupling of progradation with dry and wet seasons.

Integrating voice assistant technology can potentially enhance accessibility and user interaction in web-based platforms for geospatial data visualisation and processing. This paper presents an open-source development incorporating a voice virtual assistant into the BStreams web application platform, designed specifically for visualising geospatial data. The study aimed to mitigate the observed limitations of virtual assistants in understanding the semantics and context of user commands in the geospatial domain. A discourse framework specialised for geospatial data was developed using a structured prototype methodology. In addition, a survey assisted in identifying commonly used English terminologies for geospatial data interaction. The questionnaire results were subsequently assessed using the ChatGPT natural language processing model. The application used open-source frameworks, including Web Speech API, Leaflet, Mapbox geocoding, and JavaScript. This investigation indicates that voice virtual assistants can potentially enhance human-computer interaction for geospatial data visualisation and processing. However, challenges related to natural language understanding, accents, and domain knowledge were raised. The paper advocates for further research to refine the integration of voice technology in this domain.

Background: Processing whole-slide images (WSI) to train neural networks can be intricate and laborious. We developed an open-source library covering recurrent tasks in processing of WSI and in evaluating the performance of the trained networks for classification tasks. Methods: Two histopathology use-cases were selected. First we aimed to train a CNN to distinguish H&amp;E-stained slides obtained from neuropathologically classified low-grade epilepsy-associated dysembryoplastic neuroepithelial tumor (DNET) and ganglioglioma (GG). The second project we trained a convolutional neural network (CNN) to predict the hormone expression of pituitary adenoms only from hematoxylin and eosin (H&amp;E) stained slides. In the same approach, we addressed the issue to also predict clinically silent corticotroph adenoma. We included four clinico-pathological disease conditions in a multilabel approach. Results: Our best performing CNN achieved an area under the curve (AUC) of 0.97 for the receiver operating characteristic (ROC) for corticotroph adenoma, 0.86 for silent corticotroph adenoma and 0.98 for gonadotroph adenoma. Our DNET-GG classifier achieved an AUC of 1.00 for the ROC curve. All scores were calculated with the help of our library on predictions on a case basis. Conclusions: Our comprehensive library is most helpful to standardize the work-flow and minimize the work-burden in training CNN. It is also compatible with fastai. Indeed, our new CNNs reliably extracted neuropathologically relevant information from the H&amp;E staining only. This approach will supplement the clinico-pathological diagnosis of brain tumors, which is currently based on cost-intense microscopic examination and variable panels of immunohistochemical stainings.

The rising popularity of UAVs and other autonomous control systems, coupled with real-time operating systems, has increased the complexity of developing systems with the proper robustness, performance, and reactivity. Concurrently, the growing demand for more sophisticated computational tasks, proportionally bigger payloads, battery limitations, and smaller take-off mass necessitates higher energy efficiency for all avionics and mission computers. The purpose of this paper is to develop the technology for experimental studies of indicators of reactivity and energy consumption of the computing platform for unmanned aerial vehicles (UAVs). The paper provides an experimental assessment of the ’Boryviter’ 0.1 computing platform, which is implemented on the ATSAMV71 microprocessor and operates under the open-source FreeRTOS operating system. The results obtained are the basis for developing algorithms and energy-efficient design strategies for the mission computer to solve the optimization problem. The paper provides experimental results of measurements of the energy consumed by the microcontroller and estimates of the reduction in system energy consumption due to additional time costs for suspending and resuming the computer’s operation. The results show that the ’Boryviter’ 0.1 computing platform can be used as a UAV mission computer for typical flight control tasks that require real-time computing under the influence of external factors. As the further work direction, authors plan to carry out the investigation of the proposed energy saving algorithms in scope of the planned NASA’s F’Prime software flight framework. Such an investigation that should be done with the real flight computation load for the mission computer, will help to qualify the obtained energy saving methods and their implementation results.

Table detection is a preliminary step in extracting reliable information from tables in scanned document images. We present CasTabDetectoRS, a novel end-to-end trainable table detection framework that operates on Cascade Mask R-CNN, including Recursive Feature Pyramid network and Switchable Atrous Convolution in the existing backbone architecture. By utilizing a comparatively lightweight backbone of ResNet-50, this paper demonstrates that superior results are attainable without relying on pre and post-processing methods, heavier backbone networks (ResNet-101, ResNeXt-152), and memory-intensive deformable convolutions. We evaluate the proposed approach on five different publicly available table detection datasets. Our CasTabDetectoRS outperforms the previous state-of-the-art results on four datasets (ICDAR-19, TableBank, UNLV, and Marmot) and accomplishes comparable results on ICDAR-17 POD. Upon comparing with previous state-of-the-art results, we obtain a significant relative error reduction of 56.36%, 20%, 4.5%, and 3.5% on the datasets of ICDAR-19, TableBank, UNLV, and Marmot, respectively. Furthermore, this paper sets a new benchmark by performing exhaustive cross-datasets evaluations to exhibit the generalization capabilities of the proposed method.

Tables in the document image are one of the most important entities since they contain crucial information. Therefore, accurate table detection can significantly improve information extraction from tables. In this work, we present a novel end-to-end trainable pipeline, HybridTabNet, for table detection in scanned document images. Our two-stage table detector uses the ResNeXt-101 backbone for feature extraction and Hybrid Task Cascade (HTC) to localize the tables in scanned document images. Moreover, we replace conventional convolutions with deformable convolutions in the backbone network. This enables our network to detect tables of arbitrary layouts precisely. We evaluate our approach comprehensively on ICDAR-13, ICDAR-17 POD, ICDAR-19, TableBank, Marmot, and UNLV. Apart from the ICDAR-17 POD dataset, our proposed HybridTabNet outperforms earlier state-of-the-art results without depending on pre and post-processing steps. Furthermore, to investigate how the proposed method generalizes unseen data, we conduct an exhaustive leave-one-out-evaluation. In comparison to prior state-of-the-art results, our method reduces the relative error by 27.57% on ICDAR-2019-TrackA-Modern, 42.64% on TableBank (Latex), 41.33% on TableBank (Word), 55.73% on TableBank (Latex + Word), 10% on Marmot, and 9.67% on UNLV dataset. The achieved results reflect the superior performance of the proposed method.

Visual Question Generation (VQG) from images is a rising research topic in both fields of natural language processing and computer vision. Although there are some recent efforts towards generating questions from images in the open domain, the VQG task in the medical domain has not been well-studied so far due to the lack of labeled data. In this paper, we introduce a goal-driven VQG approach for radiology images called VQGRaD that generates questions targeting specific image aspects such as modality and abnormality. In particular, we study generating natural language questions based on the visual content of the image and on additional information such as the image caption and the question category. VQGRaD encodes the dense vectors of different inputs into two latent spaces, which allows generating, for a specific question category, relevant questions about the images, with or without their captions. We also explore the impact of domain knowledge incorporation (e.g., medical entities and semantic types) and data augmentation techniques on visual question generation in the medical domain. Experiments performed on the VQA-RAD dataset of clinical visual questions showed that VQGRaD achieves 61.86% BLEU score and outperforms strong baselines. We also performed a blinded human evaluation of the grammaticality, fluency, and relevance of the generated questions. The human evaluation demonstrated the better quality of VQGRaD outputs and showed that incorporating medical entities improves the quality of the generated questions. Using the test data and evaluation process of the ImageCLEF 2020 VQA-Med challenge, we found that relying on the proposed data augmentation technique to generate new training samples by applying different kinds of transformations, can mitigate the lack of data, avoid overfitting, and bring a substantial improvement in medical VQG.

General–purpose automatic speech recognition (ASR) systems have improved their quality and are being used for pronunciation assessment. However, the assessment of isolated short utterances, as words in minimal pairs for segmental approaches, remains an important challenge, even more for non-native speakers. In this work, we compare the performance of our own tailored ASR system (kASR) with the one of Google ASR (gASR) for the assessment of Spanish minimal pair words produced by 33 native Japanese speakers in a computer-assisted pronunciation training (CAPT) scenario. Participants of a pre/post-test training experiment spanning four weeks were split into three groups: experimental, in-classroom, and placebo. Experimental group used the CAPT tool described in the paper, which we specially designed for autonomous pronunciation training. Statistically significant improvement for experimental and in-classroom groups is revealed, and moderate correlation values between gASR and kASR results were obtained, beside strong correlations between the post-test scores of both ASR systems with the CAPT application scores found at the final stages of application use. These results suggest that both ASR alternatives are valid for assessing minimal pairs in CAPT tools, in the current configuration. Discussion on possible ways to improve our system and possibilities for future research are included.

As people grow a custom to effortless outdoor navigation there is a rising demand for similar possibility indoors as well. Unfortunately, indoor localization, being one of the necessary requirements for navigation, continues to be problem without a clear solution. In this article we are proposing a method for an indoor positioning system using a single image. This is made possible using small preprocessed database of images with known control points as the only preprocessing needed. Using feature detection with SIFT algorithm we can look through the database and find image which is the most similar to the image taken by user. Pair of images is then used to find coordinates of database image using PnP problem. Furthermore, projection and essential matrices are determined allowing for the user image localization ~ determining the position of the user in indoor environment. Benefits of this approach lies in the single image being the only input from user and no requirements for new onsite infrastructure and thus enables a simpler realization for the building management.