The classical approach to non-linear regression in physics, is to take a mathematical model describing the functional dependence of the dependent variable from a set of independent variables, and then, using non-linear fitting algorithms, extract the parameters used in the modeling. Particularly challenging are real systems, characterised by several additional influencing factors related to specific components, like electronics or optical parts. In such cases, to make the model reproduce the data, empirically determined terms are built-in the models to compensate for the impossibility of modeling things that are, by construction, impossible to model. A new approach to solve this issue is to use neural networks, particularly feed-forward architectures with a sufficient number of hidden layers and an appropriate number of output neurons, each responsible for predicting the desired variables. Unfortunately, feed-forward neural networks (FFNNs) usually perform less efficiently when applied to multi-dimensional regression problems, that is when they are required to predict simultaneously multiple variables that depend from the input dataset in fundamentally different ways. To address this problem, we propose multi-task learning (MTL) architectures. These are characterized by multiple branches of task-specific layers, which have as input the output of a common set of layers. To demonstrate the power of this approach for multi-dimensional regression, the method is applied to luminescence sensing. Here the MTL architecture allows predicting multiple parameters, the oxygen concentration and the temperature, from a single set of measurements.

Background: Animal-assisted therapy has positive effects on cognitive function, depression, performance ability, and social functioning in elderly patients. The aim of this study was to evaluate the effects of rearing pet insects on the cognitive function of healthy elderly participants, with fMRI (functional magnetic resonance imaging) being used for this purpose. Methods: Community-dwelling elderly women (≥60 years) with normal cognitive function were enrolled during April 2015. They were randomized at a 1:1 ratio into two groups: insect-rearing and control (n=16) groups, with the insect-rearing group being further classified into two groups for analysis according to the subjects’ scores in the Wisconsin Card Sorting Test, WCST) at the first fMRI: insect-rearing group I with a relatively high score (n=13), and insect-rearing group II with a relatively low score (n=6). All subjects were educated on a healthy lifestyle for better cognitive function at every visit, and the insect-rearing groups received and reared crickets as pet insects. The fMRI was performed at baseline and after 8 weeks using the WCST as a stimulus. The WCST consisted of two variations, a high level baseline (HLB) and semi-WCST version. Results: There were no significant differences in the baseline characteristics among the three groups. There was a significant difference accuracy of the HLB–semi-WCST (p<0.05) in insect-rearing group II after 8 weeks from the baseline test. In the fMRI analysis involving the WCST reaction test, increased activation was observed in the right dorsal lateral prefrontal cortex and parietal cortex in insect-rearing group II when the semi-WCST, rather than the HLB, was performed. There were no significant differences in the other groups. Conclusion: The rearing of pet insects as an animal-assisted therapy is cost-effective, easy, and occupies little space. In this study, it showed positive effects on executive functions and performance improvement in elderly women. Further larger studies on the effects of pet insects on cognitive function are warranted.

1) Background: A population-based cross-sectional study was conducted to understand how green cover and accessibility of common public open spaces in compact urban areas affect physical activeness of resident. 2) Methods: A total of 554 residents completed a structured questionnaire on quality-of-life, physical activity level, and healthy eating practice. 3) Results: The sampled population lived with green cover averaged 10.11 ± 7.95% (ranged 1.56–9.88), whereas majority (90%) of the residents performed physical activities at medium and high levels. Metabolic Equivalent of Task (MET)-minutes/week was associated with the green cover percentage (Pearson r = 0.092; p < 0.05). Irrespective of age and physical activity level, active residents commonly used public open spaces within district for performing exercise, especially parks and promenade were mostly used by older residents while sports facilities by the younger groups. 4) Conclusions: Current findings suggested promotion of exercise could be achieved by the design or redesign of built environment to include more parks accessible to the residents with the increase of vegetation.

The rapid growth in the number of interconnected devices on the Internet (referred to as the Internet of Things – IoT), along with the huge volume of data that are exchanged and processed, has created a new landscape in network design and operation. Due to the limited battery size and computational capabilities of IoT nodes, data processing usually takes place on external devices. Since latency minimization is a key concept in modern era networks, edge servers that are in close proximity to IoT nodes gather and process related data, while in some cases data offloading in the cloud might have to take place. The interconnection of a vast number of heterogeneous IoT devices with the edge servers and the cloud, where IoT, edge and cloud converge to form a computing continuum, is also known as IoT-edge-cloud (IEC) continuum. Several key challenges are associated with this new computing systems architectural approach, including: i) design of connection and programming protocols aimed at properly manipulating a huge number of heterogeneous devices over diverse infrastructures; ii) design of efficient task offloading algorithms aimed at optimizing services execution; iii) support for security and privacy enhancements during data transfer to deal with the existent and even unforeseen attacks and threats landscape; iv) scalability, flexibility and reliability guarantees to face the expected mobility for IoT systems, and; v) design of optimal resource allocation mechanisms to make the most out of the available resources. These challenges become even more significant towards the new era of sixth generation (6G) networks, which will be based on the integration of various cutting edge heterogeneous technologies. Therefore, the goal of this survey paper is to present all recent developments in the field of IEC continuum systems, with respect to the aforementioned deployment challenges. In the same context, potential limitations and future challenges are highlighted as well. Finally, indicative use cases are also presented, from an IEC continuum perspective.

Transportation tasks in nursing are common and shorten the time for direct patient care in times of nursing shortage and an aging society. There is a lack of research on transportation needs, with most studies focusing on overall nursing task analysis without a deeper analysis of transportation tasks. In July and September 2023, we conducted weekly observations of nurses to assess clinical transportation needs. We aim to understand the economic impact and the methods nurses use for transportation tasks. We conducted a participant observation using a standardized app-based form over a seven-day observation period in two rural clinics. N=1,830 transports were made by nurses and examined by descriptive analysis. Non-medical supplies account for 27.05 % (n=495) of all transports, followed by medical supplies at 17.32 % (n=317), pharmacotherapy at 14.10 % (n=258) and other other categories like meals or drinks contributing 12.68 % (n=232). Most transports had a factual transport time of under a minute, with patient transport and lab samples displaying more variability. In total, 77.15 % of all transports were made by hand. Requirements to collect items or connect transports with patient care were included in 5 % of all transports. Our economic evaluation spotlighted meals as the most costly transport, with 9,596.16€ p.a. in the observed clinics. Budget-friendly robots would amortize over one year by transporting meals. We support understanding nurses’ transportation needs via further research on assistive robots to validate our findings and determine the feasibility of transport robots.

Continual learning (CL), a.k.a lifelong learning, is an emerging research topic that has been attracting increasing interest in the field of machine learning. With human activity recognition (HAR) playing a key role in enabling numerous real-world applications, an essential step towards the long-term deployment of such systems is to extend the activity model to dynamically adapt to changes in people’s everyday behavior. Current research in CL applied to HAR domain is still under-explored with researchers exploring existing methods developed for computer vision in HAR. Moreover, analysis has so far focused on task-incremental or class-incremental learning paradigms where task boundaries are known. This impedes the applicability of such methods for real-world systems. To push this field forward, we build on recent advances in the area of continual learning and design a lifelong adaptive learning framework using Prototypical Networks, LAPNet-HAR, that processes sensor-based data streams in a task-free data-incremental fashion and mitigates catastrophic forgetting using experience replay and continual prototype adaptation. Online learning is further facilitated using contrastive loss to enforce inter-class separation. LAPNet-HAR is evaluated on 5 publicly available activity datasets in terms of its ability to acquire new information while preserving previous knowledge. Our extensive empirical results demonstrate the effectiveness of LAPNet-HAR in task-free CL and uncover useful insights for future challenges.

On the thirty-fifth anniversary of the first description of O2-sensitive K+ channels in the carotid body chemoreceptors [1], O2 sensing remains a salient issue in the literature. Whereas much has been learned about this subject, important questions such as the identity of the specific K+ channel subtype(s) responsible for O2 sensing by chemoreceptors and the mechanism(s) by which their activities are altered by hypoxia have not yet been definitively answered. O2 sensing is a fundamental biological process necessary for the acute and chronic responses to varying environmental O2 levels which allow organisms to adapt to hypoxia. Whereas chronic responses depend on the modulation of hypoxia-inducible transcription factors which determine the expression of numerous genes encoding enzymes, transporters and growth factors, acute responses rely mainly on the dynamic modulation of ion channels by hypoxia, causing adaptive changes in cell excitability, contractility, and secretory activity in specialized tissues. The most widely studied oxygen-sensitive ion channels are potassium channels but oxygen sensing by members of both the calcium and sodium channel families has also been demonstrated. Given the explosion of information on this topic, in this review we will focus on the mechanisms of physiological oxygen chemotransduction by PO2-dependent K+ channels, with particular emphasis on their function in carotid body chemoreceptor cells (CBCC) and pulmonary artery smooth muscle cells (PASMC), highlighting areas of consensus and controversy within the field. We will first describe the most well-established concepts, those reproduced in multiple laboratories, and then discuss selected observations or questions that remain unresolved, and that limit our progress in this field.

Mirror Therapy (MT) and Cognitive Therapeutic Exercise (CTE) are two neurorehabilitation techniques based on neuroplasticity and designed to enhance the functionality of the upper limb. Therefore, the aim is to assess the effectiveness of the combination of CTE and MT with task-oriented training (TOT) on fine and gross motor skills of the affected hand, as well as the functionality of the affected upper limb, autonomy, independence, and quality of life (QoL) in patients with acute stroke. This longitudinal and multicenter study involved a sample of 120 participants selected randomly and consecutively. Three groups were formed, including a control group (CG) and two treatment groups, one with MT and the other with CTE. The functionality of the upper limb, fine and gross motor skills, the quantity and quality of use of the affected upper limb, autonomy, and QoL were evaluated using various scales validated in the Spanish population at the baseline, after 20 treatment sessions, and at three months of follow-up. ANCOVA analysis was conducted, revealing statistically significant differences between assessments in the CG compared to the experimental groups (GEs), indicating a notable improvement in overall functionality, fine and gross motor skills, autonomy, and QoL in the latter. However, there were no significant differences between the GEs. Therefore, the application of either of these experimental combinations is crucial for the motor recovery, functionality, and improvement of the quality of life of individuals who have suffered a stroke.

This work presents noble composites of AAILs@MCM-48 by functionalizing ordered mesopo-rous silicon MCM-48 with two amino acid-based ionic liquids (AAILs) ([Emim][Gly] and [Em-im][Ala]) to improve carbon capture and selectivity of CO2 over nitrogen. Thermogravimetric and XRD analyses of the composites show that the MCM-48 support's thermal and structural in-tegrity was preserved after the AAILs were encapsulated. An N2 adsorption-desorption study at 77 K also confirms AAIL encapsulation in the porous support. Under post-combustion flue gas conditions, both [Emim][Gly]@MCM-48 and [Emim][Ala]@MCM-48 demonstrated improved CO2 adsorption in comparison to unmodified MCM-48, with a CO2 partial pressure of around 0.15 bar. In terms of maximal CO₂ uptake, the 40wt%-[Emim][Gly] composite outperformed the others at 303K, with values of 0.74 and 0.82 mmol g-1, respectively, at 0.1 and 0.2 bar. These numbers show a 10x and 5x increase, respectively, as compared to the pure MCM-48 under iden-tical conditions. In addition, the selectivity of the composites was improved significantly, at 0.1 bar, the selectivity of composites containing 40wt% [Emim][Ala] increased to 17, compared to 2 for pristine MCM-48.

This study presents the encapsulation of two amino acid-based ionic liquids (AAILs), 1-Ethyl-3-methylimidazolium glycine [Emim][Gly] and 1-Ethyl-3-methylimidazolium alanine [Emim][Ala], in a highly porous metal organic framework (MOF-177), to generate state-of-the-art composites for post-combustion CO2 capture. The thermogravimetric analysis (TGA) demonstrated a successful encapsulation of the AAILs, consequently dramatically reducing the composites' surface area and pore volume. Both [Emim] [Gly]@MOF-177 and [Emim][Ala]@MOF-177 had close to 3 times the CO2 uptake of MOF-177 at 20 wt.% loading, 0.2 bar, and 303 K. Additionally, 20-[Emim][Gly]@MOF-177 and 20-[Emim] [Ala]@MOF-177 enhanced their CO2/N2 selectivity from 5 (pristine MOF-177) to 13 and 11, respectively.

Biosensors and wearable sensor systems with transmitting capabilities are currently developed and used for the monitoring of health data, exercise activities, and other performance data. Unlike conventional approaches, these devices enable convenient, continuous, and unobtrusive monitoring of a user’s behavioral signals in real time. Examples include signals relative to hand an finger movement/pressure control reflected by individual grip force data. As will be shown here, these directly translate into task, skill and hand-specific (dominant versus non-dominant hand) grip force profiles for different measurement loci in the fingers and palm of the hand. On the basis of thousands of sensor data from multiple sensor locations, individual grip force profiles of an task expert, a trained user and a highly proficient user (expert) performing an image-guided and robot-assisted precision task with the dominant or the non-dominant hand are analyzed in several steps following Tukey’s “detective work” approach. Correlation analyses (Person’s Product Moment) reveal skill-specific differences in individual grip force profiles across multiple sources of variation, functionally mapped to the somatosensory brain networks which ensure grip force control and its evolution with control expertise. Implications for the real-time monitoring of individual grip force profiles and their evolution with training in complex task-user systems are brought forward.

Engineering application of additively manufactured (AM) metallic materials is quite limited by their fatigue behaviors which are very inconsistent with that of conventionally wrought or cast ones. Here, based on advanced material characterization techniques, such as focused ion beam (FIB), scanning electron microscopy (SEM) and transmission electron microscopy (TEM), the mi-crostructures underneath fracture surfaces were thoroughly investigated in an AM aluminium (AlSi10Mg) alloy with horizontal and vertical building orientation endured very-high-cycle fa-tigue (VHCF) loading under stress ratios R = –1, 0 and 0.5. Two VHCF failure specimens A and B were representatively selected to further SEM examine and TEM samples preparation via FIB milling. Specimen A was horizontally printed and failed at R = –1; specimen B was vertically printed and failed at R = 0. TEM samples A1 and B1 were lifted from locations near the crack ini-tiation sites on the fracture surfaces of specimens A and B; The locations of TEM samples A2 and B2 kept away from the crack origin sites but still within the “fish-eye” region of crack steady growth. TEM observations show that there was no characteristic microstructure induced by VHCF in different oriented specimens and under various R values.

Electricity load forecasting is a crucial undertaking within all the deregulated markets globally. In contemporary times, the transition from conventional electricity grids to Smart Grids constitutes an area where extensive research is conducted on a global scale. Among the research challenges, the investigation of Deep Transfer Learning (DTL) in the field of electricity load forecasting represents a fundamental effort that imparts generality to Artificial Intelligence applications, due to new capabilities, such as knowledge transfer and computational power reduction. In this paper a comprehensive study is conducted for day-ahead electricity load forecasting. For this purpose, three Sequence-to-Sequence (Seq2seq) Deep Learning (DL) models are used, namely the Multilayer Perceptron (MLP), the Convolutional Neural Network (CNN) and the Ensemble Learning Model (ELM), which is consisted of the weighted combination of the outputs of MLP and CNN models. Also, the study focuses on the development of different forecasting strategies based on DTL and emphasizing the way the datasets are trained and fine-tuning for higher forecasting accuracy. In order to implement the forecasting strategies using Deep Learning models, load datasets from three Greek islands, Rhodes, Lesvos, and Chios, are used. The main purpose is to apply DTL for day-ahead predictions (1-24 hours) for each month of the year for Chios dataset after training and fine-tuning the models using the datasets of the three islands in various combinations. After several trials, four DTL strategies are illustrated. In the first strategy (DTL Case 1), each of the three DL models is trained using only Lesvos dataset, while fine-tuning is performed on the dataset of Chios island, in order to create day-ahead predictions for Chios load. In the second strategy (DTL Case 2), data from both Lesvos and Rhodes concurrently are used for the DL model training period, and fine-tuning is performed on the data from Chios. The third DTL strategy (DTL Case 3) contains the training of the DL models using Lesvos dataset, and the testing period performed directly on the Chios dataset without fine-tuning. The fourth strategy is a Multi-task Deep Learning (MTDL) approach, which has been extensively studied in recent years. In MTDL, the three DL models are trained simultaneously on all three datasets and the final predictions are made on the unknown part of the dataset of Chios. In this paper, we explore the performance of DTL and compare the results with those produced with MTDL. The results demonstrated that DTL can be applied with high efficiency for day-ahead load forecasting. Specifically, the two cases with fine-tuning (DTL Case 1 and 2) outperformed MTDL in terms of load prediction accuracy. Regarding the DL models, all three exhibit very high prediction accuracy, especially in the two cases with fine-tuning. The ELM excels compared to the single models. More specifically, for conducting day-ahead predictions, it has been concluded that the MLP model presents best monthly forecasts with a MAPE of 6.24% and 6.01% for the first two cases, the CNN model presents best monthly forecasts with a MAPE of 5.57% and 5.60% respectively and the ELM model achieves best monthly forecasts with a MAPE of 5.29% and 5.31%, respectively, indicating the very high accuracy it can achieve.

Pain is a multifaceted multisystem disorder that adversely affects neuro-psychological processes. This study compares the effectiveness of central-stimulation (tDCS) and peripheral-stimulation (TENS) in pain-inhibition during a cognitive-task in healthy volunteers and to observe potential neuro-cognitive improvements. Healthy eighty volunteers went through a complete experimental protocol comprising pain-induction (CPT), administration of tDCS/TENS, EEG-data acquisition, and cognitive assessments. The protocol for this study is categorized into 04-groups: G1(Control), G2(TENS), G3(Anodal-tDCS), and G4(Cathodal-tDCS). Paired t-tests (p

In the dynamic and complex environment of industrial control rooms, operators are often inundated with a multitude of tasks and alerts, which can lead to an overwhelming situation known as task overload. This state can precipitate decision fatigue and a heavier reliance on cognitive biases, potentially compromising the decision-making process. To mitigate such risks, the implementation of decision support systems becomes crucial. These systems are designed to assist operators in making swift and well-informed decisions, particularly when they sense their own judgment may be faltering. Our research introduces an AI-based framework that leverages dynamic influence diagrams and reinforcement learning to construct an effective decision support system. The cornerstone of this framework is the creation of a robust, interpretable, and efficient tool that supports control room operators during critical process disturbances. By integrating expert knowledge, the dynamic influence diagram forms a comprehensive model that captures the uncertainties inherent in complex industrial processes. It is adept at anomaly detection and recommending optimal actions. Moreover, this model is enhanced through a strategic partnership with reinforcement learning algorithms, which fine-tune the recommendations to be more context-specific and precise. The ultimate goal of our framework is to provide operators with a live, reliable decision support system that can significantly improve their response during process upsets. This paper outlines the development of our AI framework and its application within a simulated control room environment. Our findings indicate that the use of decision support systems can lead to improved operator performance and reduced cognitive workload. However, it also reveals a trade-off with situation awareness, which tends to diminish as operators may become overly reliant on the system’s guidance. Trust emerges as a critical factor in the adoption and effectiveness of decision support systems. Our research underscores the importance of balancing the benefits of decision support with the need for maintaining operator engagement and comprehension during process operations.

The paper provides a comprehensive review of the recent advancements and methodologies in Human-Robot Collaboration (HRC) applied to the manufacturing assembly process. In modern manufacturing, the assembly process involves intricate and time-consuming operations, often necessitating flexible manual interventions. However, the cost and stability issues associated with manual labor highlight the need for collaborative solutions integrating humans and robots. HRC, as a viable solution, involves the joint effort of humans and robots in manufacturing tasks, presenting advantages in terms of precision, reproducibility, and cycle time. This review categorizes and discusses methodologies such as task allocation, reinforcement learning, and Cyber-Physical Systems (CPS)-based planning approaches that facilitate HRC in the assembly process. It also explores experiments and future trends to address challenges and enhance efficiency in manufacturing assembly through intelligent collaboration between humans and robots. The objective of this research is to provide insights and directions for further research in HRC to optimize manufacturing processes. By analyzing the existing state-of-the-art and presenting future prospects, this paper aims to guide researchers and practitioners toward more effective implementations of HRC in manufacturing assembly, ultimately leading to improved operational efficiency and productivity.

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.

Nowadays, brain signal processing is performed rapidly in various brain-computer interface (BCI) applications. Most researchers focus on developing new methods for the future or improving the basic implemented models to identify the optimum standalone feature set. Our research focuses on four ideas. One of them introduces future communication models, and the others are for improving old models or methods. These are: 1) new communication imagery model instead of speech imager using the mental task: Due to speech imagery is very difficult, and it is impossible to imagine sound for all of the characters in all of the languages. Our research introduces a new mental task model for all languages that call Lip-sync imagery. This model can use for all characters in all languages. This paper implemented two lip-sync for two sounds, characters or letters. 2) New combination Signals: Selecting an inopportune frequency domain can lead to inefficient feature extraction. Therefore, domain selection is so important for processing. This combination of limited frequency ranges proposes a preliminary for creating Fragmentary Continuous frequency. For the first model, two s intervals of 4 Hz as filter banks were examined and tested. The primary purpose is to identify the combination of filter banks with 4Hz (scale of each filter bank) from the 4Hz to 40Hz frequency domain as new combination signals (8Hz) to obtain well and efficient features using increasing distinctive patterns and decreasing similar patterns of brain activities.3) new supplement bond graph classifier for SVM classifier: When SVM linear uses in very noisy, the performance is decreased. But we introduce a new bond graph linear classifier to supplement SVM linear in noisy data. 4) a deep formula recognition model: it converts the data of the first layer into a formula model (formula extraction model). The main goal is to reduce the noise in the subsequent layers for the coefficients of the formulas. The output of the last layer is the coefficients selected by different functions in different layers. Finally, the classifier extracts the root interval of the formulas, and the diagnosis does based on the root interval. For all of the ideas achieved the results of implementing methods. The results are between 55% to 98%. Less result is 55% for the deep detection formula, and the highest result is 98% for new combination signals.

This paper presents MADS (Multi-Agents for Data Science), an innovative and simple open-source multi-agent framework designed for systemic pipeline execution in applied supervised machine learning. By leveraging the capabilities of multi-agent systems (MAS), we introduce a universal approach to optimize and streamline machine learning pipelines. Our framework highlights the differences between various types of agents, such as reinforcement learning (RL) agents and large language model (LLM) agents, and their distinct contributions to the process. While we currently employ LLM agents to automate and enhance machine learning tasks, we acknowledge the potential of incorporating RL agents in future iterations to further improve performance and adaptability. The primary objective is to enhance the efficiency, scalability, and adaptability of supervised learning applications across various domains. This integration addresses the complexity and manual effort typically associated with machine learning workflows, paving the way for more automated, robust, and scalable solutions. Our approach demonstrates significant improvements in task automation, reduced human intervention, and enhanced model performance. The MADS framework, which will soon be available as an open-source implementation, represents a pivotal contribution to the field of machine learning, promising to facilitate broader adoption and collaborative advancement.