Despite available global efforts and funding, Tuberculosis (TB) continues to affect a considerable number of patients worldwide. Policy makers and stakeholders set clear goals to reduce TB incidence and mortality, but the emergence of multidrug-resistant TB (MDR-TB) and extensively drug-resistant TB (XDR-TB) complicate the reach of these goals. Drug-resistance TB needs to be diagnosed rapidly and accurately to effectively treat patients, prevent the transmission of MDR-TB, minimise mortality, reduce treatment costs and avoid unnecessary hospitalisations. In this narrative review, we provide a comprehensive overview of laboratory methods for detecting drug resistance in MTB, focusing on phenotypic, molecular and other drug susceptibility testing (DST) techniques. We found a large variety of methods used, with the BACTEC MGIT 960 being the most common phenotypic DST and the Xpert MTB/RIF being the most common molecular DST. We emphasise the importance of integrating phenotypic and molecular DST to address issues like resistance to new drugs, heteroresistance, mixed infections and low-level resistance mutations. Notably, most of the analysed studies adhered to the outdated definition of XDR-TB and did not consider the pre-XDR definition, thus posing challenges in aligning diagnostic methods with the current landscape of TB resistance.

Background: The Covid-19 pandemic was in reality a syndemic there the balance between a multitude of biological (Nature-made) and socio-political (men-made) factors determined individually the outcome of every single infection, on a scale of nothing-to-fatal. Blaming the virus for everything bad that happened to us after January 2020 is incorrect. Now, in ~ 2 years distance from the syndemic, it may be timely and possible to uncover the human contribution to this historical event and learn of it. Objective: To identify and use some qualitative approach to estimate the objective, statistical / epidemiological ‘value’ of the virus-selected individuals (in age-normalized groups) for the rest of the society. This non-political, non-religious approach may be the only way to understand the benefit / cost ratio of actions and emergency orders we had to survive in addition to the virus itself. Methods: Statistical data were obtained from publicly available official databases, including the relevant actuarial tables and evaluated by widely used statistical methods. The estimation of statistical/epidemiological ‘value of life’ was based, strictly and exclusively, on the real age and the statistically calculated Residual Life Expectancy (RLE, i.e. remaining years alive at a given age). The American healthcare expenses were compared to the corresponding healthcare costs/capita in other similarly well-developed west-European countries. This was the only available method to get some ideas about the magnitude of the well-established (rational) healthcare spending and compare it to the American standards. Results 1) Life expectancy uninterruptedly increased in USA since 1933, due to continuous reduction of Mortality Rate of the youngest and oldest citizens. COVID didn’t stop the population growth (+8.2 M /2020-2023). 2) The total count of Excess COVID associated Deaths (XCD) is ~ 2-times as much as the official statistic was able to explain by virus infection. It is speculated that these Complementary Covid Deaths (CCD) are probably the unexpected side-effects of the pandemic related emergency orders. 3) The worldwide ‘longevity’ trend exponentially increased the healthcare expenses, especially the care of elderly. However the USA is unique in this aspect and the healthcare costs are irrationally much higher than in any other country. 4) We can see a large economic incentive in the development of excesses under COVID syndemic. Conclusions The harm of the COVID virus - as well as the benefits of the mandatory, emergency actions during the pandemic - were grossly over-estimated. Professional, unbiased (non-religious, non-political) and independent reviews are strongly recommended.

The aim of this report is to evaluate the segmental rib index (RI) from T1-T12 spinal level in mild and moderate idiopathic scoliosis (IS) curves of thoracic, thoracolumbar and lumbar type by gender. The relationship of segmental RI to the frontal plane radiological deformity presented as the Cobb angle and to the posterior truncal surface deformity presented as the scoliometric readings of Angle of Trunk Rotation (ATR) in these patients is also assessed. Any statistical significant rela-tionship of these parameters would be very important for biomechanical relations of the rib cage (RC) deformity presented as the rib hump deformity (RHD) and the deformity in the spine and it would thus provide valuable information about scoliogeny. The segmental rib index (RI) is presented in 83 boys and girls with mild and moderate IS. The measurements include the scoliometric readings for truncal asymmetry (TA), the Cobb angle assessment and the segmental RI from T1-T12. The statistical package SPSS 23 was used for statistical analysis. The TA is documented and the Cobb angle is presented by gender and curve type. The segmental RI of thoracic, thoracolumbar and lumbar curves are presented for the first time. The correlations of the segmental RI to surface deformity presented as the rib hump deformity (RHD) in all IS patients and particularly in thoracic curves, to Cobb angle by gender and age and the comparison of the segmental RI index of asymmetric but not scoliotic children to the scoliotic peers by curve (in thoracic, thoracolumbar, lumbar curves) in boys and girls are presented. The findings emphasize the significant protagonistic role of the thoracic asymmetry in relation to the spinal deformity mainly in girls for the thoracic and in boys for the thoracolumbar curves. The cut-off point of age of the examined scoliotics is 14 years, that is when the RI shows stronger correlation with the spinal deformity, namely when the thoracic deformity is decisively effective to the developing thoracic spinal deformity, in terms of Cobb angle. In summary, the results of this study provide aetiological implications for IS, as far as the role of thorax is con-cerned.

Cervical cancer is a worldwide public health problem with a high rate of illness and mortality among women. In this study, we proposed a novel framework based on Faster RCNN-FPN ar-chitecture for the detection of abnormal cervical cells in cytology images from cancer screening test. We extended the Faster RCNN-FPN model by infusing deformable convolution layers into the feature pyramid network (FPN) to improve scalability. Furthermore, we introduced a global contextual aware module alongside the Region Proposal Network (RPN) to enhance the spatial correlation between the background and the foreground. Extensive experimentations with the proposed deformable and global context aware (DGCA) RCNN were carried out using the cer-vical image dataset of “Digital Human Body" Vision Challenge from the Alibaba Cloud TianChi Company. Performance evaluation based on the mean average precision (mAP) and receiver operating characteristic (ROC) curve has demonstrated considerable advantages of the proposed framework. Particularly, when combined with tagging of the negative image samples using tra-ditional computer-vision techniques, 6-9% increase in mAP has been achieved. The proposed DGCA-RCNN model has potential to become a clinically useful AI tool for automated detection of cervical cancer cells in whole slide images of Pap smear.

Background: Cervical cancer prevention in regions with limited access to screening and HPV vaccination necessitates innovative approaches. This study explored the potential of a test-and-treat strategy using mRNA HPV tests to impact cervical cancer prevention in a high-prevalence HIV population. Methods: A cervical screening study was conducted at three South African hospitals involving 710 under-screened, non-pregnant women (25 to 65 years) without known cervical diseases. Cytology, HPV testing, colposcopy, and biopsies were performed concurrently. Histopathologists determined final histological diagnoses based on biopsy and LLETZ histology. mRNA-HPV-genotyping for 3 (16, 18, 45) to 8 (16, 18, 31, 33, 35, 45, 52, 58) high-risk types was performed on leftover liquid-based cytology material. The preventive potential of the test-and-treat approach was estimated based on published data, reporting the causative HPV types in cervical cancer tissue from South African women. Treatment was provided as needed. Results: HPV positivity rate more than doubled from 3-type (15.2%; 95% CI: 12.6-17.8) to 8-type mRNA (31.5%; 95% CI: 28.8-34.9) combinations, significantly higher among HIV-positive women. CIN3+ prevalence among HIV-positive women (26.4%) was double that of HIV-negative women (12.9%) (p<0.01). The 6-type combination showed the best balance of sensitivity, specificity and treatment group size, and effectiveness to prevent cervical cancer. A 4-type combination (16, 18, 35, 45) could potentially prevent 77.6% (95% CI: 71.2-84.0) of cervical cancer burden by treating 20% and detecting 41.1% of CIN3 cases in the study group. Similarly, a 6-type combination (16, 18, 31, 33, 35, 45), treating 25% and including 62% of CIN3 cases, might prevent 85% of cervical cancer cases (95% CI: 79.6-90.6) among HIV-positive and negative women. Conclusion: Employing mRNA HPV tests within a test-and-treat approach holds huge promise for targeted cervical cancer prevention in under-screened populations. Testing for mRNA of the 6 highest-risk HPV types in this population and treating them all is projected to effectively prevent progression from CIN3 to invasive cervical cancer while reducing overtreatment in resource-constrained settings.

Background: In chronic obstructive pulmonary disease (COPD), there are two known classifications for assessing what is called disease severity. One is the Global Initiative for Chronic Obstructive Lung Disease (GOLD) classification, which is based on the post-bronchodilator value of FEV1 (% reference). The other is the STaging of Airflow obstruction by Ratio (STAR). Methods: We reanalyzed data from our outpatient cohort study, which included 141 participants with COPD from 2015 to 2023. We compared mortality and COPD-specific health status between the GOLD 1 to 4 groups and the STAR 1 to 4 groups. Results: By simple calculation, GOLD and STAR severity classes coincided in 75 participants (53.2%). The weighted Bangdiwala B value with linear weights was 0.775. The participants were observed for up to 95 months, with a median of 54 months. Death was confirmed in 29 participants (20.5%). In univariate Cox proportional hazards analyses, there was a significant difference in mortality between the GOLD 1 and GOLD 3+4 groups, with the GOLD 1 group used as the reference [Hazard Ratio 4.222 (95% CI 1.298 - 13.733), p=0.017]. However, there was no statistically significant predictive relationship between STAR 1 and STAR 2, or between STAR 1 and STAR 3+4. The St. George's Respiratory Questionnaire (SGRQ) Total and COPD Assessment Test (CAT) scores were significantly different between all GOLD groups, except for the CAT score between GOLD 1 and GOLD 2. The SGRQ Total and CAT scores were significantly different between STAR 1 and STAR 3+4, but not between STAR 1 and STAR 2. Conclusion: From the perspective of all-cause mortality and COPD-specific health status, the GOLD classification is more discriminative than STAR.

Background: Open AI developed ChatGPT, a language model based on the GPT architecture, designed for text-based communication. Trained on diverse internet texts, ChatGPT generates contextually appropriate responses using machine learning. It understands input through analysis and context interpretation, generating coherent and contextually relevant responses. Interaction is possible through messaging platforms. Materials and Methods: In November 2023, we utilized ChatGPT 3.5, the default version at that time, to answer questions from the Italian National Residency Program Admission Tests (SSMs) of 2021, 2022, and 2023. The questions cover clinical, diagnostic, analytical, therapeutic, and epidemiological scenarios, sometimes accompanied by images. The study compared ChatGPT's answers to the official corrections on the Italian Ministry of University and Research (MUR) website. The scoring method used for evaluation was 1 point for correct answers, 0 points for unanswered questions, and -0.25 points for incorrect answers, reflecting the SSM test scoring system. Results: In summary, ChatGPT was tested with a total of 420 questions, 140 for each test. It achieved an overall accuracy of 80.48%, providing correct answers for 338 questions and incorrect answers for 82 questions. When faced with questions containing both text and images, it answered 55% correctly and 45% incorrectly. The model's performance varied over time, with an 82.14% accuracy rate in 2021 and 2022 (115 correct out of 140) and a 77.14% accuracy rate in 2023 (108 correct out of 140). Applying thisscoring method to the SSM test, ChatGPT would have scored 105 points in 2021 and 2022, and 100 points in 2023. Conclusions: ChatGPT has exhibited above-average performance in the last three SSM tests, highlighting its robust capability to interpret clinical scenarios and offer precise diagnostic and therapeutic guidance. Despite this, some limitations persist, notably the software's inability to interpret non-textual information.

This paper offers a comprehensive examination of the process involved in developing and automating supervised end-to-end machine learning workflows for forecasting and classification purposes. It offers a complete overview of the components (i.e. feature engineering, model selection, etc), principles (i.e. bias-variance decomposition, model complexity, overfitting, model sensitivity to feature assumptions and scaling, output interpretability, etc), models (i.e. neural networks, regression models, etc), methods (i.e. Cross-Validation, data augmentation, etc), metrics (i.e. Mean Squared Error, F1-score, etc) and tools that rule most supervised learning applications with numerical and categorical data, as well as their integration, automation, and deployment. The end goal and contribution of this paper is the education and guidance of the non-AI expert academic community over complete and rigorous machine learning pipelines and data science practices, from problem scoping to design and state-of-the-art automation tools, including basic principles and reasoning in the choice of methods. The paper delves into the critical stages of supervised machine learning workflow development, many of which are often omitted by researchers due to brevity, and covers foundational concepts essential for understanding and optimizing a functional machine learning workflow, thereby offering a holistic view of task-specific application development for applied researchers who are not AI experts.

Abstract: A prevalence boundary (PB) marks the point in prevalence where the false omission rate, RFO=FN/(TN+FN), exceeds the tolerance limit for missed diagnoses. Objectives were to mathematically analyze rapid antigen test (RAgT) performance, determine why PBs are breeched, and evaluate the merits of testing three times over five days, now required by the US Food and Drug Administration for asymptomatic persons. Equations were derived to compare test performance patterns, calculate PBs, and perform recursive computations. An independent July 2023 FDA-university-commercial evaluation of RAgTs provided performance data used in theoretical calculations. Tiered sensitivity/specificity comprise: Tier-1) 90%, 95%; Tier-2) 95%, 97.5%; and Tier-3) 100%, ≥99%, respectively. Repeating a T2 test improves the PB from 44.6% to 95.2% (RFO 5%). In the FDA-university-commercial evaluation, RAgTs generated sensitivity of 34.4%, which improved to 55.3% when repeated, then 68.5% with the third test. With RFO=5%, PBs were 7.37/10.46/14.22%, respectively. PB analysis suggests RAgTs should achieve clinically proven sensitivity of 91.0-91.4%. When prevalence exceeds PBs, missed diagnoses can perpetuate virus transmission. Repeating low-sensitivity RAgTs delays diagnosis. In homes, high-risk settings, and hotspots, PB breaches may prolong contagion, defeat mitigation, facilitate new variants, and transform outbreaks into endemic disease. Molecular diagnostics can help avoid these potential vicious cycles

The affinity constant, also known as the equilibrium constant, binding constant, equilibrium association constant, or the reciprocal value, the equilibrium dissociation constant (K_d), can be considered as one of the most important characteristics for any antibody-antigen pair. Many methods based on different technologies have been proposed and used to determine this value. However, since a very large number of publications and commercial datasheets do not include this information, significant obstacles in performing such measurements seem to exist. In other cases where such data are reported, the results have often proved to be unreliable. This situation may indicate that most of the technologies available today require a high level of expertise that does not seem to be available in many laboratories. In this paper, we present a simple approach based on standard immunoassay technology that is easy and quick to perform. It relies on the effect that the molar IC₅₀ approaches the K_d value in the case of infinitely small concentrations of the reagent concentrations. A two-dimensional dilution of the reagents leads to an asymptotic convergence to K_d. The approach has some similarity to the well-known checkerboard titration used for the optimization of immunoassays. A well-known antibody against the FLAG peptide, clone M2, was used as a model system and the results were compared with other methods. This approach could be used in any case where a competitive assay is available or can be developed. The determination of an affinity constant should belong to the crucial parameters in any quality control of antibody-related products and assays and should be mandatory in papers using immunochemical protocols.

The Mediterranean basin is a hot spot of climate change where the Posidonia oceanica (L.) Delile (PO) and other seagrass are under stress due to its effect on marine habitats and the rising influence of anthropogenic activities (tourism, fishery). The PO and seabed ecosystems, in the coastal environments of Pantelleria and Lampedusa, suffer additional growing impacts from tourism in synergy with specific stress factors due to increasing vessel traffic for supplying potable water, fossil fuels for electrical power generation. Earth Observation (EO) data, provided by high resolution (HR) multi/hyperspectral operative satellite sensors of the last generation (i.e. Sentinel 2 MSI and PRISMA) have been successfully tested, using innovative calibration and sea truth collecting methods, for monitoring and mapping of PO meadows under stress, in the coastal waters of these islands, located in the Sicily Channel, to better support the sustainable management of these vulnerable ecosystems. The area of interest in Pantelleria was where the first prototype of the Italian Inertial Sea Wave Energy Converter (ISWEC) for renewable energy production was installed in 2015, and sea truth campaigns on the PO meadows were conducted. The PO of Lampedusa coastal areas, impacted by ship traffic linked to the previous factors and tropicalization effects of Italy southernmost climate change transitional zone, was mapped through a multi/hyper spectral EO-based approach, using training/testing data provided by side scan sonar data, previously acquired. Some advanced machine learning algorithms (MLA) were successfully evaluated with different supervised regression/classification models to map seabed and PO meadow classes and related Leaf Area Index (LAI) distributions in the areas of interest, using multi/hyperspectral data atmospherically corrected via different advanced approaches.