Robust Validation Research Articles

A Review on Machine Learning and Deep Learning Methods on Medical Image Classification

Medical image classification, a critical component in medical diagnostics, has significantly advanced through the integration of machine learning (ML) and deep learning (DL) techniques. This review comprehensively explores the evolution, methodologies, and applications of ML and DL in medical image classification. Traditional ML methods, including support vector machines and decision trees, have provided a foundation for early advancements by utilizing handcrafted features. However, the advent of DL, particularly convolutional neural networks (CNNs), has revolutionized the field by enabling automatic feature extraction and achieving superior performance. This review examines various DL architectures, such as ResNet, VGG, and Inception, highlighting their contributions to tasks like tumor detection, organ segmentation, and disease classification. Furthermore, it addresses challenges like data scarcity, interpretability, and computational demands, discussing potential solutions like data augmentation, transfer learning, and model optimization. The review also considers the ethical implications and the need for robust validation to ensure clinical applicability. Through a comparative analysis of existing studies, this review underscores the transformative impact of ML and DL on medical imaging, emphasizing the continuous need for innovation and interdisciplinary collaboration to enhance diagnostic accuracy and patient outcomes.

PHA4GE quality control contextual data tags: standardized annotations for sharing public health sequence datasets with known quality issues to facilitate testing and training.

As public health laboratories expand their genomic sequencing and bioinformatics capacity for the surveillance of different pathogens, labs must carry out robust validation, training, and optimization of wet- and dry-lab procedures. Achieving these goals for algorithms, pipelines and instruments often requires that lower quality datasets be made available for analysis and comparison alongside those of higher quality. This range of data quality in reference sets can complicate the sharing of sub-optimal datasets that are vital for the community and for the reproducibility of assays. Sharing of useful, but sub-optimal datasets requires careful annotation and documentation of known issues to enable appropriate interpretation, avoid being mistaken for better quality information, and for these data (and their derivatives) to be easily identifiable in repositories. Unfortunately, there are currently no standardized attributes or mechanisms for tagging poor-quality datasets, or datasets generated for a specific purpose, to maximize their utility, searchability, accessibility and reuse. The Public Health Alliance for Genomic Epidemiology (PHA4GE) is an international community of scientists from public health, industry and academia focused on improving the reproducibility, interoperability, portability, and openness of public health bioinformatic software, skills, tools and data. To address the challenges of sharing lower quality datasets, PHA4GE has developed a set of standardized contextual data tags, namely fields and terms, that can be included in public repository submissions as a means of flagging pathogen sequence data with known quality issues, increasing their discoverability. The contextual data tags were developed through consultations with the community including input from the International Nucleotide Sequence Data Collaboration (INSDC), and have been standardized using ontologies - community-based resources for defining the tag properties and the relationships between them. The standardized tags are agnostic to the organism and the sequencing technique used and thus can be applied to data generated from any pathogen using an array of sequencing techniques. The tags can also be applied to synthetic (lab created) data. The list of standardized tags is maintained by PHA4GE and can be found at https://github.com/pha4ge/contextual_data_QC_tags. Definitions, ontology IDs, examples of use, as well as a JSON representation, are provided. The PHA4GE QC tags were tested, and are now implemented, by the FDA's GenomeTrakr laboratory network as part of its routine submission process for SARS-CoV-2 wastewater surveillance. We hope that these simple, standardized tags will help improve communication regarding quality control in public repositories, in addition to making datasets of variable quality more easily identifiable. Suggestions for additional tags can be submitted to PHA4GE via the New Term Request Form in the GitHub repository. By providing a mechanism for feedback and suggestions, we also expect that the tags will evolve with the needs of the community.

Abstract A003: Unwinding the complexities of helicases as compelling drug targets in oncology

Abstract In recent years, helicases have moved to the forefront of research as novel drug targets for a variety of human diseases, including cancer. As critical components of the cell cycle and DNA repair, these enzymes offer an attractive point of therapeutic intervention. Recent bioinformatic endeavours such as the Cancer DepMap have further highlighted the essentiality of these proteins in the progression of numerous cancers. Inhibitors of Werner helicase (WRN) entered Phase I evaluation in 2023, after studies highlighted its role in tumours with microsatellite instability (MSI) and impaired mismatch repair (MMR). Building on these advances, helicases now represent a significant drug target class and are the subject of intense research. However, these targets present significant complexities in their prosecution. Hit finding and robust validation of these putative active compounds remains challenging, with high attrition rates and significant sensitivity to false positives. Resultant hits require thorough biological characterisation via both biochemical and biophysical methods, alongside determination of selectivity against other closely related family members, to increase confidence in their provenance before commencing a detailed drug discovery project. To exemplify these approaches, we have profiled three commercially available WRN inhibitors from two chemical series through a selection of biochemical and biophysical assays. The three compounds exhibit excellent selectivity against 6 distinct helicase/translocase targets. In Michaelis-Menten mechanism of Inhibition studies, we defined contrasting mechanisms of inhibition for the two series and were also able to identify time dependent inhibition with one compound series. Analysing binding interactions of the compounds against protein and protein:substrate complexes, via a fluorescent thermal shift assay (FTSA) and microscale thermophoresis (MST), enabled us to further understand and interrogate the mechanism of inhibition. Herein, we describe some of our experiences, learnings and best practices when prosecuting these compelling, but challenging, targets. Citation Format: Yael Mamane, Zoe Caple, Katie Chapman, Ian Henderson, Allan Jordan, Vanessa Lyne, Cynthia Okoye, Laurent Rigoreau, Ganesh Kadamur, Stuart Thomson, Chris Tomlinson, Jana Wolf. Unwinding the complexities of helicases as compelling drug targets in oncology [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Expanding and Translating Cancer Synthetic Vulnerabilities; 2024 Jun 10-13; Montreal, Quebec, Canada. Philadelphia (PA): AACR; Mol Cancer Ther 2024;23(6 Suppl):Abstract nr A003.

Development and validation of a questionnaire about reasons for academic cheating by nursing students: A cross-sectional study.

Cheating behavior is spreading among nursing students worldwide, necessitating the development of a validated questionnaire evaluating the reasons for such behavior. Nursing students (N = 482) from 2 universities in Saudi Arabia participated in this observational study. A survey containing items on socio-demographics and the 33-item Reasons for Cheating Scale (RCS) was completed by the respondents. The RCS had a 1-factor structure; the model fit indices were similar between the 1-, 2-, and 3-factor models, but the inter-factor correlations were too high for the 2- and 3-factor models. The measures of the quality of the factor score estimates were as follows: factor determinacy index, 0.987; expected a posteriori marginal reliability, 0.974; sensitivity ratio, 6.178; and expected percentage of true differences, 97.3%. The measures of the closeness to unidimensionality for the overall RCS were as follows: unidimensional congruence, 0.957; explained common variance, 0.875; and mean item residual absolute loading, 0.223. The intraclass correlation coefficient and McDonald's omega were 0.96 (CI: 0.93-0.98) and 0.962 (95% CI: 0.958-0.967), respectively. The severity score, infit, and outfit ranged from -0.847 to -2.015, 0.813 to 1.742, and 0.837 to 1.661, respectively. For all RCS items, the thresholds ranked τi1 < τi2 < τi3 < τi4 and showed invariance between the sexes. The RCS showed robust psychometric validity for both classical and item response theory parameters. It also had excellent test-retest reliability, internal consistency, item discrimination, factorial validity, measurement invariance, and ordered threshold level for the responses. Therefore, the RCS is a valid and reliable tool for assessing cheating behavior among nursing students.

Italian Grief and Meaning Reconstruction Inventory (GMRI): Exploratory, Confirmatory and Network Analyses

The current study was conducted to develop and validate an Italian translation of the Grief and Meaning Reconstruction Inventory (GMRI) and evaluate its psychometric properties in a sample of 297 bereaved adults. Exploratory and confirmatory factor analyses established that the 16-item Italian GMRI yielded five clearly interpretable factors (Continuing Bonds, Personal Growth, Sense of Peace, Valuing Life and Emptiness & Meaninglessness). The instrument demonstrated good reliability and robust convergent validity, through the scale’s significant inverse correlation with the Inventory of Complicated Grief and general distress, depression, stress, and anxiety. Network analysis of the items indicated that Continuing Bonds and Valuing Life could potentially assume central positions in the process of constructing meaning in the face of bereavement. In summary, the study suggests that the Italian version of the GMRI is valid and reliable instrument for assessing reconstruction of meaning after the loss of a loved one among the bereaved Italian population.

The influence of agility on green purchasing practices: the role of supplier governance and supplier relationships

PurposeThis paper aims to illuminate the nuanced dynamics of green supply chain management (GSCM), specifically focusing on the intersections of supplier relationships, supplier governance and organizational agility. Recognizing a gap in the understanding of how these elements confluence to promote green purchasing, the paper uses a quantitative study on data collected from the Australian food industry. Advanced analysis techniques provide empirical evidence underscoring the pivotal roles these elements play, expanding on current GSCM literature within a resource-based view.Design/methodology/approachThis study, based on a questionnaire sent to Australian food professionals, used higher-order reflective constructs to assess supplier relationships and governance. Data was analyzed using partial least squares structural equation modeling and Hayes PROCESS, considering factors like firm revenue and manager experience. Both the reliability of measures and mediation hypotheses were stringently validated using established guidelines.FindingsThe comprehensive study validated supplier governance's key influence on green purchasing and supplier relationships. Notably, organizational agility emerged as a crucial mediator, underscoring the interplay of these constructs. Concurrently, the reflective measurement model exhibited robust validity and reliability. Interestingly, demographic factors such as company size, revenue and managerial experience showed no discernible impact on green purchasing practices.Practical implicationsIn the Australian food sector, supplier governance and relationships are pivotal for advancing green purchasing. This study emphasizes the value of organizational agility in amplifying these practices. Managers, when aligning with supplier relationships enhanced by communication and mutual aid, can foster robust green initiatives. Embracing these insights and the critical importance of supplier governance, managers can drive more sustainable, informed supply chain decisions in the industry.Originality/valueIn pursuit of understanding the relationship between supplier governance, supplier relationships and green purchasing, this research uniquely situates itself within the resource-based view (RBV) to reveal critical theoretical and practical implications. By focusing on the Australian food industry, the study spotlights the often-overlooked mediating role of organizational agility in linking supplier relationships with green purchasing efforts. In doing so, this research not only strengthens the argument for fortified supplier relationships – as a catalyst for enhancing agility and thereby green practices – but also re-contextualizes the RBV in a fresh light. This new perspective provides managers with an enriched model, emphasizing the imperative of solid supplier governance for sustainable, agile and green supply chain operations in the food domain.

Development and validation of the Chinese Rural Middle-aged and Elderly Health Literacy Scale

IntroductionA notable deficiency lies in the absence of tailored instruments for comprehensively evaluating the health literacy of middle-aged and elderly residents in rural areas of China. Given the context of...

Auto-machine learning for opportunistic thyroid nodule detection in lung cancer screening chest CT.

e13639 Background: Automated Machine Learning (Auto-ML) in medical imaging is a process that allows non-experts to utilize machine learning techniques, opening the door for non-coder physician-driven exploitation of the technology. Auto-ML was applied for opportunistic detection of thyroid nodules in the context of low-dose lung cancer screening chest CT, facilitated by an innovative platform integration. By leveraging scans originally intended for lung cancer screening, suspicious appearing asymptomatic thyroid nodules can also be screened for where technically feasible. Methods: CT scans from the National Lung Screening Trial (NLST) dataset were utilized. This dataset contains low-dose chest CT examinations initially used for lung cancer screening. Annotation was carried out within a no-code web-based platform, Gesund.ai. A board-certified diagnostic radiologist annotated 100 CT examinations. A case was labeled "suspicious" if it presented a thyroid nodule of 1 cm or greater in diameter, while cases with no such nodules were deemed "not suspicious." The Auto-ML feature was then engaged within the software platform for parameter adjustments to refine the model without the need for manual coding. This phase was characterized by an iterative exploration of model performance, facilitated by the platform’s robust validation tools. Results: Upon training and validation, the Auto-ML model, cultivated from 100 annotated cases and assessed against a separate set of 130 cases, demonstrated a discernment accuracy of 0.51 in identifying "suspicious" thyroid nodules with a sensitivity and precision of 0.51 and 0.62 respectively. The AUC for this class stands at 0.69, indicating a moderate ability to distinguish between the presence and absence of thyroid nodules. This metric, while foundational, offers insight into the model's potential efficacy in real-world diagnostic scenarios, reflecting the initial capabilities of the platform's Auto-ML integration in enhancing diagnostic processes. Conclusions: The findings from this study highlight the potential of Auto-ML in revolutionizing opportunistic screening for thyroid nodules via lung cancer CT. The initial accuracy rate underscores the necessity for further refinement and validation of the model, but also demonstrates the promising potential of Auto-ML. Institutional Auto-ML would include extraction of appropriate studies through the no-code software platform integrated directly within the PACS framework. This process would ensure data immobility and foster a secure environment for dataset creation. Annotation would be carried out by multiple physician imagers within the same facility, enhancing the data’s accuracy and comprehensiveness specific for the institution’s patient population. This exploration highlights the strategic role of Auto-ML in advancing cancer patient care through innovative technological integration.

Prediction of immune-related adverse events in urological cancer during checkpoint inhibitor immunotherapy through immunohistochemical analysis of tumorous LCP1/ADPGK.

e14699 Background: The expanded utilization of immune checkpoint inhibitors (ICIs) have successfully improved the outcome of cancer patients, yet the concomitant immune-related adverse events (irAEs) can lead to fulminant unfavorable consequences. Currently, a comprehensive approach to predict irAEs is lacking. Some relevant biomarkers derived from database analyses are still in the early stages of investigation and lack robust validation in clinical samples, including LCP1/ADPGK. Herein, we validated the performance of LCP1 and ADPGK in a retrospective patient cohort with advanced urological cancers undergoing ICIs treatment, while comparing the predictive efficacy between different algorithmic models. Methods: We retrospectively collected clinical data of patients with renal cell carcinoma (RCC), upper urinary tract epithelial carcinoma (UTUC), bladder cancer (BC) receiving anti-PD-1/PD-L1 treatment from 2020 to 2023 in Peking University First Hospital. Patients were categorized into irAEs and non-irAEs groups, with irAE assessments conducted by two expert urological oncologists. Immunohistochemistry (IHC) was performed on 5-µm-thick pre-treated FFPE tumor tissue sections. Results: A total of 112 patients were included in the study, comprising 60 with RCC, 21 with UTUC, and 31 with BC. Among them, 51 (45.5%) patients experienced irAEs, with 9 exhibiting severe irAEs of grade III-IV. Immunohistochemical analysis revealed higher expression levels of LCP1 and ADPGK in patients with irAEs compared to those without irAEs. The area under the receiver-operating characteristic curve (AUC) for predicting irAEs using LCP1 and ADPGK individually was 0.82 (p &lt; 0.0001, 95%CI = 0.74-0.90) and 0.86 (p &lt; 0.0001, 95%CI = 0.80-0.93), respectively. The combination of LCP1 and ADPGK demonstrated an improved AUC of 0.89 (p &lt; 0.0001, 95%CI = 0.83-0.95). Among different algorithms, the bivariate linear-regression and the geometric mean model show similar prediction effects. Notably, the combined LCP1+ADPGK model successfully predicted irAEs in RCC (AUC = 0.89, p &lt; 0.0001, 95%CI = 0.81-0.98), UTUC (AUC = 0.81, p = 0.0063, 95%CI = 0.63-0.99), and BC (AUC = 0.88, p = 0.0004, 95%CI = 0.75-1.00). The expression of LCP1/ADPGK was not associated with the type of cancers. Conclusions: The bivariate linear-regression model of LCP1 and ADPGK could accurately predict irAEs in patients with urological cancers, which may allow us to improve the risk-benefit balance for individuals considered for ICIs therapy.

Velocity-field based load estimation of an unsteady actuator disc

Traditional methods for determining the load of an object in unsteady cases, such as Control-volume Momentum Integration (CMI) and Noca methods, encounter challenges related to near-body acceleration and pressure estimation. This prompts the need for innovative techniques to overcome these limitations. This study introduces a novel vorticity-based approach to expand classical load determination methods. The numerical results of an unsteady actuator disc (AD) CFD simulation are used to verify and quantify the uncertainty of the three methods. In the AD model, the load is prescribed and used as a source term. The velocity and pressure fields are solved from the Navier-Stokes equations using the open-source CFD package OpenFOAM, which offers a robust validation framework. Systematic examination of the robustness and accuracy of the three methods across various load and motion unsteady scenarios reveals that the proposed vorticity-based method excels in steady-state scenarios, exhibiting the highest level of robustness and accuracy. In contrast, the traditional CMI method proves more robust and accurate in pure motion cases. In scenarios involving static and moving AD with load variations, the new vorticity-based method demonstrates superior robustness and accuracy, mainly when moderate vorticity dynamics. Conversely, in cases with increased vorticity dynamics in the wake, the accuracy of the vorticity-based method decreases, with the CMI method showing superiority.

A novel method for predicting ecological interactions with an unsupervised machine learning algorithm

Abstract This gap in knowledge regarding ecological interactions has prompted the development of various predictive approaches. Traditionally, ecological interactions have been inferred using traits. However, the lack of trait information for numerous organisms necessitates using phylogenetic data and statistical insights from interaction matrices for prediction. Previous studies have overlooked the validation of model‐predicted interactions. This study used a novel method in predicting ecological interactions using a self‐organizing map (SOM), an unsupervised machine learning algorithm. The SOM learns from the input interaction matrix by grouping the nodes into output layers based on their interactions. Subsequently, the trained model predicts the interactions as scores. To distinguish between interactions and non‐interactions, we employed F1 score maximization, setting scores above a specific threshold as interactions and the remainder as non‐interactions. We applied this method to three unipartite metawebs and one bipartite metaweb and subsequently validated the predicted interactions using two innovative approaches: taxonomic and interaction recovery validation. Our method exhibited outstanding predictive performance, particularly for large networks. Various binary classification performance indicators, including F1 score (0.84–0.97) and accuracy (0.97–0.99), indicated high performance. Moreover, the method generated minimal predicted interactions, signifying low noise in the predictions, particularly for large networks. Taxonomic validation excels in metawebs with a connectance &gt;0.1 but performs poorly in metawebs with very low connectance. In contrast, interaction recovery was most effective in larger metawebs. Our proposed method excels at making highly accurate predictions of ecological interactions with minimal noise, solely utilizing input interaction data without relying on traits or phylogenetic information regarding interacting nodes. These predictions are particularly precise for large networks, underscoring their potential to address knowledge gaps in emerging extensive metawebs. Notably, taxonomic validation and interaction recovery methods are sensitive to connectance and network size, respectively, suggesting prospects for developing robust interaction validation methods.

Diesel Fuel Injector Nozzle Reclamation

Diesel engines are integral to various industries due to their efficiency and durability. A key component of these engines is the fuel injector nozzle, which can degrade over time due to contaminants, carbon deposits, and mechanical wear. This paper examines the reclamation process for diesel fuel injector nozzles, highlighting its benefits, challenges, and the validation of reclaimed nozzles. Reclamation offers cost savings, environmental benefits, and extended component life, but requires stringent quality control and adaptation to technological advances. The paper underscores the importance of meticulous reclamation procedures and robust validation methods, such as dyno and on-road testing, to ensure the performance of reclaimed nozzles meets or exceeds original specifications. As diesel technology evolves, the reclamation industry must continuously adapt to maintain its relevance and efficacy.

Clinical Biomarkers of Tumour Radiosensitivity and Predicting Benefit from Radiotherapy: A Systematic Review.

Modern advanced radiotherapy techniques have improved the precision and accuracy of radiotherapy delivery, with resulting plans being highly personalised based on individual anatomy. Adaptation for individual tumour biology remains elusive. There is an unmet need for biomarkers of intrinsic radiosensitivity that can predict tumour response to radiation to facilitate individualised decision-making, dosing and treatment planning. Over the last few decades, the use of high throughput molecular biology technologies has led to an explosion of newly discovered cancer biomarkers. Gene expression signatures are now used routinely in clinic to aid decision-making regarding adjuvant systemic therapy. They have great potential as radiotherapy biomarkers. A previous systematic review published in 2015 reported only five studies of signatures evaluated for their ability to predict radiotherapy benefits in clinical cohorts. This updated systematic review encompasses the expanded number of studies reported in the last decade. An additional 27 studies were identified. In total, 22 distinct signatures were recognised (5 pre-2015, 17 post-2015). Seventeen signatures were 'radiosensitivity' signatures and five were breast cancer prognostic signatures aiming to identify patients at an increased risk of local recurrence and therefore were more likely to benefit from adjuvant radiation. Most signatures (15/22) had not progressed beyond the discovery phase of development, with no suitable validated clinical-grade assay for application. Very few signatures (4/17 'radiosensitivity' signatures) had undergone any laboratory-based biological validation of their ability to predict tumour radiosensitivity. No signatures have been assessed prospectively in a phase III biomarker-led trial to date and none are recommended for routine use in clinical guidelines. A phase III prospective evaluation is ongoing for two breast cancer prognostic signatures. The most promising radiosensitivity signature remains the radiosensitivity index (RSI), which is used to calculate a genomic adjusted radiation dose (GARD). There is an ongoing phase II prospective biomarker-led study of RSI/GARD in triple negative breast cancer. The results of these trials are eagerly anticipated over the coming years. Future work in this area should focus on (1) robust biological validation; (2) building biobanks alongside large radiotherapy randomised controlled trials with dose variance (to demonstrate an interaction between radiosensitivity signature and dose); (3) a validation of clinical-grade cost-effective assays that are deliverable within current healthcare infrastructure; and (4) an integration with biomarkers of other determinants of radiation response.

Artificial intelligence for gastric cancer in endoscopy: From diagnostic reasoning to market

Recognition of gastric conditions during endoscopy exams, including gastric cancer, usually requires specialized training and a long learning curve. Besides that, the interobserver variability is frequently high due to the different morphological characteristics of the lesions and grades of mucosal inflammation. In this sense, artificial intelligence tools based on deep learning models have been developed to support physicians to detect, classify, and predict gastric lesions more efficiently. Even though a growing number of studies exists in the literature, there are multiple challenges to bring a model to practice in this field, such as the need for more robust validation studies and regulatory hurdles. Therefore, the aim of this review is to provide a comprehensive assessment of the current use of artificial intelligence applied to endoscopic imaging to evaluate gastric precancerous and cancerous lesions and the barriers to widespread implementation of this technology in clinical routine.

Predicting malaria outbreak in The Gambia using machine learning techniques.

Malaria is the most common cause of death among the parasitic diseases. Malaria continues to pose a growing threat to the public health and economic growth of nations in the tropical and subtropical parts of the world. This study aims to address this challenge by developing a predictive model for malaria outbreaks in each district of The Gambia, leveraging historical meteorological data. To achieve this objective, we employ and compare the performance of eight machine learning algorithms, including C5.0 decision trees, artificial neural networks, k-nearest neighbors, support vector machines with linear and radial kernels, logistic regression, extreme gradient boosting, and random forests. The models are evaluated using 10-fold cross-validation during the training phase, repeated five times to ensure robust validation. Our findings reveal that extreme gradient boosting and decision trees exhibit the highest prediction accuracy on the testing set, achieving 93.3% accuracy, followed closely by random forests with 91.5% accuracy. In contrast, the support vector machine with a linear kernel performs less favorably, showing a prediction accuracy of 84.8% and underperforming in specificity analysis. Notably, the integration of both climatic and non-climatic features proves to be a crucial factor in accurately predicting malaria outbreaks in The Gambia.

Study on Spatialization and Spatial Pattern of Population Based on Multi-Source Data—A Case Study of the Urban Agglomeration on the North Slope of Tianshan Mountain in Xinjiang, China

Study on Spatialization and Spatial Pattern of Population Based on Multi-Source Data—A Case Study of the Urban Agglomeration on the North Slope of Tianshan Mountain in Xinjiang, China

Detecting Coseismic Landslides in GEE Using Machine Learning Algorithms on Combined Optical and Radar Imagery

Landslides, resulting from disturbances in slope equilibrium, pose a significant threat to landscapes, infrastructure, and human life. Triggered by factors such as intense precipitation, seismic activities, or volcanic eruptions, these events can cause extensive damage and endanger nearby communities. A comprehensive understanding of landslide characteristics, including spatio-temporal patterns, dimensions, and morphology, is vital for effective landslide disaster management. Existing remote sensing approaches mostly use either optical or synthetic aperture radar sensors. Integrating information from both these types of sensors promises greater accuracy for identifying and locating landslides. This study proposes a novel approach, the ML-LaDeCORsat (Machine Learning-based coseismic Landslide Detection using Combined Optical and Radar Satellite Imagery), that integrates freely available Sentinel-1, Palsar-2, and Sentinel-2 imagery data in Google Earth Engine (GEE). The approach also integrates relevant spectral indices and suitable bands used in a machine learning-based classification of coseismic landslides. The approach includes a robust and reproducible training and validation strategy and allows one to choose between five classifiers (CART, Random Forest, GTB, SVM, and Naive Bayes). Using landslides from four different earthquake case studies, we demonstrate the superiority of our approach over existing solutions in coseismic landslide identification and localization, providing a GTB-based detection accuracy of 87–92%. ML-LaDeCORsat can be adapted to other landslide events (GEE script is provided). Transfer learning experiments proved that our model can be applied to other coseismic landslide events without the need for additional training data. Our novel approach therefore facilitates quick and reliable identification of coseismic landslides, highlighting its potential to contribute towards more effective disaster management.

Psychometric validation of the Danish version of the Major Depression Inventory using data from the Lolland-Falster health study (LOFUS)

Purpose The Major Depression Inventory (MDI) is a widely used self-rating depression scale commonly in primary care in Denmark. It has not been subject to robust psychometric validation in a general population setting. The aim of this study was to evaluate the psychometric measurement properties of the MDI when applied in the general population. Methods We evaluated statistical psychometric validity using modern test theory (confirmatory factor analysis, item response theory models and Rasch measurement theory) testing local independence and differential item function across groups defined by gender, age, education, and chronic disease status. Separate analyses across different strata and across different statistical models were employed. Results Regarding structural validity we consistently identified local dependence for the item two pairs (MDI2,MDI3) and (MDI4,MDI5) across strata. This result was confirmed by bifactor CFA models and item screening. We further identified substantial differential item functioning with respect to age group and with respect to chronic disease. We identified quantified the magnitude of this lack of measurement invariance. Conclusion The MDI is psychometrically valid in homogenous sub populations, but the disclosed evidence of local dependence means that published estimates of its reliability cannot be trusted. The lack of measurement invariance means that the instrument cannot be used to compare individuals or groups unless they are similar in terms of age group and chronic disease status.

Optimizing Nanofluid Hybrid Solar Collectors through Artificial Intelligence Models

This study systematically explores and compares the performance of various artificial-intelligence (AI)-based models to predict the electrical and thermal efficiency of photovoltaic–thermal systems (PVTs) cooled by nanofluids. Employing extreme gradient boosting (XGB), extra tree regression (ETR), and k-nearest-neighbor (KNN) regression models, their accuracy is quantitatively evaluated, and their effectiveness measured. The results demonstrate that both XGB and ETR models consistently outperform KNN in accurately predicting both electrical and thermal efficiency. Specifically, the XGB model achieves remarkable correlation coefficient (R2) values of approximately 0.99999, signifying its superior predictive capabilities. Notably, the XGB model exhibits a slightly superior performance compared to ETR in estimating electrical efficiency. Furthermore, when predicting thermal efficiency, both XGB and ETR models demonstrate excellence, with the XGB model showing a slight edge based on R2 values. Validation against new data points reveals outstanding predictive performance, with the XGB model attaining R2 values of 0.99997 for electrical efficiency and 0.99995 for thermal efficiency. These quantitative findings underscore the accuracy and reliability of the XGB and ETR models in predicting the electrical and thermal efficiency of PVT systems when cooled by nanofluids. The study’s implications are significant for PVT system designers and industry professionals, as the incorporation of AI-based models offers improved accuracy, faster prediction times, and the ability to handle large datasets. The models presented in this study contribute to system optimization, performance evaluation, and decision-making in the field. Additionally, robust validation against new data enhances the credibility of these models, advancing the overall understanding and applicability of AI in PVT systems.

Uncovering novel drug targets for bipolar disorder: a Mendelian randomization analysis of brain, cerebrospinal fluid, and plasma proteomes.

There is a clear demand for innovative therapeutics for bipolar disorder (BD). We integrated the largest BD genome-wide association study (GWAS) dataset (NCase = 41 917, NControl = 371 549) with protein quantitative trait loci from brain, cerebrospinal fluid, and plasma. Using a range of integrative analyses, including Mendelian randomization (MR), Steiger filter analysis, Bayesian colocalization, and phenome-wide MR analysis, we prioritized novel drug targets for BD. Additionally, we incorporated data from the UK Biobank (NCase = 1064, NControl = 365 476) and the FinnGen study (NCase = 7006, NControl = 329 192) for robust biological validation. Through MR analysis, we found that in the brain, downregulation of DNM3, MCTP1, ABCB8 and elevation of DFNA5 and PDF were risk factors for BD. In cerebrospinal fluid, increased BD risk was associated with increased levels of FRZB, AGRP, and IL36A and decreased CTSF and LRP8. Plasma analysis revealed that decreased LMAN2L, CX3CL1, PI3, NCAM1, and TIMP4 correlated with increased BD risk, but ITIH1 did not. All these proteins passed Steiger filtering, and Bayesian colocalization confirmed that 12 proteins were colocalized with BD. Phenome-wide MR analysis revealed no significant side effects for potential drug targets, except for LRP8. External validation further underscored the concordance between the primary and validation cohorts, confirming MCTP1, DNM3, PDF, CTSF, AGRP, FRZB, LMAN2L, NCAM1, and TIMP4 are intriguing targets for BD. Our study identified druggable proteins for BD, including MCTP1, DNM3, and PDF in the brain; CTSF, AGRP, and FRZB in cerebrospinal fluid; and LMAN2L, NCAM1, and TIMP4 in plasma, delineating promising avenues to development of novel therapies.