Precision-recall Curve Research Articles

Radiomics-based predictive model for preoperative risk classification of gastrointestinal stromal tumors using multiparametric magnetic resonance imaging: aretrospective study.

The aim of this study was to develop and assess aradiomics model utilizing multiparametric magnetic resonance imaging (MRI) for the prediction of preoperative risk assessment in gastrointestinal stromal tumors (GISTs). An analysis was performed retrospectively on agroup of 121 patients who received ahistological diagnosis of GIST. They were then divided into two sets, with85 in the training set and 36 in the validation set through random partitioning. Radiomics features from five MRI sequences, totaling 600 per patient, were extracted and subjected to feature selection utilizing arandom forest algorithm. The discriminatory efficacy of the models was evaluated through receiver operating characteristic (ROC) and precision-recall (P-R) curve analyses. Model calibration was assessed via calibration curves. Subgroup analysis was performed on GISTs with apathological maximum diameter equal to or less than 5 cm. Furtherly, Kaplan-Meier (K-M) curves and log-rank tests were used to compare the differences in survival status among different groups. Cox regression analysis was employed to identify independent prognostic factors and to construct aprognostic prediction model. The clinical model (ModelC) displayed limited predictive efficacy in the context of GIST. Conversely, aradiomics model (ModelR) incorporating five parameters exhibited robust discriminative capabilities across both the training and validation sets, yielding area under the ROC curve (AUC) values of 0.893 (95% confidence interval [CI]: 0.807-0.949) and 0.855 (95% CI: 0.732-0.978), respectively. The F1max scores derived from the P‑R curves were 0.741 and 0.842 for the training and validation sets, respectively. Noteworthy was the exclusion of the two-dimensional tumor diameter and tumor location when constructing ahybrid model (ModelCR) that amalgamated radiomics and clinical features. ModelR demonstrated asubstantially enhanced discriminative capacity in the training set compared with ModelC (p < 0.005). The net reclassification improvement (NRI) corroborated the superior performance of ModelR over ModelC, thereby enhancing diagnostic accuracy and clinical applicability. Patients in the high-risk group had significantly worse recurrence-free survival (RFS, p < 0.001) and overall survival (OS, p = 0.004), and the radiomics signature is an independent risk factor for RFS. The extended model incorporating the radiomics signature outperformed the baseline model in terms of risk assessment accuracy (p < 0.001). Our investigation underscores the value of integrating radiomics analysis in conjunction with machine learning algorithms for prognostic risk stratification in GIST, presenting promising implications for informing clinical decision-making processes as well as optimizing management strategies.

An interpretable (explainable) model based on machine learning and SHAP interpretation technique for mapping wind erosion hazard.

Soil erosion by wind poses a significant threat to various regions across the globe, such as drylands in the Middle East and Iran. Wind erosion hazard maps can assist in identifying the regions of highest wind erosion risk and are a valuable tool for the mitigation of its destructive consequences. This study aims to map wind erosion hazards by developing an interpretable (explainable) model based on machine learning (ML) and Shapley additive exPlanation (SHAP) interpretation techniques. Four ML models, namely random forest (RF), support vector machine (SVM), extreme gradient boosting (XGB), and quadratic discriminant analysis (QDA) were used. Thirteen features associated with wind erosion were mapped spatially and then subjected to a multivariate adaptive regression spline (MARS) feature selection algorithm, and then, tolerance coefficient (TC) and variance inflation factor (VIF) statistical tests were used to explore multicollinearity among the variables. MARS analysis shows that eight features consisting of elevation (or DEM), soil bulk density, precipitation, aspect, slope, soil sand content, vegetation cover (or NDVI), and lithology were the most effective for wind erosion, while no collinearity existed among these variables. The ML models were used for ranking the effective features, and the research introduces the application of an interpretable ML model for the interpretation of predictive model's output. The ranking of effective features by RF-as the most typical ML model-revealed that elevation and soil bulk density were the two most important features. According to the area under the receiver operating characteristic curve (AUROC) (with a value > 90%) and precision-recall (PR) (with a value > 90%) curves, all four ML models performed with great accuracy. According to the PR curve, the SVM model performed slightly better than others, and its results revealed that 20.9%, 23%, and 16.6% of the total area in Hormozgan Province is characterized by moderate, high, and very high hazard classes to wind erosion, respectively. SHAP revealed that soil sand content and elevation are the most important variables contributing to the predictive model output. Overall, our research is one of the pioneering applications of interpretable ML models in mapping wind erosion hazards in Southern Iran. We recommend that future research should address the aspect of interpretability in order to better understand predictive model outputs.

Drug-induced cholestasis (DIC) predictions based on in vitro inhibition of major bile acid clearance mechanisms.

Drug-induced cholestasis (DIC) is recognized as a major safety concern in drug development, as it represents one of the three types of drug-induced liver injury (DILI). Cholestasis is characterized by the disruption of bile flow, leading to intrahepatic accumulation of toxic bile acids. Bile acid regulation is a multifarious process, orchestrated by several hepatic mechanisms, namely sinusoidal uptake and efflux, canalicular secretion and intracellular metabolism. In the present study, we developed a prediction model of DIC using in vitro inhibition data for 47 marketed drugs on nine transporters and five enzymes known to regulate bile acid homeostasis. The resulting model was able to distinguish between drugs with or without DILI concern (p-value = 0.039) and demonstrated a satisfactory predictive performance, with the area under the precision-recall curve (PR AUC) measured at 0.91. Furthermore, we simplified the model considering only two processes, namely reversible inhibition of OATP1B1 and time-dependent inhibition of CYP3A4, which provided an enhanced performance (PR AUC = 0.95). Our study supports literature findings suggesting a contribution not only from a single process inhibition, but a rather synergistic effect of the key bile acid clearance processes in the development of cholestasis. The use of a quantitative model in the preclinical investigations of DIC is expected to reduce attrition rate in advanced development programs and guide the discovery and development of safe medicines.

DeepmiRNATar: A deep learning-based model for miRNA targets prediction

MicroRNAs (miRNAs) play a crucial role in regulating fundamental biological processes such as the cell cycle, differentiation, and apoptosis by directly interacting with multiple genes (mRNAs). This regulatory mechanism has a profound impact on cellular function and the overall physiological condition of an organism. However, the prediction of miRNA-mRNA interactions encounters computational challenges in the field of biology due to the diverse sequences and complex data patterns. To overcome these obstacles, this research effort introduced DeepmiRNATar, a tool designed to precisely pinpoint miRNA targets, offering essential assistance in the realm of disease management. DeepmiRNATar leverages the Word2vec-based DeepLncLoc approach for encoding miRNA sequence characteristics and utilizes the DNABERT pre-trained model for in-depth semantic comprehension of target sequences. Through the integration of TextCNN, BiLSTM, and SpatialConv Attention mechanisms, the model scrutinizes structural features, temporal relationships, and overall interactions within the sequences. Following a series of experimental assessments, DeepmiRNATar attained an impressive AUC of 99.15% on the evaluation dataset, on par with the current leading prediction methodologies. Notably, the precision-recall curve, sensitivity, and F-measure values reached 99.18%, 97.43%, and 95.47%, respectively. Compared to existing miRNA target prediction models, DeepmiRNATar demonstrates a notable enhancement in overall predictive accuracy. The successful creation and experimental validation of the DeepmiRNATar model signify a significant advancement in miRNA target identification technology.

4DCT image artifact detection using deep learning.

Four-dimensional computed tomography (4DCT) is an es sential tool in radiation therapy. However, the 4D acquisition process may cause motion artifacts which can obscure anatomy and distort functional measurements from CTscans. We describe a deep learning algorithm to identify the location of artifacts within 4DCT images. Our method is flexible enough to handle different types of artifacts, including duplication, misalignment, truncation, andinterpolation. We trained and validated a U-net convolutional neural network artifact detection model on more than 23000 coronal slices extracted from 98 4DCT scans. The receiver operating characteristic (ROC) curve and precision-recall curve were used to evaluate the model's performance at identifying artifacts compared to a manually identified ground truth. The model was adjusted so that the sensitivity in identifying artifacts was equivalent to that of a human observer, as measured by computing the average ratio of artifact volume to lung volume in a givenscan. The model achieved a sensitivity, specificity, and precision of 0.78, 0.99, and 0.58, respectively. The ROC area-under-the-curve (AUC) was 0.99 and the precision-recall AUC was 0.73. Our model sensitivity is 8% higher than previously reported state-of-the-art artifact detectionmethods. The model developed in this study is versatile, designed to handle duplication, misalignment, truncation, and interpolation artifacts within a single image, unlike earlier models that were designed for a single artifact type.

Abstract Su804: Use of Machine Learning Models to Predict Neurologically Intact Survival for Advanced Age Adults Following Out-of-hospital Cardiac Arrest

Background: We recently published a multivariable logistic regression model that successfully predicted neurologic outcomes in advanced age adults (≥ 65 years old) who achieve return of spontaneous circulation following out-of-hospital cardiac arrest (OHCA). Prior to externally validating this model, we sought to compare the predictive performance of our logistic regression model against common machine learning (ML) algorithms using the same dataset. Methods: We performed a retrospective observational analysis of the Cardiac Arrest Registry to Enhance Survival database from 2013-2021. All non-traumatic OHCA occurring in adults (≥65 years) who survived to hospital admission were included. The primary outcome measure was neurologically intact survival defined as a cerebral performance category (CPC) score of 1 or 2 at hospital discharge. Our original logistic regression model was compared to Boosted Trees and Decision Tree ML algorithms and assessed using accuracy, Area Under the Receiver Operating Characteristic Curve (AUC-ROC) and Area Under the Precision-Recall Curve (AUC-PR). The model variables included age, gender, race/ethnicity, location, witnessed status, who initiated cardiopulmonary resuscitation, whether an automated external defibrillator was applied prior to emergency medical services arrival, and first monitored rhythm. Results: A total of 83,561 OHCA were included. Median age was 75 years (IQR 69-82), 58.9% were male, 53% were White, 67% experienced an OHCA at home, 52% were witnessed by a bystander, 44% received bystander CPR, and 34% were found in shockable rhythm. Neurologically intact survival (CPC 1 or 2) occurred in 23% of patients. The logistic regression model had an accuracy of 79.6%, AUC-PR 0.542, and AUC-ROC of 0.774. The Boosted Trees model had an accuracy of 79.2%, AUC-PR 0.555, and AUC-ROC of 0.774. The Decision Tree Model had an accuracy of 79.3%, AUC-PR 0.490, and AUC-ROC of 0.717. Conclusion: Boosted Trees and Decision Tree ML models performed comparably to the logistic regression model in predicting favorable versus unfavorable neurological outcomes following successful resuscitation from OHCA in advanced-age adults. Given their similar performance, logistic regression may be preferred for its simplicity and interpretability over alternative ML models. Future efforts to improve and ultimately externally validate the model are warranted.

Performance of Risk Models for Antimicrobial Resistance in Adult Patients With Sepsis

The results of prediction models that stratify patients with sepsis and risk of resistant gram-negative bacilli (GNB) infections inform treatment guidelines. However, these models do not extrapolate well across hospitals. To assess whether patient case mix and local prevalence rates of resistance contributed to the variable performance of a general risk stratification GNB sepsis model for community-onset and hospital-onset sepsis across hospitals. This was a retrospective cohort study conducted from January 2016 and October 2021. Adult patients with sepsis at 10 acute-care hospitals in rural and urban areas across Missouri and Illinois were included. Inclusion criteria were blood cultures indicating sepsis, having received 4 days of antibiotic treatment, and having organ dysfunction (vasopressor use, mechanical ventilation, increased creatinine or bilirubin levels, and thrombocytopenia). Analyses were completed in April 2024. The model included demographic characteristics, comorbidities, vital signs, laboratory values, procedures, and medications administered. Culture results were stratified for ceftriaxone-susceptible GNB (SS), ceftriaxone-resistant but cefepime-susceptible GNB (RS), and ceftriaxone- and cefepime-resistant GNB (RR). Negative cultures and other pathogens were labeled SS. Deep learning models were developed separately for community-onset (patient presented with sepsis) and hospital-onset (sepsis developed ≥48 hours after admission) sepsis. The models were tested across hospitals and patient subgroups. Models were assessed using area under the receiver operating characteristic curve (AUROC) and area under precision recall curve (AUPRC). A total of 39 893 patients with 85 238 sepsis episodes (43 207 [50.7%] community onset; 42 031 [48.3%] hospital onset) were included. Median (IQR) age was 65 (54-74) years, 21 241 patients (53.2%) were male, and 18 830 (47.2%) had a previous episode of sepsis. RS contributed to 3.9% (1667 episodes) and 5.7% (2389 episodes) of community-onset and hospital-onset sepsis episodes, respectively, and RR contributed to 1.8% (796 episodes) and 3.9% (1626 episodes), respectively. Previous infections and exposure to antibiotics were associated with the risk of resistant GNB. For example, in community-onset sepsis, 375 RR episodes (47.1%), 420 RS episodes (25.2%) and 3483 of 40 744 (8.5%) SS episodes were among patients with resistance to antimicrobial drugs (P < .001). The AUROC and AUPRC results varied across hospitals and patient subgroups for both community-onset and hospital-onset sepsis. AUPRC values correlated with the prevalence rates of resistant GNB (R = 0.79; P = .001). In this cohort study of 39 893 patients with sepsis, variable model performance was associated with prevalence rates of antimicrobial resistance rather than patient case mix. This variability suggests caution is needed when using generalized models for predicting resistant GNB etiologies in sepsis.

M-NET: Transforming Single Nucleotide Variations into Patient Feature Images for the Prediction of Prostate Cancer Metastasis and Identification of Significant Pathways.

High-performance prediction of prostate cancer metastasis based on single nucleotide variations remains a challenge. Therefore, we developed a novel biologically informed deep learning framework, named M-NET, for the prediction of prostate cancer metastasis. Within the framework, we transformed single nucleotide variations into patient feature images that are optimal for fitting convolutional neural networks. Moreover, we identified significant pathways associated with the metastatic status. The experimental results showed that M-NET significantly outperformed other comparison methods based on single nucleotide variations, achieving improvements in accuracy, precision, recall, F1-score, area under the receiver operating characteristics curve, and area under the precision-recall curve by 6.3%, 8.4%, 5.1%, 0.070, 0.041, and 0.026, respectively. Furthermore, M-NET identified some important pathways associated with the metastatic status, such as signaling by the hedgehog pathway. In summary, compared with other comparative methods, M-NET exhibited a better performance in the prediction of prostate cancer metastasis.

One Third of Alcohol Use Disorder Diagnoses are Missed by ICD Coding.

Alcohol use carries significant morbidity and mortality, yet accurate identification of alcohol use disorder (AUD) remains a multi-layered problem for both researchers and clinicians. To fine-tune a language model to AUD in the clinical narrative and to detect AUDs not accounted for by ICD-9 coding in the MIMIC-III database. We applied clinicalBERT to unique patient discharge summaries. For classification, patients were divided into nonoverlapping groups stratified by the presence/absence of AUD ICD diagnosis for model training (80%), validation (10%), and testing (10%). For detection, the model was trained (80%) and validated (20%) on 1:1 positive/negative patients, then applied to remaining negative patient population. Physicians adjudicated 600 samples from the full model confidence spectrum to confirm AUD by Diagnostic and Statistical Manual of Mental Disorders-V criteria. The model exhibited the following characteristics (mean, standard deviation): precision (0.9, 0.02), recall (0.65, 0.03), F-1 (0.75, 0.02), area under the receiver operating curve (0.97, 0.01), and area under the precision-recall curve (0.86, 0.01). Adjudication produced an estimated 4% under-documentation rate for the total study population. As model confidence increased, AUD under-documentation rate rose to 30% of the number of patients identified as positive by ICD-9 coding. Our model improves the identification of patients meeting AUD criteria, outperforming ICD codes in detecting cases of AUD. Detection discrepancy between ICD and free-text highlights clinician under documentation, not under recognition. Adjudication revealed model over-sensitivity to language around substance use, withdrawal, and chronic liver disease; future study requires application to a broader set of patient age and acuity. This model has the potential to improve rapid identification of patients with AUD and enhance treatment allocation.

“ChatGPT 4.0 Ghosted Us While Conducting Literature Search:” Modeling the Chatbot’s Generated Non-Existent References Using Regression Analysis

The integration of AI technologies like ChatGPT has transformed academic research, yet substantial gaps exist in understanding the implications of AI-generated non-existent references in literature searches. While prior studies have predominantly focused on medical and geography fields using descriptive statistics, a systematic investigation into ChatGPT 4.0’s effectiveness in generating accurate references within the realm of science and technology education remains unexplored, highlighting a significant dearth of research in this critical area. This study, therefore, investigates the reliability of AI-generated references in academic writing utilizing ChatGPT 4.0. Employing a non-experimental correlational design, the research examines the impact of prompt specificity on citation accuracy across various types of prompts, including general, specific, methodological, review, and interdisciplinary prompts. The findings indicate that specific, review, and interdisciplinary prompts correlate positively with accurate references, while general prompts frequently result in non-existent references. Visualizations, including a confusion matrix and precision-recall curve, illustrate the model’s performance. Ultimately, the study underscores the necessity of well-structured prompts to enhance reference quality and cautions against AI-induced hallucinations that produce non-existent references, which can significantly undermine research credibility.

Development and Validation of Deep Learning-Based Infectivity Prediction in Pulmonary Tuberculosis Through Chest Radiography: Retrospective Study.

Pulmonary tuberculosis (PTB) poses a global health challenge owing to the time-intensive nature of traditional diagnostic tests such as smear and culture tests, which can require hours to weeks to yield results. This study aimed to use artificial intelligence (AI)-based chest radiography (CXR) to evaluate the infectivity of patients with PTB more quickly and accurately compared with traditional methods such as smear and culture tests. We used DenseNet121 and visualization techniques such as gradient-weighted class activation mapping and local interpretable model-agnostic explanations to demonstrate the decision-making process of the model. We analyzed 36,142 CXR images of 4492 patients with PTB obtained from Severance Hospital, focusing specifically on the lung region through segmentation and cropping with TransUNet. We used data from 2004 to 2020 to train the model, data from 2021 for testing, and data from 2022 to 2023 for internal validation. In addition, we used 1978 CXR images of 299 patients with PTB obtained from Yongin Severance Hospital for external validation. In the internal validation, the model achieved an accuracy of 73.27%, an area under the receiver operating characteristic curve of 0.79, and an area under the precision-recall curve of 0.77. In the external validation, it exhibited an accuracy of 70.29%, an area under the receiver operating characteristic curve of 0.77, and an area under the precision-recall curve of 0.8. In addition, gradient-weighted class activation mapping and local interpretable model-agnostic explanations provided insights into the decision-making process of the AI model. This proposed AI tool offers a rapid and accurate alternative for evaluating PTB infectivity through CXR, with significant implications for enhancing screening efficiency by evaluating infectivity before sputum test results in clinical settings, compared with traditional smear and culture tests.

Inconsistency among evaluation metrics in link prediction

Abstract Link prediction is a paradigmatic and challenging problem in network science, which aims to predict missing links, future links and temporal links based on known topology. Along with the increasing number of link prediction algorithms, a critical yet previously ignored risk is that the evaluation metrics for algorithm performance are usually chosen at will. This paper implements extensive experiments on hundreds of real networks and 26 well-known algorithms, revealing significant inconsistency among evaluation metrics, namely different metrics probably produce remarkably different rankings of algorithms. Therefore, we conclude that any single metric cannot comprehensively or credibly evaluate algorithm performance. In terms of information content, we suggest the usage of at least two metrics: one is the area under the receiver operating characteristic curve (AUC), and the other is one of the following three candidates, say the area under the precision-recall curve (AUPR), the area under the precision curve (AUC-Precision), and the normalized discounted cumulative gain (NDCG). When the data is imbalanced, say the number of negative samples significantly outweighs the number of positive samples, the area under the generalized ROC curve (AUC-gROC) should also be used. In addition, as we have proved the essential equivalence of threshold-dependent metrics, if in a link prediction task, some specific thresholds are meaningful, we can consider any one threshold-dependent metric with those thresholds. This work completes a missing part in the landscape of link prediction, and provides a starting point toward a well-accepted criterion or standard to select proper evaluation metrics for link prediction.

Predicting temporomandibular disorders in adults using interpretable machine learning methods: a model development and validation study

IntroductionTemporomandibular disorders (TMD) have a high prevalence and complex etiology. The purpose of this study was to apply a machine learning (ML) approach to identify risk factors for the occurrence of TMD in adults and to develop and validate an interpretable predictive model for the risk of TMD in adults.MethodsA total of 949 adults who underwent oral examinations were enrolled in our study. 5 different ML algorithms were used for model development and comparison, and feature selection was performed by feature importance ranking and feature decreasing methods. Several evaluation indexes, including the area under the receiver-operating-characteristic curve (AUC), were used to compare the predictive performance. The precision-recall curve (PR), calibration curve, and decision curve analysis (DCA) further assessed the accuracy and clinical utility of the model.ResultsThe performance of the random forest (RF) model was the best among the 5 ML models. An interpretable RF model was developed with 7 features (gender, malocclusion, unilateral chewing, chewing hard substances, grinding teeth, clenching teeth, and anxiety). The AUCs of the final model on the training set, internal validation set, and external test set were 0.892, 0.854, and 0.857, respectively. Calibration and DCA curves showed high accuracy and clinical applicability of the model.DiscussionAn efficient and interpretable TMD risk prediction model for adults was successfully developed using the ML method. The model not only has good predictive performance, but also enhances the clinical application value of the model through the SHAP method. This model can provide clinicians with a practical and efficient TMD risk assessment tool that can help them better predict and assess TMD risk in adults, supporting more efficient disease management and targeted medical interventions.

A Deep Learning Model to Predict Breast Implant Texture Types Using Ultrasonography Images: Feasibility Development Study.

Breast implants, including textured variants, have been widely used in aesthetic and reconstructive mammoplasty. However, the textured type, which is one of the shell texture types of breast implants, has been identified as a possible etiologic factor for lymphoma, specifically breast implant-associated anaplastic large cell lymphoma (BIA-ALCL). Identifying the shell texture type of the implant is critical to diagnosing BIA-ALCL. However, distinguishing the shell texture type can be difficult due to the loss of human memory and medical history. An alternative approach is to use ultrasonography, but this method also has limitations in quantitative assessment. This study aims to determine the feasibility of using a deep learning model to classify the shell texture type of breast implants and make robust predictions from ultrasonography images from heterogeneous sources. A total of 19,502 breast implant images were retrospectively collected from heterogeneous sources, including images captured from both Canon and GE devices, images of ruptured implants, and images without implants, as well as publicly available images. The Canon images were trained using ResNet-50. The model's performance on the Canon dataset was evaluated using stratified 5-fold cross-validation. Additionally, external validation was conducted using the GE and publicly available datasets. The area under the receiver operating characteristic curve (AUROC) and the area under the precision-recall curve (PRAUC) were calculated based on the contribution of the pixels with Gradient-weighted Class Activation Mapping (Grad-CAM). To identify the significant pixels for classification, we masked the pixels that contributed less than 10%, up to a maximum of 100%. To assess the model's robustness to uncertainty, Shannon entropy was calculated for 4 image groups: Canon, GE, ruptured implants, and without implants. The deep learning model achieved an average AUROC of 0.98 and a PRAUC of 0.88 in the Canon dataset. The model achieved an AUROC of 0.985 and a PRAUC of 0.748 for images captured with GE devices. Additionally, the model predicted an AUROC of 0.909 and a PRAUC of 0.958 for the publicly available dataset. This model maintained the PRAUC values for quantitative validation when masking up to 90% of the least-contributing pixels and the remnant pixels in breast shell layers. Furthermore, the prediction uncertainty increased in the following order: Canon (0.066), GE (0072), ruptured implants (0.371), and no implants (0.777). We have demonstrated the feasibility of using deep learning to predict the shell texture type of breast implants. This approach quantifies the shell texture types of breast implants, supporting the first step in the diagnosis of BIA-ALCL.

Completely non-invasive prediction of IDH mutation status based on preoperative native CT images.

The isocitrate dehydrogenase (IDH) mutation status is one of the most important markers according to the 2021 WHO classification of CNS tumors. Preoperatively, this information is usually obtained based on invasive biopsies, contrast-enhanced MR images or PET images generated using radioactive tracers. However, the completely non-invasive determination of IDH mutation status using routinely acquired preoperative native CT images has hardly been investigated to date. In our study, we show that radiomics-based machine learning allows to determine IDH mutation status based on preoperative native CT images both with very high accuracy and completely non-invasively. Based on independent test data, we are able to correctly identify 91.1% of cases with an IDH mutation. Our final model, containing only six features, exhibits a high area under the curve of 0.847 and an excellent area under the precision-recall curve of 0.945. In the future, such models may be used for a completely non-invasive prediction of important genetic markers, potentially allowing treating physicians to reduce the number of biopsies and speed up further treatment planning.

MIPPIS: protein–protein interaction site prediction network with multi-information fusion

BackgroundThe prediction of protein–protein interaction sites plays a crucial role in biochemical processes. Investigating the interaction between viruses and receptor proteins through biological techniques aids in understanding disease mechanisms and guides the development of corresponding drugs. While various methods have been proposed in the past, they often suffer from drawbacks such as long processing times, high costs, and low accuracy.ResultsAddressing these challenges, we propose a novel protein–protein interaction site prediction network based on multi-information fusion. In our approach, the initial amino acid features are depicted by the position-specific scoring matrix, hidden Markov model, dictionary of protein secondary structure, and one-hot encoding. Simultaneously, we adopt a multi-channel approach to extract deep-level amino acids features from different perspectives. The graph convolutional network channel effectively extracts spatial structural information. The bidirectional long short-term memory channel treats the amino acid sequence as natural language, capturing the protein’s primary structure information. The ProtT5 protein large language model channel outputs a more comprehensive amino acid embedding representation, providing a robust complement to the two aforementioned channels. Finally, the obtained amino acid features are fed into the prediction layer for the final prediction.ConclusionCompared with six protein structure-based methods and six protein sequence-based methods, our model achieves optimal performance across evaluation metrics, including accuracy, precision, F1, Matthews correlation coefficient, and area under the precision recall curve, which demonstrates the superiority of our model.

Predicting deterioration in dengue using a low cost wearable for continuous clinical monitoring

Close vital signs monitoring is crucial for the clinical management of patients with dengue. We investigated performance of a non-invasive wearable utilising photoplethysmography (PPG), to provide real-time risk prediction in hospitalised individuals. We performed a prospective observational clinical study in Vietnam between January 2020 and October 2022: 153 patients were included in analyses, providing 1353 h of PPG data. Using a multi-modal transformer approach, 10-min PPG waveform segments and basic clinical data (age, sex, clinical features on admission) were used as features to continuously forecast clinical state 2 h ahead. Prediction of low-risk states (17,939/80,843; 22.1%), defined by NEWS2 and mSOFA < 6, was associated with an area under the precision-recall curve of 0.67 and an area under the receiver operator curve of 0.83. Implementation of such interventions could provide cost-effective triage and clinical care in dengue, offering opportunities for safe ambulatory patient management.

High-Precision Defect Detection in Solar Cells Using YOLOv10 Deep Learning Model

This study presents an advanced defect detection approach for solar cells using the YOLOv10 deep learning model. Leveraging a comprehensive dataset of 10,500 solar cell images annotated with 12 distinct defect types, our model integrates Compact Inverted Blocks (CIBs) and Partial Self-Attention (PSA) modules to enhance feature extraction and classification accuracy. Training on the Viking cluster with state-of-the-art GPUs, our model achieved remarkable results, including a mean Average Precision (mAP@0.5) of 98.5%. Detailed analysis of the model’s performance revealed exceptional precision and recall rates for most defect classes, notably achieving 100% accuracy in detecting black core, corner, fragment, scratch, and short circuit defects. Even for challenging defect types such as a thick line and star crack, the model maintained high performance, with accuracies of 94% and 96%, respectively. The Recall–Confidence and Precision–Recall curves further demonstrate the model’s robustness and reliability across varying confidence thresholds. This research not only advances the state of automated defect detection in photovoltaic manufacturing but also underscores the potential of YOLOv10 for real-time applications. Our findings suggest significant implications for improving the quality control process in solar cell production. Although the model demonstrates high accuracy across most defect types, certain subtle defects, such as thick lines and star cracks, remain challenging, indicating potential areas for further optimization in future work.

Development and Validation of Machine Learning Models for Identifying Prediabetes and Diabetes in Normoglycemia.

Prediabetes and diabetes are both abnormal states of glucose metabolism (AGM) that can lead to severe complications. Early detection of AGM is crucial for timely intervention and treatment. However, fasting blood glucose (FBG) as a mass population screening method may fail to identify some individuals who are actually AGM but with normoglycemia. This study aimed to develop and validate machine learning (ML) models to identify AGM among individuals with normoglycemia using routine health check-up indicators. According to the American Diabetes Association (ADA) criteria, participants with normoglycemia (FBG≤5.6mmol/L) were collected from 2019 to 2023, and then divided into AGM and Normalgroups using glycosylated haemoglobin (HbA1c) 5.7% as the threshold. Data from 2019 to 2022 were divided into training and internal validation sets at a 7:3 ratio, while data from 2023 were used as the external validation set. Seven ML algorithms-including logistic regression (LR), random forest (RF), support vector machine (SVM), extreme gradient boosting machine, multilayer perceptron (MLP), light gradient boosting machine (LightGBM), and categorical boosting (CatBoost)-were used to build models for identifying AGM in normoglycemia population. Model performance was evaluated using the area under the receiver operating characteristic curve (auROC) and the precision-recall curve (auPR). The feature contributions to the optimal model was visualised using the SHapley Additive exPlanations (SHAP). Finally, an intuitive and user-friendly interactive interface was developed. A total of 59,259 participants were finally enroled in this study, and then divided into the training set of 32,810, the internal validation set of 14,060, and the external validation set of 12,389. The Catboost model outperformed the others with auROC of 0.806 and 0.794 for the internal and external validation set, respectively. Age was the most important feature influencing the performance of the CatBoost model, followed by fasting blood glucose, red blood cells, haemoglobin, body mass index, and triglyceride-glucose. A well-performed ML model to identify AGM in the normoglycemia population was built, offering significant potential for early intervention and treatment of AGM that would otherwise have been missed.

A large scale multi institutional study for radiomics driven machine learning for meningioma grading

This study aims to develop and evaluate radiomics-based machine learning (ML) models for predicting meningioma grades using multiparametric magnetic resonance imaging (MRI). The study utilized the BraTS-MEN dataset’s training split, including 698 patients (524 with grade 1 and 174 with grade 2–3 meningiomas). We extracted 4872 radiomic features from T1, T1 with contrast, T2, and FLAIR MRI sequences using PyRadiomics. LASSO regression reduced features to 176. The data was split into training (60%), validation (20%), and test (20%) sets. Five ML algorithms (TabPFN, XGBoost, LightGBM, CatBoost, and Random Forest) were employed to build models differentiating low-grade (grade 1) from high-grade (grade 2–3) meningiomas. Hyperparameter tuning was performed using Optuna, optimizing model-specific parameters and feature selection. The CatBoost model demonstrated the best performance, achieving an area under the receiver operating characteristic curve (AUROC) of 0.838 [95% confidence interval (CI): 0.689–0.935], precision of 0.492 (95% CI: 0.371–0.623), recall of 0.838 (95% CI: 0.689–0.935), F1 score of 0.620 (95% CI: 0.495–0.722), accuracy of 0.729 (95% CI: 0.650–0.800), an area under the precision-recall curve (AUPRC) of 0.620 (95% CI: 0.433–0.753), and Brier score of 0.156 (95% CI: 0.122-0.200). Other models showed comparable performance, with mean AUROCs ranging from 0.752 to 0.784. The radiomics-based ML approach presented in this study showcases the potential for non-invasive and pre-operative grading of meningiomas using multiparametric MRI. Further validation on larger and independent datasets is necessary to establish the robustness and generalizability of these findings.