SEARCH · PubMed Health

Results for “Predictive Learning Models”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 37 records · Page 2Linked to original sources

PubMed2026

Systematic review of machine learning approaches for predicting sickle cell crisis and mortality risk at the climate-health nexus.

BACKGROUND: Sickle cell anemia (SCA) is a severe genetic blood disorder characterized by recurrent vaso-occlusive crises and increased mortality, with the greatest burden occurring in low- and middle-income countries. Climatic and environmental conditions, including temperature variability, humidity, rainfall, air pollution, and seasonal changes, have been associated with disease exacerbation. However, the extent to which these factors have been incorporated into predictive models remains unclear. This study systematically reviews the application of machine learning (ML) models for predicting SCA crises and mortality in relation to climate and environmental factors. METHODOLOGY: The PRISMA guidelines were used, and 34 peer-reviewed studies published between 2005 and 2026 were analyzed to identify the climate variables, ML approaches employed, and predictive performance. The reviewed studies applied a range of ML techniques, including artificial neural networks, random forests, support vector machines, decision trees, logistic regression, and deep learning models. Temperature, humidity, rainfall, wind speed, air quality indicators, and seasonal patterns were the most frequently examined environmental variables. RESULTS: The findings indicate that most existing models rely predominantly on clinical and demographic data, with limited integration of climate information and inadequate representation of high-burden regions, especially Sub-Saharan Africa. Studies incorporating environmental variables reported improved predictive performance and highlighted the potential of climate-informed early warning systems for SCA management. CONCLUSION: The review recommends development of interdisciplinary, climate-aware ML frameworks, expansion of longitudinal environmental datasets, and increased research in underrepresented regions to support climate-resilient and patient-centered SCA care.

Humans↗

PubMed2025

Development and evaluation of a machine learning model for osteoporosis risk prediction in Korean women.

BACKGROUND: The aim of this study was to develop a machine learning (ML) model for classifying osteoporosis in Korean women based on a large-scale population cohort study. This study also aimed to assess ML model performance compared with traditional osteoporosis screening tools. Furthermore, this study aimed to examine the factors influencing the risk of osteoporosis through variable importance. METHODS: Data was collected from 4199 women aged 40-69 years in the baseline survey of the Ansan and Ansung cohort of the Korean Genome and Epidemiology Study. Osteoporosis was set as the dependent variable to develop ML classification models. Independent variables included 122 factors related to osteoporosis risk, such as socio-demographic characteristics, anthropometric parameters, lifestyle factors, reproductive factors, nutrient intakes, diet quality indices, medical history, medication history, family history, biochemical parameters, and genetic factors. The six classification models were developed using ML techniques, including decision tree, random forest, multilayer perceptron, support vector machine, light gradient boosting machine, and extreme gradient boosting (XGBoost). The six ML classification models were compared with two traditional osteoporosis screening tools, including the osteoporosis risk assessment instrument (ORAI) and the osteoporosis self-assessment tool (OST). The ML model performances were evaluated and compared using the confusion matrix and area under the curve (AUC) metrics. Variable importance was assessed using the XGBoost technique to investigate osteoporosis risk factors. RESULTS: The XGBoost model showed the highest performance out of the six ML classification models, with an accuracy of 0.705, precision of 0.664, recall of 0.830, and F1 score of 0.738. Moreover, the XGBoost model showed a higher performance on AUC than ORAI and OST. Variable importance scores were identified for 69 out of the 122 variables associated with osteoporosis risk factors. Age at menopause ranked first in variable importance. Variables of arthritis, physical activities, hypertension, education level, income level; alcohol intake, potassium intake, homeostatic model assessment for insulin resistance; energy intake, vitamin C intake, gout; and dietary inflammatory index ranked in the top 20 out of the 69 variables, using the XGBoost technique. CONCLUSIONS: This study found that an XGBoost model can be utilized to classify osteoporosis in Korean women. Age at menopause is a significant factor in osteoporosis risk, followed by arthritis, physical activities, hypertension, and education level.

Humans↗

PubMed2026

An encyclopedia of human enhancer-gene regulatory interactions.

Identifying transcriptional enhancers and their target genes is essential for understanding gene regulation and the effect of human genetic variation on disease1-6. Here we create and evaluate a resource of more than 92 million enhancer-gene regulatory interactions across 1,458 biosamples covering 369 cell types and tissues, by integrating predictive models, chromatin states, three-dimensional contacts and large-scale genetic perturbations generated by the ENCODE Consortium7. We first create a systematic benchmarking pipeline to compare predictive models, assembling a dataset of 10,356 element-gene pairs measured in CRISPR perturbation experiments, more than 30,000 fine-mapped expression quantitative trait loci and 569 fine-mapped genome-wide association study (GWAS) variants linked to a probable causal gene. Using this framework, we develop ENCODE-rE2G, a predictive model achieving state-of-the-art performance across several prediction tasks, demonstrating that iterative perturbations and supervised machine learning can build increasingly accurate predictive models of enhancer regulation. Using ENCODE-rE2G, we build an encyclopedia of enhancer-gene regulatory interactions in the human genome, revealing global properties of enhancer networks, identifying differences in regulatory complexity across genes and improving analyses linking noncoding variants to target genes and cell types for common complex diseases. By interpreting the model, we find that beyond enhancer activity and three-dimensional enhancer-promoter contacts, additional features that guide enhancer-promoter communication include promoter class and enhancer-enhancer synergy. These genome-wide maps of enhancer-gene regulatory interactions, benchmarking software, predictive models and insights about enhancer function provide a valuable resource for future studies of gene regulation and human genetics.

Humans↗

PubMed2026

Clinical Variable-Based Machine Learning for Predicting Early mCRPC Using Exclusively Clinical Variables: Development and Multicenter External Validation.

BACKGROUND AND OBJECTIVE: Metastatic hormone-sensitive prostate cancer (mHSPC) exhibits heterogeneous progression patterns, with early progression to metastatic castration-resistant prostate cancer (mCRPC) within 12 months indicating aggressive tumor biology and poor prognosis. Current risk stratification tools (CHAARTED, LATITUDE) offer limited individualized prediction. Machine learning approaches are increasingly applied to predict prostate cancer progression, but most models show modest performance (AUC 0.68-0.72), limited external validation, or require genomic variables unavailable in routine practice. This study aimed to develop and externally validate a novel RINH algorithm for predicting early mCRPC progression (≤ 12 months) using exclusively clinical variables, positioning it as a superior alternative to conventional ML classifiers. METHODS: This multicenter study enrolled 412 patients with de novo mHSPC from seven Spanish academic centers using mixed retrospective-prospective data collection. Twenty clinical variables were recorded, including demographics, PSA, ISUP grade, metastatic localization, CHAARTED/LATITUDE classifications, and treatment modalities. Following RINH-based outlier exclusion (55 patients), 357 patients (29 with early progression, 8.1%) were used to train six ML algorithms: RINH, Logistic Regression, Linear Discriminant, Support Vector Machine, Random Forest, and Subspace Discriminant. A two-tiered validation strategy integrated stratified fivefold cross-validation across all centers and formal external validation using center 1 (n = 121, 19 events) for training and centers 2-7 (n = 207, 10 events) for independent testing. Performance metrics included AUC, sensitivity, specificity, accuracy, and F1-score. KEY FINDINGS AND LIMITATIONS: Artificial intelligence and machine learning (ML) are transforming oncology, promising personalized risk stratification beyond traditional clinical criteria. In metastatic hormone-sensitive prostate cancer (mHSPC), early progression to castration resistance (mCRPC) within 12 months signals aggressive biology and poor prognosis, yet current tools (CHAARTED, LATITUDE) offer limited individualized prediction. Multiple ML models have been proposed with variable success: most achieve modest performance (AUC 0.68-0.72), lack robust external validation, or rely on genomic variables inaccessible in routine practice. We propose a novel approach using the Rivality Index Neighborhood (RINH) algorithm, demonstrating superior predictive capacity in an initial multicenter validation with exclusively clinical variables. This study provides rigorous multicenter external validation, advancing toward implementable precision oncology tools. CONCLUSIONS AND CLINICAL IMPLICATIONS: The RINH algorithm achieves superior predictive performance for early mCRPC progression using exclusively clinical variables, representing a significant advance toward implementable risk stratification. However, low reliability scores in external validation underscore that excellent performance metrics alone do not guarantee stability. Before clinical deployment, validation in substantially larger cohorts with higher progression events is essential. If validated, this model could enable personalized, risk-adapted therapeutic strategies, refining patient selection for treatment intensification or de-escalation.

Humans↗

PubMed2026

Machine learning for population-level risk prediction of future cholangiocarcinoma.

BACKGROUND: The poor prognosis of cholangiocarcinoma (CCA) is largely driven by rapid, asymptomatic disease progression, which usually results in a late diagnosis in the absence of established screening strategies. An early, cost-effective, and universally applicable risk assessment strategy would therefore be valuable. METHODS: We developed machine learning (ML) models on prospective, multimodal data from 487,495 UK Biobank (UKB) participants, of whom 649 developed CCA during follow-up. Data from England (80%) were utilised for ML development via five-fold cross-validation, and then all models were tested on withheld data from Scotland, Wales, and Newcastle (20%). Iterative ablation studies reduced inputs from >150 features across demographic data, lifestyle, health records, blood parameters, genomics, and metabolomics to models built on five and ten routinely available clinical parameters. These were externally validated in the Penn Medicine Biobank (PMBB; n = 2638; 28 CCA), All of Us Research Program (AOU; n = 330,433; 362 CCA), Japan Medical Data Centre Claims Database (JMDC; n = 8,425,522; 723 CCA) and TriNetX (n = 728,886; 1592 CCA). FINDINGS: We show that ML models integrating biliary-disease associated health records and Gamma glutamyltransferase can stratify risk of future CCA. Evaluation on the UKB test set as well as three independent cohorts revealed robust performance and generalisability across ethnicities. We achieved AUROCs of 0.71 [95% CI: 0.703-0.711], 0.77 [95% CI: 0.764-0.778 ], 0.796 [95% CI: 0.795-0.798] and 0.8 [95% CI: 0.794-0.805] for UKB, PMBB, AOU, and JMDC respectively, with respective AUPRCs of 0.014 [95% CI: 0.009-0.018], 0.042 [95% CI: 0.037-0.048], 0.038 [95% CI: 0.033-0.042] and 0.001 [95% CI: 0.001-0.001]. In AOU, application of the Youden J-optimised threshold yielded a number needed to screen of 79. Separate models for intra- and extrahepatic CCA did not improve performance. In line with the pathophysiology, performance declined for longer intervals between assessment and event. A group-level analysis in the TriNetX cohort revealed hazard ratios of up to 82.5 [95% CI: 26.4-257.96]. We provide extensive interpretability results and release all source codes used to develop the presented models. INTERPRETATION: We provide a comprehensive framework for early CCA risk stratification in the general population, identifying key predictors, and demonstrating the potential of data-driven models in personalised screening for hepatobiliary cancer. FUNDING: German Cancer Aid (grant #70115730), Junior Principal Investigator Fellowship programme of RWTH Aachen Excellence strategy.

Humans↗

PubMed1992

Predictive and diagnostic learning within causal models: asymmetries in cue competition.

Several researchers have recently claimed that higher order types of learning, such as categorization and causal induction, can be reduced to lower order associative learning. These claims are based in part on reports of cue competition in higher order learning, apparently analogous to blocking in classical conditioning. Three experiments are reported in which subjects had to learn to respond on the basis of cues that were defined either as possible causes of a common effect (predictive learning) or as possible effects of a common cause (diagnostic learning). The results indicate that diagnostic and predictive reasoning, far from being identical as predicted by associationistic models, are not even symmetrical. Although cue competition occurs among multiple possible causes during predictive learning, multiple possible effects need not compete during diagnostic learning. The results favor a causal-model theory.

Adult↗

PubMed2026

Machine learning to differentiate colonization from infection in multidrug-resistant Gram-negative bacteria: implications for further research.

PURPOSE OF REVIEW: Machine learning has emerged as a promising tool to support antimicrobial decision-making in infectious diseases. In colonized patients, distinguishing multidrug-resistant Gram-negative bacteria (MDR-GNB) colonization from true infection remains a major clinical challenge, as both delayed appropriate therapy in severe infections and unnecessary broad-spectrum antimicrobial use may adversely affect patient outcomes and antimicrobial stewardship. This review discusses the current evidence on machine learning models for predicting or detecting MDR-GNB infection in colonized patients, highlights key methodological limitations of the available literature, and outlines future research priorities. RECENT FINDINGS: Current evidence specifically evaluating machine learning models beyond logistic regression in MDR-GNB-colonized patients remains limited. Overall, while machine learning may achieve encouraging discriminatory performance, important methodological limitations persist. Most notably, predictive models are frequently developed in heterogeneous populations that do not reflect the clinically relevant populations of colonized patients in which treatment decisions are made. Furthermore, improvements in predictive performance remain modest, possibly reflecting limited sample sizes and data granularity rather than insufficient algorithmic complexity. In our opinion, future advances could require multicenter datasets enriched with longitudinal clinical, microbiological, and genomic information, together with automated feature extraction from electronic health records. SUMMARY: The main challenge for machine learning in predicting MDR-GNB infection in colonized patients may lie not in developing increasingly sophisticated algorithms, but in generating clinically representative datasets and adopting rigorous methodological standards for model development, validation, calibration, and implementation. Future research should prioritize clinically meaningful target populations and demonstrate improvements in patient outcomes and antimicrobial stewardship beyond conventional measures of predictive performance.

antimicrobial resistance↗

PubMed2026

Diagnostic performance of machine learning models versus established risk stratification for intracranial aneurysm rupture: a systematic review and bivariate meta-analysis.

BACKGROUND: Machine learning (ML) models have been proposed to improve the discrimination of intracranial aneurysm rupture status beyond established clinical risk stratification tools. However, reported performance is heterogeneous and the relative contribution of model architecture and feature dominance remains unclear. METHODS: We performed a Preferred Reporting Items for Systematic Reviews and Meta-Analyses-diagnostic test accuracy systematic review and diagnostic meta-analysis of studies evaluating ML models for intracranial aneurysm rupture discrimination. PubMed, Embase and CENTRAL were searched to February 2026. Sensitivity and specificity were pooled using a bivariate random-effects model, with summary receiver operating characteristic curves generated across training, internal testing and external validation datasets. Models were compared with regression-based approaches and Population, Hypertension, Age, Size of aneurysm, Earlier subarachnoid haemorrhage, Site of aneurysm (PHASES) scores. Subgroup and meta-regression analyses explored associations between algorithm family and feature domain. RESULTS: Sixty-two retrospective cohorts (29 709 patients 209 models) met the inclusion criteria. In training datasets, pooled sensitivity and specificity for ML were 0.81 (95% CI 0.75 to 0.85) and 0.83 (0.80-0.86), with an area under the curve (AUC) of 0.878, exceeding PHASES (AUC 0.667). In testing datasets, ML retained higher discrimination (AUC 0.837) than regression models (0.806) and PHASES (0.646). In external validation, sensitivity was preserved (0.82), but specificity declined (0.66). Deep learning demonstrated the highest AUCs (training and testing). Incorporation of haemodynamic or radiomic features improved pooled discrimination relative to morphology alone. Evidence of small-study effects and mostly unclear Prediction Model Risk Of Bias Assessment Tool ratings were observed. CONCLUSIONS: ML approaches demonstrate higher pooled discrimination for aneurysm rupture status than conventional risk scores in retrospective datasets, but reduced external validation specificity and heterogeneity limit confidence for clinical translation. Prospective, externally validated, calibrated models are required before integration into routine cerebrovascular risk stratification.

Humans↗

PubMed2026

ToxiVerse: chemical bioprofiling, toxicity data sharing and customizable predictive modeling.

MOTIVATION: Chemical toxicity assessment is critical for drug development and environmental safety. Computational models have emerged as a promising alternative to animal testing and now play a significant role in efficiently evaluating new chemicals. To address the urgent need for user-friendly machine learning tools in computational toxicology, we developed ToxiVerse, a public web-based platform. RESULTS: ToxiVerse provides automatic chemical bioprofiling, curated toxicity datasets, and a predictive modeling interface designed for researchers who lack programming expertise. The platform comprises three integrated modules: (i) Bioprofiler, which provides chemical descriptors by combining chemical-bioactivity data from PubChem assays with a machine learning-based data gap-filling procedure; (ii) Database, which hosts ∼50 000 curated chemicals covering diverse toxicity endpoints; and (iii) Cheminformatics, which enables dataset upload, chemical curation, and automatic generation of quantitative structure-activity relationship models for toxicity prediction. AVAILABILITY: The tool is accessible at www.toxiverse.com, and source code is available at https://github.com/zhu-research-group/toxiverse.

Quantitative Structure-Activity Relationship↗

PubMed2026

Machine learning-based prediction of unplanned readmission and construction of an online calculator for elderly patients with mild ischemic stroke.

OBJECTIVE: To screen for independent risk factors for unplanned readmission in elderly patients with mild ischemic stroke, and to construct and validate an online risk prediction calculator based on an interpretable machine learning model, thereby providing a promising practical tool for accurate clinical assessment of 30‑day all‑cause unplanned readmission risk in this population. METHODS: A prospective cohort study was conducted, including 1050 patients aged ≥ 60 years with mild ischemic stroke admitted between August 2023 and September 2024. Participants were randomly divided into a training set (840 cases) and a test set (210 cases) at a ratio of 8:2. Risk factors were screened by univariate analysis and multivariable Logistic regression. Four machine learning models, namely LightGBM, XGBoost, Random Forest, and K‑Nearest Neighbors (KNN), were developed and their performance was evaluated using AUC, accuracy, sensitivity, and specificity as metrics. The SHAP framework was used for interpretability analysis, and an online calculator was subsequently developed based on the optimal model. RESULTS: Univariate analysis showed significant differences (P < 0.05) in 13 factors including age, smoking, AIP, TyG index, HALP score, etc. Multivariable Logistic regression identified age (OR = 9.752), smoking (OR = 5.171), AIP (OR = 6.691), TyG index (OR = 4.393), HALP score (OR = 2.831), and ≥ 2 comorbidities (OR = 3.664) as independent risk factors. All four machine learning models demonstrated good predictive performance. Based on a comprehensive evaluation of multiple metrics and computational efficiency, the LightGBM model exhibited the best predictive performance (AUC = 0.884, accuracy = 0.829, sensitivity = 0.812, specificity = 0.875). SHAP analysis showed that age, AIP, TyG index, smoking, and HALP score were key predictors. An online calculator developed based on this model enables individualized risk predictions. CONCLUSION: Key risk factors associated with 30‑day unplanned readmission in elderly patients with mild ischemic stroke were identified. The LightGBM model demonstrated high predictive accuracy, and together with the interpretability analysis and online calculator, offers a practical tool to support clinical risk assessment. However, this tool requires future external validation.

Humans↗

PubMed1991

Disease versus social-learning models of alcoholism in the prediction of alcohol problem recognition, help-seeking, and stigma.

Subjects (N = 254) were presented with either a disease or social-learning view of alcoholism and their subsequent attitudes were examined. Presentation of a disease view strengthened endorsement of that model, but attitudes toward alcoholics, treatment effectiveness, problem recognition, and help-seeking were not significantly different between the experimental conditions. Differences in attitudes between high and low disease and social-learning belief subjects were examined in 2 x 2 ANOVAs. Separate multiple regression analyses with alcohol problem recognition, help-seeking, and stigma as composite dependent variables and etiologic beliefs, drinking measures, family history of alcoholism, and responsibility for drinking problems as independent variables were performed. Results show that neither disease nor social-learning beliefs are particularly effective in predicting the dependent variables.

Adolescent↗

PubMed2025

Predicting the First Onset of Suicidal Thoughts and Behaviors in Adolescents Using Multimodal Risk Factors: A 4-Year Longitudinal Study.

OBJECTIVE: Suicide is one of the leading causes of death among youth worldwide, yet existing studies that aimed to predict the first onset of suicidal thoughts and behaviors (STB) included a limited number of data modalities and/or focused on adult populations. This study aimed to prospectively predict first-onset STB across 4-year follow-ups in adolescents using an existing STB history classification model that was previously applied to baseline data and a new machine learning model with 195 biopsychosocial features. METHOD: Participants were 7,503 unrelated adolescents (54.5% female, ages 9-11 years at baseline) from the multisite, longitudinal Adolescent Brain Cognitive Development (ABCD) Study. An existing baseline STB history classification model was applied to predict longitudinal first-onset STB in adolescents compared with healthy controls and clinical controls (individuals with a mental health disorder but no STB). A new elastic net logistic regression model with 195 features was trained on data from 14 sites (n = 5,220), and the resulting top 15 features were validated at 7 independent sites (n = 2,283). RESULTS: The previously developed model to classify STB lifetime history also prospectively predicted first-onset STB in adolescents with an area under the curve (AUC) [95% CI] of 0.73 [0.70, 0.75], p < .001, compared with healthy controls and AUC [95% CI] of 0.63 [0.60, 0.66], p < .001, compared with clinical controls. The newly trained model with top 15 features performed similarly with AUC [95% CI] of 0.73 [0.71, 0.76], p < .001, and AUC [95% CI] of 0.64 [0.60, 0.66], p < .001, for the same comparison groups. The most consistent predictors across models included female sex, sleep disturbances, and maladaptive home and school environments. CONCLUSION: The models predicted first-onset STB in adolescents with moderate accuracy. This study also confirmed the roles of well-established psychological risk factors for STB and identified several novel neurocognitive and brain imaging risk factors. Future studies should validate these models in large-scale diverse samples before clinical translation. PLAIN LANGUAGE SUMMARY: This study followed over 7,500 adolescents for 4 years and tested 2 machine learning models using psychological, social, and brain data to identify those at risk of experiencing suicidal thoughts or behaviors. Both models predicted first-time suicidal thoughts or behaviors with moderate accuracy. Key risk factors that were identified included being female, experiencing sleep problems, and negative home and school environments. DIVERSITY & INCLUSION STATEMENT: We worked to ensure sex and gender balance in the recruitment of human participants. We worked to ensure race, ethnic, and/or other types of diversity in the recruitment of human participants. We worked to ensure that the study questionnaires were prepared in an inclusive way. Diverse cell lines and/or genomic datasets were not available. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented racial and/or ethnic groups in science. One or more of the authors of this paper self-identifies as a member of one or more historically underrepresented sexual and/or gender groups in science. We actively worked to promote sex and gender balance in our author group. One or more of the authors of this paper received support from a program designed to increase minority representation in science. We actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our author group. While citing references scientifically relevant for this work, we also actively worked to promote sex and gender balance in our reference list. While citing references scientifically relevant for this work, we also actively worked to promote inclusion of historically underrepresented racial and/or ethnic groups in science in our reference list. The author list of this paper includes contributors from the location and/or community where the research was conducted who participated in the data collection, design, analysis, and/or interpretation of the work.

Adolescent↗

PubMed2025

Comparing artificial and convolutional neural networks with traditional models for Genomic prediction in wheat.

With the rapid development of sequencing technology, the application of genomic prediction has become more and more common in breeding schemes of livestocks and crops. Selecting an appropriate statistical model is of central importance to achieve high prediction accuracy. Recently, machine learning models have been expected to upgrade genomic prediction into a new era. However, the perspective still suffers from lack of evidence that machine learning models can generally outperform the traditional ones on empirical data sets. In this study, we compared two machine learning models based on artificial neural network (ANN) and convolutional neural network (CNN) with four traditional models, including genomic best linear unbiased prediction (GBLUP), Bayesian ridge regression (BRR), BayesA and BayesB, using three published data sets for grain yield in wheat. For each model, we considered two variants: modeling and ignoring the genotype-by-environment ([Formula: see text]) interaction. In the comparison, we considered two strategies of cross-validation: predicting genotypes that have not been evaluated in any environment (CV1) and predicting genotypes that have been tested in other environments (CV2). Our results showed that traditional Bayesian models (BayesA, BayesB, and BRR) outperformed GBLUP, ANN and CNN when considering [Formula: see text] interaction. The accuracies of ANN and CNN were higher than traditional models only in CV1 and when [Formula: see text] interaction was ignored. It was also found that the performance of the two machine learning models was significantly affected by the interaction between the CV strategy and the way of treating the [Formula: see text] interaction, while that of the four traditional models was only influenced by whether the [Formula: see text] interaction was considered or not. Thus, machine learning models can be a powerful complementary to the traditional ones and their superiority may depend on the prediction scenario. Among the two machine learning models, we observed that the accuracy of ANN was higher than CNN in most cases, indicating that it is still challenging to adapt complex machine learning models such as CNN to genomic prediction.

ANN↗

PubMed2025

Prematurity and Genetic Liability for Autism Spectrum Disorder.

BACKGROUND: Autism Spectrum Disorder (ASD) is a neurodevelopmental condition characterized by diverse presentations and a strong genetic component. Environmental factors, such as prematurity, have also been linked to increased liability for ASD, though the interaction between genetic predisposition and prematurity remains unclear. This study aims to investigate the impact of genetic liability and preterm birth on ASD conditions. METHODS: We analyzed phenotype and genetic data from two large ASD cohorts, the Simons Foundation Powering Autism Research for Knowledge (SPARK) and Simons Simplex Collection (SSC), encompassing 78,559 individuals for phenotype analysis, 12,519 individuals with genome sequencing data, and 8,104 individuals with exome sequencing data. Statistical significance of differences in clinical measures was evaluated between individuals with different ASD and preterm status. We assessed the rare variants burden using generalized estimating equations (GEE) models and polygenic load using ASD-associated polygenic risk score (PRS). Furthermore, we developed a machine learning model to predict ASD in preterm children using phenotype and genetic features available at birth. RESULTS: Individuals with both preterm birth and ASD exhibit more severe phenotypic outcomes despite similar levels of genetic liability for ASD across the term and preterm groups. Notably, preterm ASD individuals showed an elevated rate of de novo variants identified in exome sequencing (GEE model, p=0.005) in comparison to the non-ASD preterm group. Additionally, a GEE model showed that a higher ASD PRS, preterm birth, and male sex were positively associated with a higher predicted probability for ASD, reaching a probability close to 90% in SPARK. Lastly, we developed a machine learning model using phenotype and genetic features available at birth with limited predictive power (AUROC = 0.65). CONCLUSIONS: Preterm birth may exacerbate the multimorbidity present in ASD, which was not due to the ASD genetic factors. However, increased genetic factors may elevate the likelihood of a preterm child being diagnosed with ASD. Additionally, a polygenic load of ASD-associated variants had an additive role with preterm birth in the predicted probability for ASD, especially for boys. We propose that incorporating genetic assessment into neonatal care could benefit early ASD identification and intervention for preterm infants.

Autism Spectrum Disorder↗

PubMed2026

AI-Supported, Integrative Prediction of Postoperative Delirium: Protocol for the CONFUSED Study.

BACKGROUND: Postoperative delirium (POD) is a frequent and serious complication in older surgical patients, characterized by acute cognitive dysfunction and fluctuating levels of consciousness. POD is associated with prolonged hospitalization, long-term cognitive decline, reduced quality of life, and increased mortality. Despite its clinical relevance, the underlying pathophysiological mechanisms remain poorly understood, and reliable biomarkers for early prediction and prevention are lacking. OBJECTIVE: The CONFUSED study aims to identify molecular and clinical predictors of POD by integrating clinical data with proteomic, transcriptomic, and epigenetic analyses. The primary objective is to develop predictive models for POD using multimodal data. Secondary objectives include the identification of delirium-associated genes, proteins, and epigenetic signatures, as well as the exploration of patient subgroups at increased risk for POD. METHODS: CONFUSED is a prospective observational cohort study conducted at a German university hospital. Adult patients undergoing major surgery under general anesthesia will be enrolled until 100 cases of POD have been observed, which is expected to require a total sample size of approximately 200 to 300 patients. Blood samples are collected at 4 predefined time points: before premedication, immediately after surgery, and on postoperative days 2 and 5. Samples undergo comprehensive proteomic profiling, transcriptomic analysis using RNA microarrays, DNA methylation analysis, and genotyping of selected polymorphisms. Clinical data, including demographics, comorbidities, perioperative variables, medications, and delirium assessments using the Confusion Assessment Method (CAM) and CAM for the intensive care unit, are systematically recorded. Statistical analyses include univariate and multivariate methods, as well as machine learning approaches such as random forests and support vector machines, to identify relevant biomarkers and develop predictive models. The study protocol follows STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) and TRIPOD (Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis) guidelines and was approved by the responsible ethics committees. RESULTS: The study was registered in the German Clinical Trials Register (DRKS00033854) on March 18, 2024. Recruitment started in January 2024 and is ongoing at the time of manuscript submission. As of now, 135 patients have been enrolled. Sample collection and laboratory analyses are ongoing. Data analysis began in January 2026, with first results anticipated in July 2026. Final data lock is anticipated after the completion of recruitment. CONCLUSIONS: By integrating multimodal molecular data with clinical parameters and applying advanced machine learning techniques, the CONFUSED study aims to improve the prediction and understanding of POD. The results are expected to support the development of personalized preventive strategies and contribute to improved perioperative care for patients at risk of POD.

Humans↗

PubMed2026

Predicting cellular responses to perturbation across diverse contexts with State.

While machine learning models offer potential for predicting transcriptomic effects of perturbation, they currently struggle to generalize across cellular contexts. Here, we introduce State, a machine learning model that predicts perturbation effects while accounting for cellular heterogeneity within and across experiments. State is trained using single-cell gene expression data to predict perturbation effects across sets of cells. State improved discrimination of effects on large datasets by more than 30% and identified differentially expressed genes across genetic, signaling, and chemical perturbations with significantly improved accuracy compared with baselines. Its cell embeddings trained on observational data from 167 million cells enable the identification of strong perturbations in cellular contexts where no perturbations were observed during training. We further introduce Cell-Eval, a comprehensive evaluation framework that can be used to evaluate future models. Overall, the performance and flexibility of State set the stage for scaling the development of AI models of cell state.

Machine Learning↗

PubMed2026

Prediction of antimicrobial minimum inhibitory concentration from bacterial genomes using a scalable and interpretable machine learning approach.

Although machine learning models can predict antimicrobial susceptibility from bacterial whole genome sequencing (WGS), state-of-the-art approaches are computationally demanding or dependent on knowledge of genetic resistance determinants. Here, we describe an efficient data-driven approach to predicting minimum inhibitory concentration (MIC) by progressively extending and refining predictive genome segments, independent of prior knowledge of resistance determinants. Resultant models had high interpretability - known and potentially novel resistance determinants were captured. Using 762 clinical E. coli strains, 71.6% of predictions were within one dilution of the measured MIC. Models trained with this algorithm generalised better onto external data (F1 score = 0.85) compared with alternative models trained on annotated resistance determinants (F1 = 0.82) or k-mer counts (F1 = 0.74). Computational demands were low (RAM usage 23.6GB vs 38.8GB for k-mer model). These advantages represent an important advance in predicting antimicrobial susceptibility from WGS, with potential applications for clinical diagnostics, drug development, and surveillance.

Journal Article↗

PubMed2026

Machine Learning-Driven Prediction of Coronary Artery Disease Risk Based on UK Biobank Plasma Proteomics.

BACKGROUND: Coronary artery disease (CAD) is a leading global cause of mortality, yet the predictive accuracy of conventional risk models is limited. Here, we integrate conventional risk factors, polygenic risk scores, and large-scale proteomics to develop a unified model for enhanced CAD risk prediction. METHODS: Using data from UK Biobank, participants with plasma proteomics and genetic risk data were included after excluding prevalent CAD. Participants from England were split into training (n=32 330) and internal validation (n=13 857) sets, and Scotland/Wales participants formed an external validation set (n=5775). Incident CAD was ascertained from linked health records. A 202-protein proteomic risk score was derived by least absolute shrinkage and selection operator Cox regression, and CatBoost models were trained using conventional risk factors alone and with incremental addition of polygenic risk scores and protein proteomic risk scores; Shapley Additive Explanations-guided forward selection identified a compact protein panel. RESULTS: Across cohorts, the median age was 58 years and ∼45% were men. Protein proteomic risk score was dose-dependently associated with CAD risk. Compared with conventional risk factors alone, integrating polygenic risk scores and protein proteomic risk scores improved discrimination, with the area under the curve increasing from 0.750 (95% CI, 0.732-0.767) to 0.789 (95% CI, 0.772-0.805) in internal validation and from 0.717 (95% CI, 0.683-0.750) to 0.762 (95% CI, 0.732-0.791) in external validation. A 9-protein panel (GDF15 [growth differentiation factor 15], MMP12 [matrix metalloproteinase 12], NPPB [natriuretic peptide B], PGF [placental growth factor], REN [renin], ADGRG2 [adhesion G-protein coupled receptor], ACE2 [angiotensin-converting enzyme 2], CDCP1 [CUB domain-containing protein 1], CXCL17 [C-X-C motif chemokine ligand 17)]) captured most proteomic predictive information. CONCLUSIONS: Our findings demonstrate that integrating conventional risk factors, polygenic risk scores, and proteomic data improves CAD risk prediction. This study highlights the utility of proteomics in precision cardiovascular medicine and simplified risk stratification tools.

Humans↗