SEARCH · PubMed Health

Results for “Machine Learning”

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 55 records · Page 3Linked to original sources

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

Research on identification of key genes and immune-metabolic mechanisms in atrial fibrillation through integrated multi-cohort transcriptomic analysis and machine learning.

This study aimed to integrate multiple datasets for the identification of atrial fibrillation (AF)-related differentially expressed genes (DEGs), analyze their underlying mechanisms through functional enrichment and machine learning, construct diagnostic models, and explore immune-metabolic interactions to provide novel biomarkers and theoretical foundations. Gene expression datasets were integrated and normalized, with batch effects removed using principal component analysis. Differential expression analysis, functional enrichment analysis (Gene Ontology and Kyoto Encyclopedia of Genes and Genomes pathways), and machine learning-based feature gene selection and model construction were performed. Shapley additive explanations analysis was utilized to interpret the constructed models, while gene set enrichment analysis, gene set variation analysis, and immune cell infiltration analysis were conducted to investigate the associations between feature genes and immune infiltration. After integrating and normalizing gene expression data and eliminating batch effects via principal component analysis, 6 DEGs were identified, including 4 upregulated and 2 down-regulated ones. Functional enrichment analysis showed these DEGs were significantly enriched in neuro-related biological processes and pathways, indicating their key roles in AF pathogenesis. Five key feature genes were selected using LASSO, random forest, and support vector machine-recursive feature elimination algorithms. They had significant expression differences between the AF and control groups (P < .001) and were located on distinct chromosomes. The constructed random forest and support vector machine models performed excellently (area under the curve ≥ 0.85). Shapley additive explanations analysis revealed TNNI1 contributed most to model prediction, with its expression significantly positively correlated with immune cell infiltration. Gene set enrichment analysis and gene set variation analysis analyses further showed feature genes participated in AF pathogenesis by regulating immune modulation, metabolic pathways, and autophagy. Immune cell infiltration analysis found altered proportions of T-cell subsets and M0 macrophages in the AF group, along with complex links between feature gene expression and immune cell function. This study systematically elucidated the unique gene expression patterns and key regulatory pathways associated with AF, clarifying the crucial roles of feature genes in immune regulation, metabolic imbalance, and cellular dysfunction. These findings provide a theoretical basis and potential therapeutic targets for understanding AF pathogenesis and developing targeted treatment strategies.

Atrial Fibrillation↗

PubMed2024

Prediction of metabolic syndrome using machine learning approaches based on genetic and nutritional factors: a 14-year prospective-based cohort study.

INTRODUCTION: Metabolic syndrome is a chronic disease associated with multiple comorbidities. Over the last few years, machine learning techniques have been used to predict metabolic syndrome. However, studies incorporating demographic, clinical, laboratory, dietary, and genetic factors to predict the incidence of metabolic syndrome in Koreans are limited. In the present study, we propose a genome-wide polygenic risk score for the prediction of metabolic syndrome, along with other factors, to improve the prediction accuracy of metabolic syndrome. METHODS: We developed 7 machine learning-based models and used Cox multivariable regression, deep neural network (DNN), support vector machine (SVM), stochastic gradient descent (SGD), random forest (RAF), Naïve Bayes (NBA) classifier, and AdaBoost (ADB) to predict the incidence of metabolic syndrome at year 14 using the dataset from the Korean Genome and Epidemiology Study (KoGES) Ansan and Ansung. RESULTS: Of the 5440 patients, 2,120 were considered to have new-onset metabolic syndrome. The AUC values of model, which included sex, age, alcohol intake, energy intake, marital status, education status, income status, smoking status, dried laver intake, and genome-wide polygenic risk score (gPRS) Z-score based on 344,447 SNPs (p-value < 1.0), were the highest for RAF (0.994 [95% CI 0.985, 1.000]) and ADB (0.994 [95% CI 0.986, 1.000]). CONCLUSIONS: Incorporating both gPRS and demographic, clinical, laboratory, and seaweed data led to enhanced metabolic syndrome risk prediction by capturing the distinct etiologies of metabolic syndrome development. The RAF- and ADB-based models predicted metabolic syndrome more accurately than the NBA-based model for the Korean population.

Humans↗

PubMed2026

Integrated transcriptome analysis and machine learning to construct a homeostatic model of acetylation for bladder cancer and validate the key gene CES1.

BACKGROUND: Bladder cancer (BLCA) is one of the most common malignant tumors of the urinary system. Protein acetylation (PA) plays a critical role in regulating multiple biological processes (BPs), cellular homeostasis, and cancer-related signaling pathways. This study aimed to construct a homeostatic model of acetylation for BLCA using integrated transcriptome analysis and machine learning and to validate the key gene CES1. METHODS: RNA sequencing (RNA-seq) and clinical data were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Acetylation-related differentially expressed genes (DEGs) in BLCA were screened using differential expression analysis (DEA). An acetylation homeostatic model was constructed via univariate, machine learning-based least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression analyses, followed by validation in multiple cohorts. Single-cell RNA-seq analysis was used to explore gene expression patterns in diverse cell types. Enrichment analysis (EA), immune infiltration, and drug sensitivity analysis (DSA) were performed to characterize molecular features of different risk groups. Finally, the biological function of CES1 as the key gene was verified by in vitro knockdown experiments. RESULTS: We established a robust acetylation homeostatic model consisting of five genes, which effectively predicted overall survival (OS) and served as an independent prognostic factor in BLCA. High-risk patients showed significantly poorer prognosis, distinct immune infiltration profiles, and differential drug sensitivity. CES1 was identified and validated as the key gene in this model, which was highly expressed in BLCA and associated with poor prognosis. Knockdown of CES1 markedly suppressed cell proliferation, invasion, and migration, and reduced intracellular coenzyme A (CoA) levels, thereby regulating PA homeostasis. CONCLUSIONS: We developed and validated a novel acetylation homeostatic model for survival stratification and personalized treatment guidance in BLCA, based on integrated transcriptome analysis and machine learning. CES1 is closely associated with intracellular CoA levels and the malignant progression of BLCA. Its potential association with PA homeostasis requires further mechanistic validation, and it may act as a candidate therapeutic biomarker for BLCA.

Bladder cancer (BLCA)↗

PubMed2025

Predicting natural variation in the yeast phenotypic landscape with machine learning.

Most organismal traits result from the complex interplay of many genetic and environmental factors, making their prediction difficult. Here, we used machine learning (ML) models to explore phenotype predictions for 223 traits measured across 1011 genome-sequenced Saccharomyces cerevisiae strains isolated worldwide. We benchmarked a ML pipeline with multiple linear and non-linear models to predict phenotypes from genotypes and gene expression, and determined gradient boosting machines as the best-performing model. Gene function disruption scores and gene presence/absence emerged as best predictors, suggesting a considerable contribution of the accessory genome in controlling phenotypes. The prediction accuracy broadly varied among phenotypes, with stress resistance being easier to predict compared to growth across nutrients. ML identified relevant genomic features linked to phenotypes, including high-impact variants with established relationships to phenotypes, despite these being rare in the population. Near-perfect accuracies were achieved when other phenomics data mostly in similar conditions were used, suggesting that useful information can be conveyed across phenotypes. Overall, our study underscores the power of ML to interpret the functional outcome of genetic variants.

Genetic Variation↗

PubMed2026

Dynamic evolution of chaperone-mediated autophagy is associated with tumor microenvironment remodeling and prognostic stratification in lung adenocarcinoma: insights from single-cell transcriptomics, ensemble machine learning, and experimental validation.

BACKGROUND: Lung adenocarcinoma (LUAD) shows prognostic heterogeneity, and tumor-node-metastasis (TNM) staging is limited for individualized management. Chaperone-mediated autophagy (CMA) maintains proteostasis, but its role during adenocarcinoma in situ (AIS)-minimally invasive adenocarcinoma (MIA)-invasive adenocarcinoma (IAC) progression remains unclear. METHODS: Single-cell RNA sequencing (scRNA-seq) data from GSE189357 and bulk transcriptomes from The Cancer Genome Atlas (TCGA)-LUAD and Gene Expression Omnibus (GEO) cohorts were integrated. CMA activity, cell-cell communication, weighted gene co-expression network analysis (WGCNA), tumor-normal differential expression, machine-learning survival modeling, tumor microenvironment (TME) features, drug sensitivity, and EPC1 function were analyzed. RESULTS: CMA-high tumor epithelial cells increased from AIS (58.1%) to MIA (65.7%) but declined in IAC (44.4%; p < 0.001). CMA-low cells preferentially received fibroblast-derived extracellular matrix cues. A CMA-negatively correlated module identified 69 core genes. Random survival forest (RSF) performed best among 117 machine-learning combinations (mean concordance index > 0.873). High-risk patients had worse survival across cohorts, and the risk score was independently associated with overall survival (hazard ratio = 16.013, 95% confidence interval: 9.579-26.768, p < 0.001). High-risk tumors showed proliferative activation and M0 macrophage enrichment, whereas low-risk tumors showed stronger immune-related signaling. EPC1 overexpression suppressed malignant phenotypes in A549 cells. CONCLUSION: CMA dynamics are associated with stromal and immune remodeling during LUAD progression. A CMA-based model provides robust prognostic stratification and may offer a basis for future TME-guided studies.

Chaperone-mediated autophagy↗

PubMed2026

Investigating the mechanisms of PhIP-induced colorectal cancer through network toxicology, machine learning, and molecular dynamics simulation.

BACKGROUND: Over the past few years, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP)- a compound from grilled or processed meats-has emerged as a major player in cancer development, especially colorectal cancer (CRC). This work dives into its potential links to CRC and uncovers the key genes that bridge this connection. METHODS: We tapped into various databases to pinpoint target genes tied to PhIP and CRC, then ran protein-protein interaction (PPI) analyses for visualization. Next, we explored underlying mechanisms through Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment. To nail down predictions, we tested 107 machine learning pipelines and picked the best one, validating its accuracy and the core genes' prognostic value across datasets. Next, molecular docking and dynamics simulations probed the interactions between these genes and PhIP. Finally, cell proliferation was assessed using Cell Counting Kit-8 (CCK-8) and 5-ethynyl-2'-deoxyuridine (EdU) assays, and polymerase chain reaction (PCR) was performed to validate the expression levels of the hub genes. RESULTS: Our analysis identified 39 overlapping genes, from which a machine learning model (glmBoost + Enet) identified six candidate targets: CDK4, CEBPB, COMT, SOX9, TIMP1, and TOP2A. To prioritize these, a hierarchical screening framework was applied. Molecular docking and dynamics simulations identified CDK4, COMT, and TIMP1 as the most stable interactors with PhIP. Functional assays confirmed that PhIP treatment significantly enhanced the proliferation of CRC cells. Crucially, quantitative PCR (qPCR) validation in multiple CRC cell lines identified TIMP1 as the primary target, showing the most consistent and significant upregulation upon PhIP exposure. CONCLUSIONS: In essence, these genes drive PhIP is role in CRC, offering novel insights into its molecular pathways. This could reshape how we tackle food-related pollutants, paving the way for better prevention and targeted therapies.

Colorectal cancer (CRC)↗

PubMed2026

Integrated single-cell transcriptomics, Mendelian randomization, and machine learning identify CEBPZ as an immune-related biomarker in oral lichen planus.

BACKGROUND: Oral lichen planus (OLP) is a chronic, immune-mediated oral mucosal disease with complex pathophysiology and potential for malignant transformation. Understanding its molecular basis is critical for the development of precise diagnostic and therapeutic strategies. OBJECTIVES: We aimed to identify key immune-related biomarkers and characterize cellular dynamics in OLP, with a particular focus on the role of CEBPZ in disease pathogenesis. MATERIAL AND METHODS: We analyzed single-cell RNA sequencing (scRNA-seq) data from OLP lamina propria samples (GSE211630) to identify disease-specific T-cell subpopulations using high-dimensional weighted gene co-expression network analysis (hdWGCNA) for oxidative stress-related gene modules.-data-based Mendelian randomization (SMR) integrated FinnGen genome-wide association study (GWAS; 342,499 Europeans) data with Genotype-Tissue Expression (GTEx) expression quantitative trait loci (eQTL) data to identify causal genes. Machine learning (ML) models (least absolute shrinkage and selection operator (LASSO) and convolutional neural network (CNN)) were developed using bulk RNA-seq datasets (GSE52130 and GSE38616) for diagnostic purposes. RESULTS: We identified OLP-specific T-cell populations (clusters 0, 3, 5, 7, 13, and 15) with enhanced migration inhibition factor (MIF) pathway signaling toward B cells and monocytes. Two oxidative stress-associated modules contained hub genes, including CEBPZ. Summary-data-based Mendelian randomization analysis identified 231 OLP-associated genes, with CEBPZ uniquely intersecting LASSO-selected markers (odds ratio (OR) = 1.057, 95% confidence interval (95% CI) = 1.013-1.102, p = 0.010). Machine learning models achieved area under the curve (AUC) values ranging from 0.653 to 0.745, with the CNN model reaching a validation accuracy of 0.735. CEBPZ showed elevated expression in OLP T cells and correlated with enhanced MIF-(CD74+CXCR4) signaling. CONCLUSIONS: This integrative approach identifies CEBPZ as a pivotal biomarker linking genetic susceptibility, oxidative stress, and immune dysregulation in OLP. Our diagnostic models offer promising tools for OLP management.

CEBPZ↗

PubMed2026

Machine learning and multi-omics clustering to map cellular rewiring and immune evasion in ccRCC.

Immune checkpoint blockade (ICB) efficacy in clear cell renal cell carcinoma (ccRCC) is limited by tumor microenvironment (TME) heterogeneity. Because traditional bulk-derived models lack spatial resolution, we developed an integrated framework connecting macroscopic survival risks to microscopic TME structures. We applied ten algorithms to establish multi-omics subtypes and evaluated 101 machine-learning combinations across three independent cohorts to generate a Consensus Machine Learning-driven Signature (CMLS). The signature's spatial and cellular origins were decoded using spatial transcriptomics (ST) and a 140,000-cell scRNA-seq atlas. Expression of key genes was experimentally validated via RT-qPCR in 17 paired ccRCC clinical tissues. We identified two molecular subtypes with distinct clinical and epigenetic profiles. SuperPC optimization yielded a 24-gene CMLS serving as an independent prognostic factor. scRNA-seq and ST deconvolution revealed these signals predominantly originate from cancer-associated fibroblasts (CAFs) and malignant epithelial cells, which collaborate to drive spatial immune exclusion. RT-qPCR confirmed significant overexpression of five core CMLS genes in ccRCC versus adjacent normal tissues. Low CMLS scores correlated with enhanced ICB responsiveness, whereas high-CMLS tumors demonstrated specific vulnerability to dasatinib and dabrafenib. The CMLS translates spatial immune-exclusion dynamics into a quantifiable metric, outperforming tumor mutational burden in predicting ICB benefits, providing a robust tool for patient stratification in ccRCC.

Humans↗

PubMed2026

Revealing potential biomarkers and metabolic mechanisms of ovarian aging in hens during late laying period based on machine learning and metabolomics.

Ovarian function decline during the late laying period represents a major bottleneck for the economic efficiency of the global poultry industry. However, the underlying metabolic mechanisms and reliable early-warning biomarkers for ovarian aging remain poorly understood. In this study, we performed the first untargeted LC-MS/MS metabolomics analysis of ovarian tissues from Taihe silky fowls at peak laying (30 weeks) and late laying (50 weeks) stages, and employed an ensemble machine learning strategy integrating LASSO, random forest, and support vector machine (SVM) algorithms to identify high-confidence core biomarkers of ovarian aging. Gene expression analysis was further conducted to validate the potential molecular mechanisms. Our results showed that the metabolic profiles of ovarian tissues differed significantly between the two groups. A total of 6 core biomarkers were identified, 4 of which were long-chain acylcarnitines. Mechanistic analysis revealed that downregulation of key genes in the carnitine shuttle system led to impaired mitochondrial fatty acid β-oxidation, which in turn triggered excessive oxidative stress and compromised ovarian endocrine function. In conclusion, this study identifies long-chain acylcarnitines as potential metabolic biomarkers for ovarian aging in Taihe silky fowls. These findings provide novel insights into the metabolic basis of poultry ovarian aging and lay a theoretical foundation for the precise regulation of reproductive performance in indigenous poultry breeds.

Animals↗

PubMed2025

MicroRNAs signatures in small extracellular vesicles for psychological resilience in young adults using machine learning.

AIMS: Psychological resilience refers to an individual's capacity to adapt to adverse events. MicroRNAs (miRNAs) play a crucial role in regulating post-transcriptional processes, while small extracellular vesicles (sEVs) act as transport vehicles. This study aimed to employ genome-wide profiling to identify and validate differences in the expression of resilience-associated sEV-miRNAs between low resilience (LR) and high resilience (HR) in young adults. METHODS: Eighty participants were divided into LR or HR based on the Connor - Davidson Resilience Scale (CD-RISC). The expression levels of the target sEV-miRNAs in LR and HR were compared and analyzed. RESULTS: Expression analyses demonstrated significant differences in let-7b, miR-151b, miR-335, and miR-193a between LR and HR (p < 0.01), with let-7b showing the highest discriminative ability. The AUC values for each sEV-miRNA ranged from 0.74 to 0.94, based on logistic regression and three machine learning models: random forest, support vector machine, and eXtreme gradient boosting. Based on leave-one-out cross-validation in different models, the combined four sEV-miRNAs demonstrated strong performance for detecting LR (AUC = 0.87-0.90). Sex-specific differences were also observed, with female participants showing more pronounced resilience signatures in targeted sEV-miRNAs. CONCLUSIONS: These findings suggest that sEV-miRNAs hold potential as biomarkers for psychological resilience in young adults.

Humans↗

PubMed2025

Assessing individual genetic susceptibility to metabolic syndrome: interpretable machine learning method.

BACKGROUND: Genome-wide association studies have provided profound insights into the genetic aetiology of metabolic syndrome (MetS). However, there is a lack of machine-learning (ML)-based predictive models to assess individual genetic susceptibility to MetS. This study utilized single-nucleotide polymorphisms (SNPs) as variables and employed ML-based genetic risk score (GRS) models to predict the occurrence of MetS, bringing it closer to clinical application. METHODS: Feature selection was performed using Least Absolute Shrinkage and Selection Operator. Six ML algorithms were employed to construct GRS models. A fivefold cross-validation was utilized to aid in the internal validation of models. The receiver operating characteristic (ROC) curve was used to select the better-performing GRS model. The SHapley Additive exPlanations (SHAP) was then applied to interpret the model. After extracting GRS, stratified analysis of BMI, age and gender was performed. Finally, these conventional risk factors and GRS were integrated through multivariate logistic regression to establish a combined model. RESULTS: A total of 17 SNPs were selected for analysis. Among the GRS models, the extreme gradient boosting (XGBoost) model demonstrated superior discriminative performance (AUC = 0.837). The XGBoost's optimal robustness was also validated through five-fold cross-validation (mean ROC-AUC = 0.706). The XGBoost-based SHAP algorithm not only elucidated the global effects of 17 SNPs across all samples, but also described the interaction between SNPs, providing a visual representation of how SNPs impact the prediction of MetS in an individual. There was a strong correlation between GRS and MetS risk, particularly observed among young individuals, males and overweight individuals. Furthermore, the model combining conventional risk factors and GRS exhibited excellent discriminative performance (AUC = 0.962) and outstanding robustness (mean ROC-AUC = 0.959). CONCLUSION: This study established a reliable XGBoost-based GRS model and a GRS prediction platform (https://metabolicsyndromeapps.shinyapps.io/geneticriskscore/) to assess individual genetic susceptibility to MetS. This model has high interpretability and can provide personalized reference for determining the necessity of primary prevention measures for MetS. Additionally, there may be interactions between traditional risk factors and GRS, and the integration of both in a comprehensive model is useful in the prediction of MetS occurrence.

Humans↗

PubMed2026

Artificial Intelligence and Machine Learning Applications in Fibromuscular Dysplasia: Transforming Diagnosis, Risk Stratification, and Clinical Decision-Making.

Fibromuscular dysplasia (FMD) is a non-atherosclerotic vascular disorder with heterogeneous presentations, making diagnosis and management highly dependent on imaging and clinical expertise. This narrative review examines how artificial intelligence (AI) and machine learning (ML) are transforming FMD care. AI-enhanced imaging, particularly convolutional neural network-based analysis, improves detection of the characteristic "string-of-beads" pattern on CT angiography, magnetic resonance angiography, and ultrasound, although FMD-specific validation remains limited. ML models facilitate risk stratification, prediction of disease progression, and early identification of complications such as aneurysms and stroke by integrating clinical, imaging, and genomic data. AI-driven clinical decision support systems further enable personalized treatment selection through pharmacogenomic insights and robot-assisted interventions. Despite promising real-world applications, challenges persist, including limited large-scale datasets, workflow integration, regulatory barriers, and algorithmic bias affecting underrepresented populations. Future advances in explainable AI, federated learning, and digital health integration may enable a shift toward predictive, patient-centered FMD management.

Humans↗

PubMed2026

Diagnostic Performance of Machine Learning for Systemic Lupus Erythematosus: Systematic Review and Meta-Analysis.

BACKGROUND: Early and accurate diagnosis of systemic lupus erythematosus (SLE) and its organ involvement is essential. Previous reviews of machine learning (ML) in SLE combined heterogeneous tasks and validation strategies and may have overinterpreted model performance. OBJECTIVE: This study evaluated the diagnostic performance of ML and deep learning (DL) models for 3 clinically distinct SLE-related tasks: SLE classification or diagnosis, lupus nephritis (LN) diagnosis, and neuropsychiatric systemic lupus erythematosus (NPSLE) discrimination. We also assessed methodological quality and certainty of evidence. METHODS: PubMed, Embase, Cochrane Library, Web of Science, and IEEE Xplore were searched from January 2014 to April 2026. Eligible peer-reviewed diagnostic accuracy studies developed or validated ML or DL models for 1 of the 3 prespecified tasks, used an accepted reference standard, and provided data for a 2×2 contingency table. Bivariate random-effects meta-analyses with the Hartung-Knapp-Sidik-Jonkman adjustment were used to pool sensitivity and specificity. We reported 95% prediction intervals (PIs), assessed risk of bias using the Quality Assessment of Diagnostic Accuracy Studies for Artificial Intelligence tool (QUADAS-AI; Viknesh Sounderajah [Imperial College London]), and evaluated certainty of evidence using the Grading of Recommendations Assessment, Development, and Evaluation framework for diagnostic test accuracy. RESULTS: Twenty-nine studies were included: 17 for SLE classification, 5 for LN diagnosis, and 7 for NPSLE discrimination. In the primary task-stratified analysis, pooled sensitivity was 0.91 (95% CI 0.86-0.94; 95% PI 0.56-0.99), and pooled specificity was 0.94 (95% CI 0.91-0.96; 95% PI 0.69-0.99), with low heterogeneity (I²=23.9% and 22.9%, respectively). DL models showed a sensitivity of 0.93 and specificity of 0.95, compared with 0.88 and 0.94 for traditional ML models. Certainty of evidence was high for most analyses but low for LN diagnosis because of inconsistency and imprecision. All studies were retrospective, and only 9 of 29 (31%) performed independent external validation. Overall risk of bias was high or unclear in 22 of 29 (75.9%) studies. No study reported model calibration, decision-curve analysis, or net clinical benefit. CONCLUSIONS: ML models showed promising diagnostic accuracy across 3 distinct SLE-related tasks, but wide PIs, limited external validation, and pervasive risk of bias restrict conclusions about real-world generalizability. Prospective multicenter studies with standardized tasks and reference standards, independent external validation, and formal assessment of calibration and clinical utility are required before clinical implementation.

Humans↗

PubMed2026

Diagnostic performance of machine learning models for malignant and non-malignant pleural effusion: Systematic review and meta-analysis.

BACKGROUND: Accurately distinguishing malignant pleural effusion (MPE) from non-malignant pleural effusion is clinically important, but the generalisability and methodological quality of machine-learning (ML) models remain uncertain. METHODS: We searched eight databases to 23 April 2026. Diagnostic performance was pooled using random-effects and Reitsma bivariate models, and study quality was assessed using PROBAST+AI. RESULTS: Forty-two studies were included; 17 contributed to the AUC meta-analysis and 14 to the bivariate analysis. The pooled AUC was 0.90 (95 % CI 0.85-0.94; 95 % prediction interval 0.62-0.98), with sensitivity of 0.80 (95 % CI 0.77-0.83) and specificity of 0.87 (95 % CI 0.79-0.92). Only nine studies reported external, temporal or independent validation. Externally validated studies had a lower pooled AUC than studies without external validation (0.83 vs 0.92), with lower specificity observed in the two externally validated studies contributing sensitivity and specificity data. All 42 development assessments had high overall quality concerns, and all 42 model evaluations were judged at high risk of bias. CONCLUSIONS: ML models showed good apparent accuracy for distinguishing MPE from non-MPE, but the evidence was limited by substantial heterogeneity, high risk of bias and scarce external validation. The pooled estimates reflect the average performance of different selected models rather than the expected accuracy of a single clinical test. ML models should be regarded as adjuncts to existing diagnostic pathways until they are confirmed by rigorous multicentre prospective external validation and clinical-impact studies.

Humans↗

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

Multimodal artificial intelligence and machine learning in oncology: from data integration to precision cancer care.

Cancer remains a major global health burden, with approximately 20 million new cases and 9.7 million cancer-related deaths reported globally in 2022. While advances in radiological imaging, molecular profiling, and clinical data have enhanced the interpretation of disease progression, the availability of multiple such modalities still does not meet the needs of a large patient population. This narrative review focuses on the role of multimodal artificial intelligence and machine learning in bridging the gap in interpreting heterogeneous modalities to improve risk prediction, prognostic assessment, and treatment decision-making in precision oncology. Multimodal frameworks such as Pathomic Fusion illustrate how complementary histopathological and genomic information can be integrated for cancer diagnosis and prognostic modeling. Multimodal models have demonstrated potential in virtual biopsy, cancer screening, prognostic prediction, radiotherapy planning, intraoperative guidance, and clinical-trial design using digital twins and synthetic control arms. The major limitations of incorporating multimodal artificial intelligence and machine learning in oncology include data heterogeneity, demographic or institutional biases, and reproducibility challenges that hinder translation. Accordingly, appropriate data-governance strategies, fairness audits, and privacy-preserving approaches such as federated learning should be considered where appropriate. Future progress will depend on the development of standardized benchmarking datasets, robust external validation, seamless integration with electronic health records and picture archiving and communication systems, and the implementation of explainable, secure, and clinically validated multimodal artificial intelligence frameworks that support precision oncology in routine clinical practice.

deep learning↗

PubMed2025

Identification of Biomarkers for Right Ventricular Dysfunction in Idiopathic Dilated Cardiomyopathy Via Urinary Proteomics and Machine Learning.

BACKGROUND: Right ventricular dysfunction (RVD) is a common complication of idiopathic dilated cardiomyopathy linked to poor outcomes. However, reliable noninvasive biomarkers for RVD remain lacking. This study aimed to identify urinary proteomic markers using mass spectrometry and machine learning. METHODS: In this prospective cohort, patients with idiopathic dilated cardiomyopathy were classified by cardiac magnetic resonance imaging into groups with RVD (RV ejection fraction <45%) and without RVD groups. Baseline urine samples were profiled by data-independent acquisition mass spectrometry. Differentially expressed proteins were identified and selected by least absolute shrinkage and selection operator regression to build a diagnostic model, developed in a training set, and validated in a test set. The primary end point was a composite of cardiovascular death, heart failure rehospitalization, left ventricular assist device implantation, or heart transplantation. RESULTS: The study enrolled 147 patients with idiopathic dilated cardiomyopathy (64 with RVD, 83 without), with a median follow-up of 19.3 months. Of 3579 quantified urinary proteins, 46 were differentially expressed between groups. A 3-protein panel (RARRES1 [retinoic acid receptor responder protein 1], MVB12B [multivesicular body subunit 12B], GSK3A [glycogen synthase kinase 3 alpha]) was identified and showed excellent diagnostic accuracy (training area under the curve 0.946; validation area under the curve0.935), outperforming both NT-proBNP (N-terminal pro-brain natriuretic peptide) and tricuspid annular plane systolic excursion. The risk score derived from this panel effectively stratified patients, with the high-risk group exhibiting significantly worse outcomes than the low-risk group (hazard ratio, 3.24 [95% CI, 1.56-6.71], P=0.002). CONCLUSIONS: The urinary proteomic panel developed in this study demonstrates diagnostic and prognostic potential for identifying RVD in idiopathic dilated cardiomyopathy, providing a promising noninvasive tool for precise detection and clinical risk stratification.

Humans↗