Harnessing Machine Learning, a Subset of Artificial Intelligence, for Early Detection and Diagnosis of Type 1 Diabetes: A Systematic Review DOI Open Access
Rahul Mittal, Matthew B. Weiss,

Arnaldo Rendon

et al.

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(9), P. 3935 - 3935

Published: April 22, 2025

Type 1 diabetes (T1D) is an autoimmune condition characterized by the destruction of insulin-producing pancreatic beta cells, leading to lifelong insulin dependence and significant complications. Early detection T1D essential delay disease onset improve outcomes. Recent advancements in artificial intelligence (AI) machine learning (ML) have provided powerful tools for predicting diagnosing T1D. This systematic review evaluates current landscape AI/ML-based approaches early detection. A comprehensive search across PubMed, EMBASE, Science Direct, Scopus identified 1447 studies, which 10 met inclusion criteria narrative synthesis after screening full-text review. The studies utilized diverse ML models, including logistic regression, support vector machines, random forests, neural networks. datasets encompassed clinical parameters, genetic risk markers, continuous glucose monitoring (CGM) data, proteomic metabolomic biomarkers. included involved a total 49,172 participants employed case–control, retrospective cohort, prospective cohort designs. Models integrating multimodal data achieved highest predictive accuracy, with area under curve (AUC) values reaching up 0.993 sex-specific models. CGM plasma biomarkers, such as CXCL10 IL-1RA, also emerged valuable identifying at-risk individuals. While results highlight potential AI/ML revolutionizing stratification diagnosis, challenges remain. Data heterogeneity limited model generalizability present barriers widespread implementation. Future research should prioritize development universal frameworks real-world validation enhance reliability integration these tools. Ultimately, technologies hold transformative practice enabling earlier guiding targeted interventions, improving long-term patient These could clinicians making more informed, timely decisions, thus reducing diagnostic delays paving way personalized prevention strategies both pediatric adult populations.

Language: Английский

Harnessing Machine Learning, a Subset of Artificial Intelligence, for Early Detection and Diagnosis of Type 1 Diabetes: A Systematic Review DOI Open Access
Rahul Mittal, Matthew B. Weiss,

Arnaldo Rendon

et al.

International Journal of Molecular Sciences, Journal Year: 2025, Volume and Issue: 26(9), P. 3935 - 3935

Published: April 22, 2025

Type 1 diabetes (T1D) is an autoimmune condition characterized by the destruction of insulin-producing pancreatic beta cells, leading to lifelong insulin dependence and significant complications. Early detection T1D essential delay disease onset improve outcomes. Recent advancements in artificial intelligence (AI) machine learning (ML) have provided powerful tools for predicting diagnosing T1D. This systematic review evaluates current landscape AI/ML-based approaches early detection. A comprehensive search across PubMed, EMBASE, Science Direct, Scopus identified 1447 studies, which 10 met inclusion criteria narrative synthesis after screening full-text review. The studies utilized diverse ML models, including logistic regression, support vector machines, random forests, neural networks. datasets encompassed clinical parameters, genetic risk markers, continuous glucose monitoring (CGM) data, proteomic metabolomic biomarkers. included involved a total 49,172 participants employed case–control, retrospective cohort, prospective cohort designs. Models integrating multimodal data achieved highest predictive accuracy, with area under curve (AUC) values reaching up 0.993 sex-specific models. CGM plasma biomarkers, such as CXCL10 IL-1RA, also emerged valuable identifying at-risk individuals. While results highlight potential AI/ML revolutionizing stratification diagnosis, challenges remain. Data heterogeneity limited model generalizability present barriers widespread implementation. Future research should prioritize development universal frameworks real-world validation enhance reliability integration these tools. Ultimately, technologies hold transformative practice enabling earlier guiding targeted interventions, improving long-term patient These could clinicians making more informed, timely decisions, thus reducing diagnostic delays paving way personalized prevention strategies both pediatric adult populations.

Language: Английский

Citations

0