PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
February 20, 2026SHILAP Revista de lepidopterología2 citationsOpen Access

Application of AI and digital health tools in public health management of T2DM: from mechanism prediction to personalized treatment

CYChonger Yu

Key Result

XGBoost reduced incident T2DM by 18% and 0.5% HbA1c in a Han Chinese cohort, DL models improved TIR by 15% without increasing hypoglycemia, and digital twins decreased severe hypoglycemia by 35% in T2DM patients.

Key Points

  • The aim is to assess the impact of AI and digital health tools on the management of type 2 diabetes.
  • Systematic review of research on AI applications in T2DM management
  • Integration of multi-omics and environmental data
  • Evaluation of AI in personalized treatment and public health interventions
  • AI enhances early screening and risk prediction of T2DM by revealing key mechanisms
  • Integration of digital tools improves treatment adherence and management efficiency
  • AI optimizes public health resource allocation and chronic disease prevention strategies

Structured PICO

P
Population
Populations at risk for or diagnosed with Type 2 diabetes mellitus (T2DM), with a specific focus on applications and disparities in low- and middle-income countries (LMICs).
I
Intervention
Artificial intelligence (AI) and digital health technologies (e.g., machine learning, deep learning, continuous glucose monitoring, wearable devices, and mobile health apps).

AI and digital health tools offer significant potential for precise prediction and personalized management of T2DM, but equitable global application requires addressing algorithmic bias, data privacy, and the digital divide in LMICs.

Main Result

Effect estimate: XGBoost model reduced incident T2DM by 18% with 0.5% HbA1c reduction; DL models increased TIR by 15% without hypoglycemia increase; Federated learning reduced HbA1c by 0.7%; Digital twins reduced severe hypoglycemia by 35%; Edge AI reduced diagnostic delays by 40%

Limitations

  • Lack of cross-ethnic validation for AI models; cross-population applicability limited due to genetic and environmental diversity
  • Limited scalability of digital twin and deep learning models in low- and middle-income countries
  • Data privacy and ethical concerns especially related to genetic data in LMICs
  • Overfitting of models on homogeneous or small datasets reduces generalizability
  • Digital divide and infrastructural limitations restrict AI and digital health technology deployment in LMICs
  • Model interpretability and explainability remain challenges affecting clinical adoption
  • Overfitting in AI-driven T2DM prediction models, particularly in multi-omics integration
  • Poor cross-population applicability and generalizability of models due to differences in genetic backgrounds and environments
  • Lack of LMIC representation in global genetic databases (over 90% of data comes from HICs)
  • Ethical dilemmas and data privacy concerns, especially in LMICs lacking dedicated regulations
  • Opaque internal working mechanisms of AI models ('black box' nature) limiting clinical trust

Abstract

Type 2 diabetes mellitus (T2DM) poses a significant global public health challenge, with its prevalence escalating continuously and disproportionately affecting low- and middle-income countries (LMICs), imposing a substantial burden on healthcare systems. Traditional management models have limitations in disease prediction, personalized treatment, and public health intervention. Artificial intelligence (AI) and digital health technologies provide novel insights for precise prediction and intelligent management of T2DM. This review systematically summarizes research progress in AI’s role in deciphering T2DM pathogenesis, personalized treatment, and public health management. By integrating multi-omics and environmental data, AI reveals key mechanisms including gene–environment (G × E) interactions, β -cell dysfunction, and inflammatory pathways, significantly enhancing early screening and risk prediction. In clinical management, AI combined with digital health tools e.g., continuous glucose monitoring (CGM), wearable devices, and mobile health (mHealth) apps facilitates remote monitoring, medication optimization, and personalized interventions, improving treatment adherence and health management efficiency. At the public health level, AI optimizes resource allocation and disease burden assessment, promoting chronic disease prevention and control model transformation. Future efforts should prioritize developing low-resource-adapted tools, strengthening data privacy protection tailored to LMICs, and addressing algorithmic fairness and the digital divide to ensure safe, equitable, and sustainable AI application in global T2DM management. Overall, AI and digital health integration is driving T2DM management towards an intelligent and precision-based era, with the potential to reduce disparities in LMICs.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Chonger Yu (2026) conducted a review in Adults with type 2 diabetes mellitus (T2DM) in low- and middle-income countries and diverse populations at risk of T2DM. AI-based prediction models including XGBoost, deep learning (DL), federated learning (FL), digital twins, edge AI vs. Traditional statistical methods or usual care was evaluated on 5-year incidence of type 2 diabetes mellitus or reduction in HbA1c and time-in-range (TIR) (XGBoost model reduced incident T2DM by 18% with 0.5% HbA1c reduction; DL models increased TIR by 15% without hypoglycemia increase; Federated learning reduced HbA1c by 0.7%; Digital twins reduced severe hypoglycemia by 35%; Edge AI reduced diagnostic delays by 40%). XGBoost reduced incident T2DM by 18% and 0.5% HbA1c in a Han Chinese cohort, DL models improved TIR by 15% without increasing hypoglycemia, and digital twins decreased severe hypoglycemia by 35% in T2DM patients.

synapsesocial.com/papers/6997f984ad1d9b11b345249dhttps://doi.org/10.3389/fpubh.2026.1756755
Ask AI
Helpful
Bookmark
Share
View Full Paper