Generative Artificial Intelligence for Diabetes Mellitus Prediction and Intelligent Clinical Decision Support

Sangeeta Lalwani — 2026-09-10

Diabetes mellitus is a major chronic metabolic disorder characterized by persistent dysregulation of blood glucose and progressive cardiovascular, renal, neurological, and ophthalmic complications. The global prevalence of diabetes has increased substantially, reaching approximately 14% among adults in 2022, with nearly 830 million people living with the disease. The rapid expansion of electronic health records, continuous glucose monitoring, laboratory databases, medical imaging, wearable technologies, and patient-generated health data has created opportunities for artificial intelligence (AI)-based diabetes prediction and clinical decision support. Conventional machine learning models demonstrate potential for diabetes detection and complication prediction; however, challenges remain regarding heterogeneous datasets, external validation, calibration, dataset shift, interpretability, and clinical implementation. This study proposes a systematic framework integrating generative AI, particularly large language models (LLMs), with validated predictive models for diabetes risk assessment and intelligent clinical decision support. A secondary evidence-synthesis approach was used, integrating findings from systematic reviews, clinical trials, diabetes technology research, and recent LLM literature. Evidence indicates that only 36% of evaluated machine learning models for diabetes complications demonstrated clearly useful discrimination. Generative AI has shown promise in treatment support, although performance decreases with increasing clinical complexity. ChatGPT-4 achieved 69.2–84.6% agreement with clinician prescribing in selected monotherapy cases but only 47.1–49.8% for dual- or triple-therapy cases. The findings support using generative AI as an orchestration layer for prediction, evidence retrieval, contextualization, explanation, documentation, and workflow support rather than as a replacement for validated prediction or clinical judgment. Safe implementation requires calibrated models, authoritative retrieval, uncertainty representation, auditability, privacy, fairness assessment, human oversight, and prospective clinical validation.

Safety and Efficacy of Finerenone in Chronic Kidney Disease (CKD) Patients

Mukherjee Nilanjan — 2026-10-07

Background: Finerenone is a selective non-steroidal mineralocorticoid receptor antagonist with established kidney and cardiovascular benefits in chronic kidney disease (CKD) associated with type 2 diabetes. Short-term real-world data from broader CKD populations, including non-diabetic CKD, remain less extensive. We evaluated renal responses and safety during routine nephrology care.

Methods: We conducted a prospective, observational, single-centre study in adults aged 18–75 years with CKD stages 1–4 and significant albuminuria/proteinuria. Twenty-eight eligible participants receiving or initiated on finerenone were followed at baseline and at 1, 3 and 6 months. Finerenone was used at 10–20 mg once daily with dose adjustment according to eGFR and serum potassium. Primary outcomes were longitudinal changes in eGFR and urine albumin-to-creatinine ratio (uACR); treatment-emergent adverse events were the principal safety outcome. Descriptive subgroup analyses compared diabetic and non-diabetic CKD.

Results: All 28 participants had recorded renal measurements at 1, 3 and 6 months. Mean eGFR increased from 45.5 ± 25.1 mL/min/1.73 m² at baseline to 54.8 ± 24.2 at 1 month, 55.9 ± 24.2 at 3 months and 55.4 ± 24.6 at 6 months; the mean 6-month change was 9.9 mL/min/1.73 m² (95% CI 5.7–14.2; p < 0.001). Median uACR decreased from 1040.0 mg/g (IQR 619.8–1485.8) to 313.5 mg/g (IQR 137.0–387.2), a 69.9% reduction in cohort medians. At 6 months, 25 participants (89.3%) had an increase in eGFR from baseline, and 19 (67.9%) had a patient-level uACR reduction of at least 30% in an exploratory analysis. Treatment-emergent adverse events were recorded in 18 participants (64.3%); hyperkalemia was recorded in 7 (25.0%), and 5 (17.9%) had maximum follow-up potassium ?5.5 mmol/L. Hypotension occurred in 4 (14.3%) and renal deterioration/mild eGFR decline in 2 (7.1%). No cardiovascular composite event or hospitalization was documented.

Conclusions: In this small, uncontrolled observational cohort, finerenone exposure was associated with progressive reduction in albuminuria and an observed increase in mean eGFR over 6 months. Hyperkalemia and haemodynamic events remained clinically relevant and support structured monitoring. The findings are hypothesis-generating and cannot establish causality or long-term cardiorenal benefit.

International Journal of Pharmaceutics and Drug Analysis

Generative Artificial Intelligence for Diabetes Mellitus Prediction and Intelligent Clinical Decision Support

Safety and Efficacy of Finerenone in Chronic Kidney Disease (CKD) Patients