• Deep Vasoya
• Dr. Mohit Bhadla

Since diabetes mellitus is becoming more common in many populations, it has become a worldwide health concern. For early diagnosis and intervention, predictive models that are transparent and accurate are essential. Though their black-box character frequently restricts clinical application due to a lack of interpretability, traditional machine learning methods have shown substantial promise in diabetes prediction. By offering insights into model decisions, Explainable AI (XAI) tackles this issue and builds patient and healthcare professional trust. In this work, we investigate how to include XAI methods into diabetes prediction machine learning models while maintaining accuracy and transparency. Current research emphasizes the significance of interpretable models for individualized diagnosis and treatment planning by highlighting the role of genetic, metabolic, and behavioural risk factors in the onset of diabetes. Numerous researches have shown how model-specific interpretability techniques, SHAP values, and feature importance analysis can help close the gap between AI predictions and human comprehension. Additionally, improvements in explain ability and diagnostic accuracy have been made possible by developments in deep learning and hybrid prediction models. This study assesses different XAI methods used to forecast diabetes and examines how well they work to give doctors insightful explanations. The results open the door for AI-driven, patient-centred healthcare solutions by highlighting the significance of explainable models in actual clinical situations.

Machine Learning in Healthcare, Explainable AI (XAI), Diabetes Risk Factors, Diabetes Prediction, XGBoost for Medical Diagnosis

"Explainable AI for Predicting Diabetes using ML", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 4, page no.n171-n176, April-2025, Available :http://www.jetir.org/papers/JETIR2504D21.pdf

"Explainable AI for Predicting Diabetes using ML", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 4, page no. ppn171-n176, April-2025, Available at : http://www.jetir.org/papers/JETIR2504D21.pdf