"Towards Explainable and Deployable AI for Personalized Diabetes Prediction: A Pipeline with SMOTE, Feature Selection, and SHAP-Based Model Interpretation"
Abstract
The growing global burden of diabetes needs improved predictive analytics for tailored risk assessment and prompt intervention. While Artificial Intelligence (AI) and Machine Learning (ML) show great potential in this field, their "black box" aspect sometimes stifles clinical acceptance and confidence. This study describes a thorough and explainable AI pipeline for personalized diabetes prediction, including key steps to improve model performance, interpretability, and deployability.
The suggested methodology uses the Synthetic Minority Oversampling Technique (SMOTE) to solve the intrinsic class imbalance in medical datasets, boosting model resilience and predictive accuracy. Robust feature selection approaches are then used to identify the most important physiological and clinical predictors, resulting in more concise and interpretable models.
Importantly, the pipeline includes SHAP (Shapley Additive Explanations) for post-hoc model interpretation, which provides visible, detailed insights into how various characteristics contribute to each prediction. This explainable approach promotes clinician comprehension and patient engagement, allowing for more informed decision-making and individualized therapeutic options. The goal of combining these strategies is to create a framework that not only achieves excellent prediction performance but also overcomes interpretability issues, opening the path for the ethical and practical deployment of AI technologies in customized diabetes management.
References
Abhari, S., Kalhori, S. R. N., Ebrahimi, M., Hasannejadasl, H., & Garavand, A. (2019). Artificial intelligence applications in type 2 diabetes mellitus care: Focus on machine learning methods. Healthcare Informatics Research, 25(4), 248–261.
Ahmed, A., Aziz, S., Abd Alrazaq, A., Farooq, F., & Sheikh, J. (2022). Overview of artificial intelligence driven wearable devices for diabetes: Scoping review. Journal of Medical Internet Research, 24(8), e36010.
Ahmed, M. G., Zaghloul, I., & Abd El Fattah, M. T. (2023). A survey on personalization of diabetes treatment using artificial intelligence techniques. Benha Journal of Applied Sciences, 8(5), 229–236.
Broome, D. T., Hilton, C. B., & Mehta, N. (2020). Policy implications of artificial intelligence and machine learning in diabetes management. Current Diabetes Reports, 20, 1–5.
Campanella, S., Paragliola, G., Cherubini, V., Pierleoni, P., & Palma, L. (2024). Towards personalized AI based diabetes therapy: A review. IEEE Journal of Biomedical and Health Informatics.
Chaki, J., Ganesh, S. T., Cidham, S. K., & Theertan, S. A. (2022). Machine learning and artificial intelligence based diabetes mellitus detection and self-management: A systematic review. Journal of King Saud University – Computer and Information Sciences, 34(6), 3204–3225.
Chang, V., Bailey, J., Xu, Q. A., & Sun, Z. (2023). Pima Indians diabetes mellitus classification based on machine learning (ML) algorithms. Neural Computing and Applications, 35(22), 16157–16173.
Dankwa Mullan, I., Rivo, M., Sepulveda, M., Park, Y., Snowdon, J., & Rhee, K. (2019). Transforming diabetes care through artificial intelligence: The future is here. Population Health Management, 22(3), 229–242.
Gautier, T., Ziegler, L. B., Gerber, M. S., Campos Náñez, E., & Patek, S. D. (2021). Artificial intelligence and diabetes technology: A review. Metabolism, 124, 154872.
Guan, Z., Li, H., Liu, R., Cai, C., Liu, Y., Li, J., Wang, X., Huang, S., Wu, L., Liu, D., & Yu, S. (2023). Artificial intelligence in diabetes management: Advancements, opportunities, and challenges. Cell Reports Medicine, 3(9), 100852. https://doi.org/10.1016/j.xcrm.2022.100852
Guo, Q., Fu, B., Tian, Y., Xu, S., & Meng, X. (2024). Recent progress in artificial intelligence and machine learning for novel diabetes mellitus medications development. Current Medical Research and Opinion, 40(9), 1483–1493.
Iftikhar, M., Saqib, M., Qayyum, S. N., Asmat, R., Mumtaz, H., Rehan, M., Ullah, I., Ud Din, I., Noori, S., Khan, M., & Rehman, E. (2024). Artificial intelligence driven transformations in diabetes care: A comprehensive literature review. Annals of Medicine and Surgery, 86(9), 5334–5342.
Li, J., Huang, J., Zheng, L., & Li, X. (2020). Application of artificial intelligence in diabetes education and management: Present status and promising prospect. Frontiers in Public Health, 8, 173.
Lyngdoh, A. C., Choudhury, N. A., & Moulik, S. (2021, March). Diabetes disease prediction using machine learning algorithms. In 2020 IEEE EMBS Conference on Biomedical Engineering and Sciences (IECBES) (pp. 517–521). IEEE.
Makino, M., Yoshimoto, R., Ono, M., Itoko, T., Katsuki, T., Koseki, A., Kudo, M., Haida, K., Kuroda, J., Yanagiya, R., & Saitoh, E. (2019). Artificial intelligence predicts the progression of diabetic kidney disease using big data machine learning. Scientific Reports, 9(1), 11862.
Makroum, M. A., Adda, M., Bouzouane, A., & Ibrahim, H. (2022). Machine learning and smart devices for diabetes management: Systematic review. Sensors, 22(5), 1843.
Nomura, A., Noguchi, M., Kometani, M., Furukawa, K., & Yoneda, T. (2021). Artificial intelligence in current diabetes management and prediction. Current Diabetes Reports, 21(12), 61.
Papazafiropoulou, A. K. (2024). Diabetes management in the era of artificial intelligence. Archives of Medical Sciences. Atherosclerotic Diseases, 9, e122.
Patil, A. R., Mane, S. C., Patil, M. A., Gangurde, N. A., Rahate, P. G., & Dhanke, J. A. (2024). Artificial intelligence and machine learning techniques for diabetes healthcare: A review. Journal of Chemical Health Risks, 1058–1063.
Preetha, S., Chandan, N., Darshan, N. K., & Gowrav, P. B. (2020). Diabetes disease prediction using machine learning. International Journal of Modern Trends in Engineering Research, 6, 37–43.
Rashid, M. M., Askari, M. R., Chen, C., Liang, Y., Shu, K., & Cinar, A. (2022). Artificial intelligence algorithms for treatment of diabetes. Algorithms, 15(9), 299.
Rigla, M., García Sáez, G., Pons, B., & Hernando, M. E. (2018). Artificial intelligence methodologies and their application to diabetes. Journal of Diabetes Science and Technology, 12(2), 303–310.
Sarwar, A., & Sharma, V. (2014). Comparative analysis of machine learning techniques in prognosis of type II diabetes. AI & Society, 29, 123–129.
Sharma, A., Prasad, K. D. V., Chakrasali, S. V., Gowda, D., Kumar, C., Chaturvedi, A., & Pazhani, A. A. J. (2023). Computer vision based healthcare system for identification of diabetes & its types using AI. Measurement: Sensors, 27, 100751.
Sheng, B., Pushpanathan, K., Guan, Z., Lim, Q. H., Lim, Z. W., Yew, S. M. E., Goh, J. H. L., Bee, Y. M., Sabanayagam, C., Sevdalis, N., & Lim, C. C. (2024). Artificial intelligence for diabetes care: Current and future prospects. The Lancet Diabetes & Endocrinology, 12(8), 569–595. https://doi.org/10.1016/S2213-8587(24)00123-4
Sowah, R. A., Bampoe Addo, A. A., Armoo, S. K., Saalia, F. K., Gatsi, F., & Sarkodie Mensah, B. (2020). Design and development of diabetes management system using machine learning. International Journal of Telemedicine and Applications, 2020(1), 8870141.
Tasin, I., Nabil, T. U., Islam, S., & Khan, R. (2023). Diabetes prediction using machine learning and explainable AI techniques. Healthcare Technology Letters, 10(1–2), 1–10.
Tuppad, A., & Patil, S. D. (2022). Machine learning for diabetes clinical decision support: A review. Advances in Computational Intelligence, 2(2), 22.
Vettoretti, M., Cappon, G., Facchinetti, A., & Sparacino, G. (2020). Advanced diabetes management using artificial intelligence and continuous glucose monitoring sensors. Sensors, 20(14), 3870.
Zhuhadar, L. P., & Lytras, M. D. (2023). The application of AutoML techniques in diabetes diagnosis: Current approaches, performance, and future directions. Sustainability, 15(18), 13484.
Wang, S., Zhang, L., & He, H. (2021). Adaptive dynamic programming for diabetes optimal control: A review. IEEE Transactions on Cybernetics, 51(1), 387–398.
Wang, Y., Gao, H., & Zhang, X. (2020). PPG signal-based blood glucose estimation using deep learning models. Sensors, 20(3), 887.
Wang, Y., Zhang, X., Guan, Z., Liu, R., & Wang, H. (2021). Recent advances in AI-based diabetes management. Current Diabetes Reports, 21, 10.
Wen, D., Zeng, W., Liu, X., Hu, Z., Song, B., Yin, K., & Zhang, X. (2022). An interpretable diabetes prediction model based on machine learning algorithms. BMC Medical Informatics and Decision Making, 22(1), 218.
Wu, X., Liu, X., & Zeng, Q. (2022). Artificial intelligence for diabetes management and decision support. Frontiers in Public Health, 10, 895658.
Xu, X., Luo, Y., & Sun, B. (2021). Early diabetes prediction using XGBoost classifier with feature selection. International Journal of Environmental Research and Public Health, 18(13), 7146.
Yang, L., Zhang, Y., Liu, J., Li, Y., & Li, Z. (2023). Personalized diabetes management using deep reinforcement learning. IEEE Transactions on Neural Networks and Learning Systems, 34(3), 1559–1570.
Yin, Z., Zhao, M., & He, Q. (2021). Explainable AI for diabetes risk prediction: Performance comparison and explainability analysis. Artificial Intelligence in Medicine, 119, 102147.
Yu, D., Lee, T. M. C., & Lee, J. (2023). Artificial intelligence in diabetes care: A review of current developments and future prospects. The Lancet Digital Health, 5(3), e170–e180. https://doi.org/10.1016/S2589-7500(22)00240-2
Zhang, D., Guo, X., & Wu, J. (2020). Deep learning for personalized glucose forecasting in diabetes management. IEEE Journal of Biomedical and Health Informatics, 24(5), 1249–1260.
Zhang, Z., Wang, Q., & Zhang, Y. (2022). AI-based personalized insulin dosing system for diabetes management. IEEE Transactions on Biomedical Engineering, 69(9), 2812–2821.
Zhou, T., Du, W., Li, Y., Gao, L., Sun, Y., Huang, X., & Liu, J. (2022). Machine learning in diabetes management: Current status and future prospects. Frontiers in Public Health, 10, 930404.
Rong, G., Mendez, A., Assi, E. B., Zhao, B., & Sawan, M. (2021). Artificial intelligence in healthcare: Review and prediction case studies. Engineering, 7(5), 672–685. https://doi.org/10.1016/j.eng.2021.02.011
Yin, Z., Zhao, M., & He, Q. (2021). Explainable AI for diabetes risk prediction: Performance comparison and explainability analysis. Artificial Intelligence in Medicine, 119, 102147.
Rong, G., Mendez, A., Assi, E. B., Zhao, B., & Sawan, M. (2021). Artificial intelligence in healthcare: Review and prediction case studies. Engineering, 7(5), 672–685. https://doi.org/10.1016/j.eng.2021.02.011
Zhang, Y., Zhang, Y., Wang, L., Yin, Z., & Yin, Y. (2020). Machine learning approaches for diabetes prediction using electronic health records: A systematic review. BMC Medical Informatics and Decision Making, 20(1), 247.
Yang, L., Zhang, Y., Liu, J., Li, Y., & Li, Z. (2023). Personalized diabetes management using deep reinforcement learning. IEEE Transactions on Neural Networks and Learning Systems, 34(3), 1559–1570.
Guan, Z., Li, H., Liu, R., Cai, C., Liu, Y., Li, J., Wang, X., Huang, S., Wu, L., Liu, D., & Yu, S. (2023). Artificial intelligence in diabetes management: Advancements, opportunities, and challenges. Cell Reports Medicine, 3(9), 100852. https://doi.org/10.1016/j.xcrm.2022.100852
Sheng, B., Pushpanathan, K., Guan, Z., Lim, Q. H., Lim, Z. W., Yew, S. M. E., Goh, J. H. L., Bee, Y. M., Sabanayagam, C., Sevdalis, N., & Lim, C. C. (2024). Artificial intelligence for diabetes care: Current and future prospects. The Lancet Diabetes & Endocrinology, 12(8), 569–595. https://doi.org/10.1016/S2213-8587(24)00123-4
