A Systematic Literature Review of Machine Learning and Explainable AI Approaches for Diabetes Prediction: Taxonomy, Gaps, and Future Directions (2020–2025)

References

[1] International Diabetes Federation, IDF Diabetes Atlas, 10th ed. Brussels, Belgium: IDF, 2021.

[2] World Health Organization, Global Report on Diabetes. Geneva, Switzerland: WHO, 2016.

[3] Z. Obermeyer and E. J. Emanuel, 'Predicting the Future — Big Data, Machine Learning, and Clinical Medicine,' N. Engl. J. Med., vol. 375, no. 13, pp. 1216-1219, 2016.

[4] R. D. Joshi and C. K. Dhakal, 'Predicting type 2 diabetes using logistic regression and machine learning approaches,' Int. J. Environ. Res. Public Health, vol. 18, no. 14, p. 7346, 2021.

[5] I. Tasin, T. U. Nabil, S. Islam, and R. Khan, 'Diabetes prediction using machine learning and explainable AI techniques,' Healthc. Technol. Lett., vol. 10, no. 1-2, pp. 1-10, 2023.

[6] M. O. Olusanya et al., 'Accuracy of machine learning classification models for prediction of type 2 diabetes,' Int. J. En viron. Res. Public Health, vol. 19, no. 21, p. 14280, 2022.

[7] E. Chowdhury et al., 'Risk prediction of cardiovascular disease for diabetic patients using ML,' arXiv:2511.04971, 2025.

[8] M. Ali et al., 'Stacking classifier with RF meta classifier for diabetes classification,' Procedia Comput. Sci., vol. 207, p. 3459, 2022.

[9] S. Naeem et al., 'Diabetes mellitus prediction using KNN and Naive Bayes,' J. Healthc. Eng., 2023.

[10] 'RF with feature selection and XAI for diabetes,' Front. Artif. Intell., 2024.

[11] S. Naaz et al., 'Comparative analysis of ML algorithms for diabetes prediction,' IEEE Access, 2025.

[12] 'A two-level stacking classifier for diabetes,' Comput. Biol. Med., 2024.

[13] U. Allani, 'Interactive diabetes risk prediction using explainable ML: a Dash-based approach,' arXiv:2505.05683, 2025.

[14] M. K. Hasan et al., 'Diabetes prediction using ensembling of different ML classifiers,' IEEE Access, vol. 8, pp. 76516-76531, 2020.

[15] A. Young et al., 'Quantum-inspired stacking model for diabetes classification,' Quant. Comput. Healthc., 2025.

[16] 'Generalizable diabetes prediction pipeline,' J. Med. Inform., 2025.

[17] 'Hybrid SVM and logistic regression for diabetes,' IEEE Trans. Biomed. Eng., 2025.

[18] O. S. Isiaka et al., 'Enhancing personalized treatment in diabetes using genomic data and deep learning,' J. Comput. Sci. Appl., vol. 31, no. 2, pp. 52-71, 2024.

[19] 'Cluster-personalised explainable AI for diabetes,' Comput. Methods Programs Biomed., 2025.

[20] 'Feature selection with Boruta and SHAP for diabetes,' Expert Syst. Appl., 2025.

[21] 'Analyzing classification algorithms with SHAP/LIME for diabetes,' Front. Digit. Health, 2024.

[22] 'Predicting early stage diabetes using RF and SHAP,' Inform. Med. Unlocked, 2021.

[23] 'Ensemble deep learning for diabetes prediction,' Appl. Soft Comput., 2024.

[24] 'Temporal biomarker analysis for diabetes using BiLSTM,' IEEE J. Biomed. Health Inform., 2025.

[25] 'GAN augmented CNN-LSTM for diabetes,' Pattern Recognit., 2024.

[26] 'XGBoost and deep learning for cardiovascular risk in BRFSS,' PLOS ONE, 2025.

[27] 'Nested cross-validation stacking for transparent diabetes prediction,' Artif. Intell. Med., 2025.

[28] 'Towards transparent ML for diabetes: multi-XAI empirical analysis,' J. Biomed. Inform., 2025.

[29] 'XGBoost and DNN hybrid for early stage diabetes,' Comput. Biol. Med., 2025.

[30] 'AMNN and KAN models for diabetes classification,' Neural Netw., 2025.

[31] 'Comparative analysis of 8 ML algorithms for diabetes survey dataset,' Health Inform. J., 2025.

[32] 'Quantum-inspired GAN data augmentation for imbalanced diabetes data,' IEEE Access, 2025.

[33] 'Cascaded ensemble classification for diabetes using Pima dataset,' Knowl.-Based Syst., 2023.

[34] S. M. Ganie and M. B. Malik, 'An ensemble machine learning approach for diabetes prediction,' Int. J. Inf. Technol., 202 2.

[35] K. Kumar et al., 'Random forest algorithm for the prediction of diabetes,' in Proc. ICSCAN, 2021.

[36] S. M. Lundberg and S.-I. Lee, 'A unified approach to interpreting model predictions,' in Adv. Neural Inf. Process. Syst., vol. 30, pp. 4765-4774, 2017.

[37] M. T. Ribeiro, S. Singh, and C. Guestrin, 'Why should I trust you?: Explaining the predictions of any classifier,' in Proc. KDD, pp. 1135- 1144, 2016.

[38] H. He, Y. Bai, E. A. Garcia, and S. Li, 'ADASYN: Adaptive synthetic sampling for imbalanced learning,' in Proc. IJCNN, pp. 1322- 1328, 2008.

[39] J. W. Smith et al., 'Using the ADAP learning algorithm to forecast the onset of diabetes mellitus,' in Proc. SCAMC, pp. 261-265, 1988.

[40] Centers for Disease Control and Prevention, Behavioral Risk Factor Surveillance System Survey Data. Atlanta: CDC, 2015.

@article{1718378,
      author = {Smarika Singh, Prof. (Dr.) Ritu Sindhu, Dr. Shivani Sharma},
      title = {A Systematic Literature Review of Machine Learning and Explainable AI Approaches for Diabetes Prediction: Taxonomy, Gaps, and Future Directions (2020–2025)},
      journal = {Iconic Research And Engineering Journals},
      year = {2026},
      volume = {9},
      number = {11},
      pages = {4378-4387},
      issn = {2456-8880},
      url = {https://www.irejournals.com/formatedpaper/1718378.pdf},
      abstract = {Diabetes mellitus is a rapidly escalating global public health crisis affecting over 537 million adults worldwide, with projections suggesting 783 million by 2045. The exponential growth of health data repositories and machine learning research has produced a vast body of literature on automated diabetes risk prediction. However, this literature suffers from recurring methodological inconsistencies — particularly over-reliance on a single benchmark dataset, data leakage from improper SMOTE application, absence of threshold optimisation for real-world imbalanced distributions, and shallow explainability that hinders clinical adoption. This paper presents a systematic literature review of 34 studies published between 2020 and 2025, covering classical machine learning, ensemble methods, deep learning architectures, and explainable AI frameworks applied to diabetes prediction. We propose a six-category taxonomy of approaches, analyse comparative performance across datasets, and identify eight specific research gaps that remain unaddressed in the existing literature. We also present our own empirical work — a SMOTE-enhanced hybrid RF+MLP stacking framework validated across Pima Indian and BRFSS 2015 datasets achieving consistent AUC of 0.808 and 0.816 respectively with 90.7% clinical recall — as a representative study addressing several identified gaps. Based on the review, we outline seven concrete future research directions essential for bridging the gap between research prototypes and clinically deployable diabetes screening tools.},
      keywords = {Diabetes Prediction, Machine Learning, Ensemble Methods, Explainable AI, SHAP, LIME, SMOTE, Cross-Dataset Validation, Systematic Review, Random Forest, Xgboost, Deep Learning},
      month = {May},
      doi = {https://doi.org/10.64388/IREV9I11-1718378}
  }