Explainable AI for Disease Diagnosis Using Machine Learning

Author

Samira Farooqui, Mrs. Deepti Gupta

Abstract

 A model's predictive performance is not enough to make sure whether machine learning models are appropriate for diagnosis. This work provides a simple leakage-free pipeline for binary classification and subsequent explanation for the Wisconsin Diagnostic Breast Cancer (WDBC) dataset. It is tested on the WDBC dataset. In this paper, we compare four classifiers, which include logistic regression, support vector machine (SVM), k-nearest neighbors, and random forest. To ensure that the experiment is conducted honestly, stratification is used when splitting the data, and all of the preprocessing techniques are applied only to the training set to avoid any leakage of information from the test set. The models are chosen based on the results of the five-fold cross-validation and tested once on the independent test data. The SVM achieved the highest accuracy (97.37%) and F1-score (96.30%), while logistic regression achieved the highest ROC-AUC (99.60%). SHAP and permutation importance revealed that the models are mostly dependent on shape and texture features of tumors, including the worst texture, concavity, and radius.

Keywords

Explainable AI, Disease Diagnosis, Machine Learning, SHAP, Breast Cancer

DOI : https://doi.org/10.5281/zenodo.23610108

Full Text:

Download Paper PDF

References

[1] T. Khater, A. Hussain, R. Bendar­daf, I. M. Talaat, H. Tawfik, S. Ansari, and S. Mahmoud, “An Explainable Artificial Intelligence Model for the Classification of Breast Cancer,” IEEE Access, vol. 13, pp. 5618–5630, 2025, doi: 10.1109/ACCESS.2023.3308446.

[2] S. Bai, S. Nasir, R. A. Khan, A. Meyer, and H. Konik, “Breast Cancer Diagnosis With Explainable Artificial Intelligence (XAI): Uncovering Strengths and Biases,” IEEE Access, vol. 13, pp. 205197–205223, 2025, doi: 10.1109/ACCESS.2025.3639184.

[3] A. Basaad, S. Basurra, E. Vakaj, M. Aleskandarany, and M. M. Abdelsamea, “GraphX-Net: A Graph Neural Network-Based Shapley Values for Predicting Breast Cancer Occurrence,” IEEE Access, vol. 12, pp. 93993–94008, 2024, doi: 10.1109/ACCESS.2024.3424526.

[4] T. Shahzad, M. Saleem, M. S. Farooq, S. Abbas, M. A. Khan, and K. Ouahada, “Developing a Transparent Diagnosis Model for Diabetic Retinopathy Using Explainable AI,” IEEE Access, vol. 12, pp. 149700–149716, 2024, doi: 10.1109/ACCESS.2024.3475550.

[5] K. R. Makkena and K. Natarajan, “Enhancing Liver Cirrhosis Diagnosis Using Machine Learning With Explainable AI and Cross-Validated Hyperparameter Tuning Techniques,” IEEE Access, vol. 13, pp. 93323–93336, 2025, doi: 10.1109/ACCESS.2025.3568918.

[6] O. Jyoti, H. B. Kibria, Z. T. Pear, M. Nahiduzzaman, M. F. Ahamed, K. R. Islam, J. Kumar, and M. E. H. Chowdhury, “A Clinically Interpretable Approach for Early Detection of Autism Using Machine Learning With Explainable AI,” IEEE Access, vol. 13, pp. 121512–121531, 2025, doi: 10.1109/ACCESS.2025.3586314.

[7] M. Naz, A. Khalid, A. Hameed, R. Taj, W. Mumtaz, F. A. Alotaibi, and M. M. Alnfiai, “Meta-Ensemble Learning for Heart Disease Prediction: A Stacking-Based Approach With Explainable AI,” IEEE Access, vol. 13, pp. 137271–137290, 2025, doi: 10.1109/ACCESS.2025.3588683.

[8] M. Rashed-Al-Mahfuz, A. Haque, A. K. M. Azad, S. A. Alyami, J. M. W. Quinn, and M. A. Moni, “Clinically Applicable Machine Learning Approaches to Identify Attributes of Chronic Kidney Disease (CKD) for Use in Low-Cost Diagnostic Screening,” IEEE Journal of Translational Engineering in Health and Medicine, vol. 9, pp. 4900511, 2021, doi: 10.1109/JTEHM.2021.3073629.

[9] D. Muhammad, I. Ahmed, M. O. Ahmad, and M. Bendechache, “Randomized Explainable Machine Learning Models for Efficient Medical Diagnosis,” IEEE Journal of Biomedical and Health Informatics, vol. 29, no. 9, pp. 6474–6486, 2025, doi: 10.1109/JBHI.2024.3491593.

[10] H. Younis, H. A. Rehman, and R. Minghim, “Somatic Mutation-Driven Cancer Stage Classification via Autoencoder Clustering and Explainable Deep Learning,” IEEE Access, vol. 14, pp. 45213–45226, 2026, doi: 10.1109/ACCESS.2026.3676284.

[11] S. M. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” in Proc. 31st Conf. Neural Information Processing Systems (NeurIPS), 2017, pp. 4765–4774.

[12] M. T. Ribeiro, S. Singh, and C. Guestrin, “Why Should I Trust You? Explaining the Predictions of Any Classifier,” in Proc. 22nd ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, 2016, pp. 1135–1144.

[13] F. Pedregosa et al., “Scikit-learn: Machine Learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.

 

 

Share your valuable work from Social Media Buttons