Explainable & Trustworthy AI Frameworks for Healthcare with Blockchain Integration

Abstract

Artificial Intelligence (AI) is increasingly used in healthcare for disease prediction and clinical decision support; however, issues related to model transparency, trust, and data security limit its practical adoption. This research presents an integrated healthcare prediction framework combining Machine Learning, Explainable Artificial Intelligence (XAI), and Blockchain technology. Several machine learning models were implemented and evaluated on a healthcare dataset, including Logistic Regression, KNN, SVM, Gradient Boosting, and Random Forest. Based on experimental results, the Random Forest classifier achieved the best performance in terms of accuracy, reliability, and stability. Explainable AI techniques such as feature importance analysis, confusion matrix, ROC curve, and precision–recall curve were applied to improve interpretability and clinical understanding of model predictions. In addition, Blockchain technology was incorporated to ensure secure data storage, tamper-proof logging, access control, and auditability of healthcare data and AI outputs. The proposed framework enhances transparency, trust, and security, making it suitable for reliable and ethical healthcare decision support systems.

Ähnliche Arbeiten