A Enhancing Healthcare Outcomes with Explainable AI (XAI) for Disease Prediction: A Comprehensive Review

Abstract

Nowadays, diseases are caused by harmful ingredients such as origins, pathogens, microbes, microorganisms, and viruses that can enter the human body. Early prediction of diseases is crucial in saving human lives, and it remains one of the most challenging issues in clinical data analysis. The recent advancements in Artificial Intelligence (AI) have significantly improved the efficiency and accuracy of disease prediction. AI is a field of computational intelligence that encompasses various applications such as expert systems, Natural Language Processing (NLP) etc.. While some previous AI approaches are manual and prone to errors, Explainable Artificial Intelligence (XAI) forecast approaches allow for automatic identification and forecast, thereby minimizing detection faults compared to depending exclusively on human skill. This paper provides a comprehensive review of advanced approaches such as RF, VGG-16, SVM, SMOTE, ECNNs, CNN, NB, and XG Boost, all of which utilize XAI methods. Among these techniques, the Random Forest (RF) method with explainable techniques such as SMOTE and ADASYN achieved the highest accuracy of 98%. This finding highlights the potential of XAI in enhancing the accuracy and reliability of disease prediction models. By providing explanations for the predictions made by the model, XAI techniques increase transparency and trustworthiness, enabling medical professionals to better understand the reasoning behind the diagnosis