Advanced CNN and Explainable AI Based Architecture for Interpretable Brain MRI Analysis

Convolutional Neural Networks (CNNs) serve as a foundational component in the domain of Computer Vision (CV). In order to enhance the Interpretability of CNN models, a critical aspect for clinical adoption, this study incorporates Explainable AI (XAI) methodologies. Through applying CNNs and XAI to a dataset comprising 5285 Brain MRI Images, a classification accuracy of 86% was achieved. The LIME framework was employed to generate localized explanations, thereby augmenting the model’s transparency and facilitating a deeper understanding of its decision-making process. This research explores the potential of synergistically integrating deep learning and XAI to foster the development of more reliable and comprehensible medical image analysis systems. Such systems hold the promise of improving diagnostic accuracy and clinical decision-making by providing healthcare professionals with transparent and explainable insights into the model’s predictions, ultimately leading to more informed and effective patient care.