An Approach for XAI Visualizations for Explainability of Alzheimer’s Detection

In the field of medical image analysis, the integration of machine learning (ML) models has significantly enhanced the process of identifying and diagnosing Alzheimer’s disease. However, the opacity of these models, often termed as "black boxes," raises concerns regarding their interpretability and reliability in clinical settings. Explainable artificial intelligence (XAI) techniques have emerged as crucial tools to address these challenges by clarifying the reasoning behind model outputs. This paper offers an extensive review of literature concerning the employment of Explainable Artificial Intelligence (XAI) methodologies, such as SHAP, LIME, GradCAM, DeepLIFT, Saliency Maps, and LRP to augment the comprehensibility of ML models applied in the detection of Alzheimer’s disease.