<p><span style="color: rgb(48, 49, 51); font-family: " helvetica="" pingfang="" hiragino="" sans="" microsoft="" font-size:="" text-wrap-mode:="" background-color:="">An Intelligent Clinical Decision Support System for Cardiovascular Dise

<p>Coronary Artery Disease</p><p>(CAD) is diagnosed with the help of Cardiac Magnetic Resonance Imaging (CMRI),</p><p>which relies on the analysis of the heart images to detect the occurrence of</p><p>blockage in the coronary arteries to help make a good diagnosis and further</p><p>treatment. This research work developed an Intelligent Clinical Support System</p><p>(ICSS) for CAD diagnosis that integrates an Explainable Artificial Intelligence</p><p>(XAI) with Deep Learning (DL). The proposed model implies CMRI for diagnosis</p><p>and Modified Gradient-Weighted Class Activation Mapping (MG-WCAM) for</p><p>segmenting cardiac images. The Shapley Additive Explanations (SHAP) for Feature</p><p>Selection (FS) then accepts the segmented images, which are required to</p><p>simplify further prediction. The FS acts as input for the Transformer-aided</p><p>Deep Neural Network (T-DNN), which is used for classification. The T-DNN learns</p><p>contextual relations, which, in turn, increases the system’s overall</p><p>efficiency. The proposed system’s segmentation mechanism is evaluated using the</p><p>Dice Similarity Coefficient (DSC) and Pixel Accuracy (PA). The DSC of the model</p><p>achieved 0.93, demonstrating a good overlap between the predicted and the</p><p>ground-truth segmented regions. The achieved classification accuracy is 88%,</p><p>surpassing the state-of-the-art techniques. These results demonstrate that</p><p>combining XAI with DL is accurate and explainable for CAD diagnosis and</p><p>treatment procedures, thereby automating and simplifying clinical measures. </p><p><br></p>