Impact of HPC on Multi-Modal Data Integration for Heart Failure Prediction in Clinical Decision Support Systems

Introduction: High-performance computing (HPC) integrated CDSS has played a vital role in making healthcare decisions using the latest computational techniques. This study emphasizes the impact of HPC-enhanced methods on decision support systems (DSS) in healthcare informatics. It focuses on improvements in accuracy, reliability, and usability, and their effects on patient outcomes and healthcare delivery efficiency. Objectives: Our study aims at designing and implementing an HPC-integrated CDSS framework for the prediction of heart failure. Thus, it can be used for concatenating multimodal data and examining the impact of HPC-integrated CDSS on various types of healthcare informatics, such as EHR, ECG, and medical imaging. In order to improve predictive accuracy, reliability, and operational efficiency with ensemble learning methods. Methods: The study integrates multimodal data, including Electronic Health Records (EHR), ECG signals, and medical imaging, with advanced machine learning techniques. The proposed ensemble model is benchmarked against traditional approaches like 3D-CNN, RNN, VGGNet, and YOLOv5. The performance of the model was evaluated on several criteria, such as accuracy, precision, sensitivity, and specificity. ROC curve analyses were conducted to examine the performance improvements of the model-based fusion approach over concatenation fusion. Results: The proposed ensemble model demonstrated excellent outcomes over traditional approaches. Also, it shows superior accuracy, precision, sensitivity, FPR, FNR, and F1 measure.ROC curve analyses demonstrated that the model-based fusion approach significantly enhanced predictive precision as well as clinical decision-making. Conclusions: This study provides empirical evidence on the benefits of HPC in healthcare decision-making. HPC-integrated CDSS impacts the quality of patient care outcomes and enhances the efficiency of healthcare delivery.