Early Detection of Heart Blockages: The Power of AI and Feedforward Neural Networks

Cardiovascular disease remains one of the leading causes of death globally, and therefore early detection of arterial blockage is required for effective intervention. The work here proposes an automated diagnosis system based on a feedforward neural network (FNN) for diagnosing ECG signals in addition to angiographic images for early blockage detection. The dataset used in this work was collected using an AD8232 sensor to provide high fidelity in ECG signal acquisition for effective model training. Our approach tries to minimize reliance on lengthy human analysis while reducing the chances of diagnostic errors simultaneously. The proposed FNN is trained on a large database of diverse ECG recordings to identify subtle, clinically meaningful patterns that are generally difficult for experienced clinicians to identify. Experimental data show the network to be 95% sensitive and 97% accurate in all cases, which indicates its clinical worth. These measurements of performance were contrasted with routine diagnostic measurements in a controlled test, which highlighted the FNN's ability to identify true positive cases and reject false positives. The implementation of this neural network in clinical settings has the potential to shorten the time for diagnostic interventions as well as enhance access, especially in remote or low-resource areas. This innovation not only has the potential to improve diagnostic efficiency but also sets the stage for future innovation in precision medicine and prevention. In summary, the system developed here represents a major step forward in the use of artificial intelligence in cardiovascular diagnosis, providing a cost-effective, scalable solution.