Improving Cardiac Rhythm Identification via Neural Signal Categorization Integrated With Immersive Threedimensional Cardiac Visualization

Cardiovascular diseases (CVDs) remain a predominant cause of global mortality, underscoring the critical necessity for prompt and precise diagnosis. Electrocardiography (ECG) is an important diagnostic tool for finding and keeping an eye on different heart problems by looking at the heart's electrical activity. However, manually identifying ECG features and categorising heartbeats is a difficult and time-consuming task that requires a lot of skill. To solve this problem, we made ArythmiAR, a new system that combines Convolutional Neural Networks (CNN) with Augmented Reality (AR) to allow for interactive diagnosis with 3D visualisation and real-time interaction. ArythmiAR has a number of important new features, such as deep learning-based ECG classification for accurate arrhythmia detection, 3D heart modelling and assembly for better visualisation, an AR interface for deploying CNN models, 3D localisation of heart sub-regions that cause arrhythmia anomalies, and better 3D visualisation and interaction. Our research examines diverse ECG classification methodologies, utilising data rebalancing techniques to improve model efficacy, particularly emphasising Multilayer Perceptron (MLP) and Convolutional Neural Network (CNN) models, which exhibited exceptionally competitive performance on the PhysioNet MIT-BIH Arrhythmia dataset, attaining an accuracy of 99.07% with the MLP model. This work also includes using the ECG classification deep learning model in an AR setting, which is a prototype for augmented rendering that lets users find, see, and interact with specific parts of the heart that cause arrhythmias. This platform gives doctors the tools they need to make better diagnoses and come up with better treatment plans, which improves the care of all patients.