Cloud-Enabled ECG Data Analysis for Heart Disease Prediction

Abstract

Cardiovascular diseases (CVDs) are proliferating, necessitating the development of scalable and efficient systems capable of doing real-time electrocardiogram (ECG) analysis. This study introduces a cloud-based architecture utilizing machine learning approaches for precise arrhythmia detection. ECG signals are acquired from the MIT BIH dataset, undergo preprocessing to remove noise, and are then securely stored in Amazon S3. The cloud architecture offers adaptable storage, rapid data retrieval, and effortless integration for machine learning model training. Upon processing, the ECG data is extracted and examined utilizing deep learning frameworks, including Convolutional Neural Networks (CNNs) and Long Short-Term Memory (LSTM) networks, for categorization and illness forecasting. This method facilitates real-time surveillance, alleviates the computing load on local devices, and guarantees safe data management. Experimental assessment reveals a classification accuracy of 99.5%, underscoring the efficacy of the proposed system for arrhythmia diagnosis and remote cardiac monitoring in clinical settings.