Analysis of Privacy Enabled Asynchronous Federated Learning for Detecting Heart Disorders
Pawan Hegde · 2024
Heart disorders are a leading cause of mortality worldwide, necessitating efficient and accurate predictive models to assist in early diagnosis and treatment. Traditional machine learning models often rely on centralized data collection, which raises privacy concerns and data security issues. Federated Learning (FL) offers a solution by enabling collaborative model training across distributed devices without sharing raw data. However, the synchronous nature of FL can lead to increased training time and inefficiency, especially in large-scale distributed systems. To address this, an Asynchronous Federated Learning (AsynFL) model for heart disorder detection, with temporal weighted aggregation method is proposed. The proposed approach allows faster convergence and reduced training latency while maintaining privacy. Experimental results demonstrate that AsynFL outperforms traditional FL in terms of both training efficiency and accuracy, making it a promising solution for privacy-preserving heart disorder prediction in distributed healthcare environments.