Modeling an Intelligent Health Monitoring System for Atrial Fibrillation Detection

This article is a research paper that presents the development and evaluation of an Intelligent Health Monitoring System (IHMS) designed for the detection of Atrial Fibrillation (AFib) using one-dimensional Convolutional Neural Networks (1D-CNNs). The study highlights the increasing prevalence of AFib and the necessity for continuous monitoring due to its potential to occur without symptoms. The paper details the design and training of three 1D-CNN models, assessing their performance in terms of inference efficiency on resource-constrained edge devices compared to cloud-based architectures. The authors conduct experiments to measure end-to-end delay and prediction time, demonstrating the feasibility of on-device AFib detection. The findings suggest that the proposed system can effectively operate in low-cost environments, providing insights into the selection of suitable architectures for embedded deployment. The paper is structured to include related work, system architecture, performance evaluation strategy, and results discussion, concluding with future directions for research.