Remote Heart Diagnostic System

Abstract

For people in remote regions, limited access to specialists like cardiologists can turn treatable conditions into a serious threat. In this work, a heart diagnostic system capable of collecting and analyzing an electrocardiogram (ECG) signal in a remote location and sending the results to be reviewed by a cardiologist is designed and implemented. This system is controlled using a Raspberry Pi 4 microcomputer, with a physical subsystem designed to collect the ECG signal. This signal is utilized to diagnose certain heart conditions using an Artificial Intelligence (AI) based system.  A touchscreen mini-terminal guides the user through collecting and compiling patient data, sending this data to the AI agent for classification, and then storing and transferring the results to another location for further examination. A user-friendly website is implemented to give the health professionals remote access to the collected data. The entire system is housed in a 3-D printed enclosure for protection. It is important to note that the AI system is trained using multiple data sets obtained from Physionet, as the collection of training data using this system is a multiyear project requiring a staff of health professionals, which is beyond the scope of this project.