ABSTRACT:
Cardiovascular diseases (CVDs) remain the leading cause of death globally, with the heaviest burden falling on populations in remote areas with limited access to specialist medical personnel. Within the context of "Health Without Borders," technological innovation is required to bring high-level diagnostic capabilities directly to patients. This research proposes the development of an affordable and efficient Edge Intelligence framework for remote cardiac monitoring, specifically designed for resource-constrained environments. The study adopts an electrical engineering and digital signal processing approach by utilizing secondary data from PhysioNet, specifically the PTB Diagnostic ECG Database and the MIT-BIH Arrhythmia Database. The methodology involves the pre-processing of electrocardiogram (ECG) signals for noise reduction and feature extraction using Lightweight Deep Learning algorithms. The primary focus is optimizing the model to run on low-power hardware (Edge Computing) without requiring constant internet connectivity or expensive server infrastructure. Simulation results demonstrate that the developed AI model can detect various types of arrhythmias with accuracy levels exceeding basic clinical standards, while maintaining very low power and memory consumption. This proves that advanced technological solutions can be implemented inclusively without relying on expensive medical devices that are often inaccessible to health facilities in border regions. Practically, this research provides a proof of concept that the integration of electrical engineering and public health can create independent and sustainable early warning systems. In conclusion, the use of Edge-based AI represents a concrete step toward realizing health inclusion, where geographical and economic boundaries no longer hinder access to quality cardiac diagnosis. This project supports the vision of universal health equity for all levels of society.