Design and Implementation of a Multi-Parameter Health Monitoring SoC using RISC-V and TinyML for Real-Time Edge-Based Diagnostics

The increasing demand for continuous cardiac monitoring has created the need for portable, low-power, and intelligent health-tracking systems capable of operating reliably in edge environments. Traditional cloud-dependent solutions suffer from latency, bandwidth limitations, and privacy concerns, making them unsuitable for real-time diagnostics. This work presents a compact multi-parameter health monitoring System-on-Chip (SoC) implemented on an ESP32-C3 RISC-V microcontroller with integrated TinyML for on-device analytics. The system acquires ECG signals using the AD8232 module and measures heart rate and$\text{SpO}_{2}$using the MAX30100 optical sensor, while HRV metrics are computed directly from ECG-derived RR intervals. A lightweight TinyML model deployed on the microcontroller enables local classification of cardiac conditions without cloud support, achieving low latency and reduced energy consumption. Experimental evaluation demonstrates stable ECG waveform detection, accurate heart rate, and$\text{SpO}_{2}$estimation, and reliable HRV computation, supported by real-time visualization through serial and dashboard interfaces. The results validate the feasibility of a low-cost, scalable, and edge-based IoMT system capable of early detection of cardiac anomalies such as arrhythmia and hypoxemia, highlighting its potential for wearable and remote healthcare applications.