Intellectual Property India Publishes Patent Application for A Risc-V Edge Ai Powered Wearable System For Real Time Physiological Signal Monitoring And Anomaly Detection

MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202641072543 A) filed by The Principal, Mepco Schlenk Engineering College on June 11, 2026, for A Risc-V Edge Ai Powered Wearable System For Real Time Physiological Signal Monitoring And Anomaly Detection.

Inventors include Dr. D. Dinesh Kumar; V. Arun Raj; and Dr. S. Sridhar Raj.

The application for the patent was published on June 19, 2026, under issue no. 25/2026.

Abstract: ABSTRACT OF THE INVENTION: A RISC-V Edge Al-Powered Wearable System for Real Time Physiological Signal Monitoring and Anomaly Detection This invention discloses an indigenous, low-latency wearable prototype built on the RJSC-V open-source Instruction Set Architecture (!SA) for real-time physiological signal monitoring and on-device anomaly detection using Edge Artificial Intelligence (Edge AI). The system is implerim1ted on the VSDSquadron PRO development board (SiFive FE31 O-G002 RISC-V core, RV32lMAC ISA) and integrates a multimodal biomedical sensor suite comprising ECG, PPG/SpOz, skin temperature, and 3-axis accelerometer modules connected via FC/SPI interfaces. Lightweight neural network models-Tiny-VGG, ID-CNN, and CNN-LSTMare trained on physiological datasets (MIT-BIH, PhysioNet) with synthetic data augmentation using GANs/V AEs, then optimized via post-training quantization to 8-bit integer Tensor Flow Lite format (::SI MB footprint). The quantized model is deployed as C-header files. on the RJSC-V MCU, where a bare-metal integer-arithmetic inference engine performs real-time classification of physiological anomalies-including ARRHYTHMIA, HYPERTENSION, HYPOTENSION, STRESS, FEVER, and HYPOXIA-within ::;50 ms latency. Multimodal sensor fusion at the feature level enhances diagnostic robustness and reduces false positives. An optional BLE 5.0 module enables wireless alert transmission to healthcare professionals. The invention eliminates cloud dependency, preserves patient data privacy, operates under ultra-low power consumption, and advances India's indigenous RlSC-V semiconductor ecosystem for healthcare applications, targeting TRL 7 for commercialization.

Disclaimer: Curated by HT Syndication.