Ultra-low Power Analog Folded Neural Network for Cardiovascular Health Monitoring
A wearable heart guardian that can run with extremely low power and continuously check ECG signals.

Overview
Most health wearables trade battery life for intelligence. This project redesigns that balance. We built an analog folded neural network that works like a compact, energy-sipping specialist for ECG patterns. Folded means the same hardware is reused cleverly over time, so we do more with less power and less chip area. For non-experts: imagine one skilled doctor examining patients one by one very efficiently instead of hiring a full hospital for every check. The result is continuous heart monitoring with tiny energy needs, opening the door to lighter, longer-lasting, and more practical preventive healthcare devices.
Real-World Impact
Makes always-on heart monitoring more realistic for everyday users, not just clinical environments.
Technologies & Techniques
Key Achievements
Near-batteryless operating profile through serialized computation
Optimized 6-layer model (hidden size 30) for ECG anomaly screening
Continuous monitoring suitable for day-long wearable use
Strong detection performance for key cardiovascular warning patterns
Low thermal noise and compact on-chip footprint
Lower peak power compared with conventional analog neural implementations
Publications
Citation data synced from scholarly cache.
Neural network design via voltage-based resistive processing unit and diode activation function-a new architecture
2021 IEEE International Midwest Symposium on Circuits and Systems (MWSCAS)
2021
Cited by 18
Ultra-low Power Analog Recurrent Neural Network Design Approximation for Wireless Health Monitoring
2022 IEEE 19th International Conference on Mobile Ad Hoc and Smart Systems (MASS)
2022
Cited by 14
Hybrid analog-digital sensing approach for low-power real-time anomaly detection in drones
2021 IEEE 18th International Conference on Mobile Ad Hoc and Smart Systems (MASS)
2021
Cited by 12
Ultra-low power analog folded neural network for cardiovascular health monitoring
IEEE Journal of Biomedical and Health Informatics
2024
Cited by 6
Related Projects
AI-Assisted S-Parameter Prediction
Like a GPS for high-speed circuits: AI predicts signal behavior in seconds instead of waiting hours for simulations.
Learn more →
Spiking Neural Networks for Signal Processing
Brain-inspired AI that fires only when needed, reducing power use for always-on sensing.
Learn more →
Want to Discuss This Research?
I'm always excited to discuss technical details, potential applications, or collaborative opportunities.