A 48.6-to-105.2µW machine-learning assisted cardiac sensor SoC for mobile healthcare monitoring

A machine-learning (ML) assisted cardiac sensor SoC (CS-SoC) is designed for healthcare monitoring with mobile devices. The architecture realizes the cardiac signal acquisition with versatile feature extractions and classifications, enabling higher order analysis over traditional DSPs. Besides, the dynamic standby controller further suppresses the leakage power dissipation. Implemented in 90nm CMOS, the CS-SoC dissipates 48.6/105.2μW at 0.5-1.0V for real-time arrhythmia/myocardial infarction syndrome detection with 95.8/99% accuracy.

[1]  Kaushik Roy,et al.  Exploring Asynchronous Design Techniques for Process-Tolerant and Energy-Efficient Subthreshold Operation , 2010, IEEE Journal of Solid-State Circuits.

[2]  Mario Konijnenburg,et al.  A voltage-scalable biomedical signal processor running ECG using 13pJ/cycle at 1MHz and 0.4V , 2011, 2011 IEEE International Solid-State Circuits Conference.

[3]  Ray-Jade Chen,et al.  A sub-100µW multi-functional cardiac signal processor for mobile healthcare applications , 2012, 2012 Symposium on VLSI Circuits (VLSIC).

[4]  Chen-Yi Lee,et al.  A frequency accuracy enhanced sub-10µW on-chip clock generator for energy efficient crystal-less wireless biotelemetry applications , 2010, 2010 Symposium on VLSI Circuits.

[5]  Chin-Teng Lin,et al.  A vectorcardiogram-based classification system for the detection of Myocardial infarction , 2011, 2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[6]  J. Kwong,et al.  An Energy-Efficient Biomedical Signal Processing Platform , 2010, IEEE Journal of Solid-State Circuits.

[7]  Fan Zhang,et al.  A batteryless 19μW MICS/ISM-band energy harvesting body area sensor node SoC , 2012, 2012 IEEE International Solid-State Circuits Conference.

[8]  H. L. Gray,et al.  Applied time series analysis , 2011 .