A 1.2nW Analog Electrocardiogram Processor Achieving a 99.63% QRS Complex Detection Sensitivity.

Area of Science:

• Biomedical Engineering
• Analog Integrated Circuit Design
• Low-Power Electronics

Background:

• Real-time QRS detection is crucial for cardiac monitoring.
• Existing methods often face challenges with power consumption and complexity.
• Need for efficient processors in wearable and implantable ECG devices.

Purpose of the Study:

• To develop an energy-efficient analog processor for real-time QRS detection.
• To implement the Pan-Tompkins algorithm in the analog domain for reduced power.
• To minimize hardware complexity and power consumption for wearable ECG applications.

Main Methods:

• Algorithm implementation in the analog domain using subthreshold-biased electronics.
• Utilization of operational transconductance amplifiers (OTAs) for processing blocks.
• Current-signal squaring and fixed-level thresholding to reduce complexity.

Main Results:

• Processor designed in 65 nm TSMC CMOS technology with a 0.078 mm² footprint.
• Achieved ultra-low power consumption of 1.2 nW at a 1 V supply.
• Demonstrated high QRS detection accuracy: 99.63% sensitivity and 99.47% positive predictivity on MIT-BIH database.

Conclusions:

• The proposed analog ECG processor offers significant energy efficiency for real-time QRS detection.
• The analog implementation and simplified thresholding effectively reduce power and area.
• This design is well-suited for low-power, high-performance wearable cardiac monitoring systems.