Multi-Lead Wireless ECG

Goel, Nisha

Student Work

Multi-Lead Wireless ECG

Public

This report focuses on the design of a multi-lead wireless electrocardiogram (ECG) monitor, the design of an atrial fibrillation detector, and the design and implementation of an R-wave detector in MATLAB. The circuit has leads that collect ECG signals from a subject and send the signals to a Nordic microcontroller (nRF52840) that uses Bluetooth to wirelessly transfer the data to a computer. Signal processing is done on the collected signal using an R-wave detector and atrial fibrillation detector. An ECG circuit using an instrumentation amplifier-based front end, linear filtering, and ADC conversion was designed on a protoboard, solderboard, and is now ready to be manufactured on a PCB. We also implemented an atrial fibrillation detector in MATLAB that classifies beat segments based on calculated values for Shannon entropy and the time varying coherence function. Finally, an R-wave detector was implemented in MATLAB by cascading a bandpass filter (fourth-order Butterworth, 5-15 Hz passband), squaring operation, lowpass filter (second-order, 5 Hz cutoff), and a peak detector. Our R-wave detection algorithm works well for both the arrhythmia and atrial fibrillation databases, producing a QRS sensitivity of 85% or higher for 64/71 of the records we tested from both the Physionet arrhythmia and atrial fibrillation databases, and a QRS positive predictivity of 91% or higher for 69/71 of the records from both databases. (NOTE: Due to a team member (N. Goel) graduating earlier than the rest of the team, only the R-wave detection algorithm is complete for this report and the other sections are making headway for completion in C-Term 2022.)

This report represents the work of one or more WPI undergraduate students submitted to the faculty as evidence of completion of a degree requirement. WPI routinely publishes these reports on its website without editorial or peer review.

Creator