ECE496Y Final Report: Edge Machine Learning for Detecting Freezing of Gait in Parkinson's Patients

TL;DR

This project developed an edge machine learning device that detects freezing of gait in Parkinson's patients with 83.7% accuracy, enabling real-time monitoring using a microcontroller.

Contribution

It introduces a novel edge ML algorithm deployed on a microcontroller for real-time detection of gait freezing in Parkinson's patients.

Findings

Achieved 83.7% accuracy in validation

Successfully deployed on Arduino Nano 33 BLE Sense

Validated real-time operation with streamed data

Abstract

Parkinson's disease is a common neurological disease, entailing a multitude of motor deficiency symptoms. In this project, we developed a device with an uploaded edge machine learning algorithm that can detect the onset of freezing of gait symptoms in a Parkinson's patient. The algorithm achieved an accuracy of 83.7% in a validation using data from ten patients. The model was deployed in a microcontroller Arduino Nano 33 BLE Sense Board model and validated in real-time operation with data streamed to the microcontroller from a computer.