CovidAlert -- A Wristwatch-based System to Alert Users from Face Touching

TL;DR

CovidAlert is a smartwatch-based system that detects face touching using accelerometer and gyroscope data, providing real-time alerts to help reduce virus transmission during the pandemic.

Contribution

The paper introduces a novel, energy-efficient smartwatch system using machine learning to detect face touching and alert users, addressing a common behavioral challenge during COVID-19.

Findings

Achieved 88.4% accuracy in detecting face touching.

Implemented system on a commercial smartwatch (Fossil Gen 5).

Reduced false positives and negatives with energy-efficient algorithms.

Abstract

Worldwide 2019 million people have been infected and 4.5 million have lost their lives in the ongoing Covid-19 pandemic. Until vaccines became widely available, precautions and safety measures like wearing masks, physical distancing, avoiding face touching were some of the primary means to curb the spread of virus. Face touching is a compulsive human begavior that can not be prevented without making a continuous consious effort, even then it is inevitable. To address this problem, we have designed a smartwatch-based solution, CovidAlert, that leverages Random Forest algorithm trained on accelerometer and gyroscope data from the smartwatch to detects hand transition to face and sends a quick haptic alert to the users. CovidALert is highly energy efficient as it employs STA/LTA algorithm as a gatekeeper to curtail the usage of Random Forest model on the watch when user is inactive. The overall accuracy of our system is 88.4% with low false negatives and false positives. We also demonstrated the system viability by implementing it on a commercial Fossil Gen 5 smartwatch.