eudasense.png
MIMS Final Project 2016

EudaeSense

The vast majority of mood tracking applications today have a significant drawback: they require too much user input and effort. Our solution is a system that passively collects user data, which is used to infer early signs of potential depressed mood, and suggests behavioral interventions.

We use wearable biometric sensors to measure and track physiological signals that reflect emotion such as heart rate, sweat, temperature, muscle tension, and breathing rate. We also use smartphone and wearable sensors to monitor behavior associated with mood changes such as sleep quality, activity level, mobility and social interaction.

Our smartphone/smartwatch application collects this data and transmits it to a back-end system that uses generalized machine learning models to predict possible mood changes.

Lastly, our application suggests behavioral interventions. Interventions were derived from existing research on cognitive behavioral therapy, in tandem with expert interviews and user studies which uncovered new and novel approaches to digital mental health management. These contextually relevant, just-in-time interventions are pushed to users discreetly on their smartwatch. 

Last updated: October 7, 2016