Related Experiment Video
Updated: Dec 28, 2025
12:51
Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students
Published on: June 16, 2018
7.8K
Evaluating depression with multimodal wristband-type wearable device: screening and assessing patient severity
Yuuki Tazawa1, Kuo-Ching Liang1, Michitaka Yoshimura1
1Keio University School of Medicine, Tokyo, Japan.
Heliyon
|February 15, 2020
Summary
Wearable devices and machine learning show promise for identifying depression and assessing its severity. The algorithm achieved 0.76 accuracy in detecting symptomatic patients using sensor data.
Area of Science:
- Digital health
- Machine learning in medicine
- Wearable sensor technology
Background:
- Depression screening and severity assessment are crucial for effective treatment.
- Traditional methods can be subjective and time-consuming.
- Wearable devices offer continuous, objective data collection.
Purpose of the Study:
- To develop and validate a machine learning algorithm for depression screening and severity assessment.
- To leverage data from wearable devices for mental health monitoring.
Main Methods:
- Utilized a wearable device measuring physiological and movement data (steps, heart rate, sleep, etc.).
- Collected continuous data from 45 depressed patients and 41 healthy controls.
- Developed machine learning models using XGBoost with 10-fold cross-validation.
Main Results:
- Significant differences in heart rate, steps, and sleep patterns were observed between depressed patients and controls.
- The model achieved 0.76 accuracy in identifying symptomatic patients using 7 days of data.
- Skin temperature and sleep-related features were key predictors in the machine learning model.
Conclusions:
- Machine learning algorithms analyzing wearable device data show potential for depression identification.
- Wearable technology may offer a scalable solution for monitoring depression severity.
- Further validation in larger, independent datasets is warranted.
