Related Experiment Video
Updated: Sep 10, 2025
12:51
Evaluation of Commercial-Off-The-Shelf Wrist Wearables to Estimate Stress on Students
Published on: June 16, 2018
7.6K
Smartphone sensor-based depression detection in campus environments: a proof-of-concept study with small-sample
Yichen Bai1, Yueze Liu1, Yang Zhang2
1School of Information Science and Engineering, Lanzhou University, Lanzhou, China.
Frontiers in Psychiatry
|August 25, 2025
Summary
Smartphone sensors can detect depression in Chinese university students, correlating with lifestyle factors like sleep and diet. This technology offers a novel approach to early mental health intervention.
Area of Science:
- Digital health
- Mental health technology
- Behavioral science
Background:
- Depression is a growing global concern, especially among adolescents.
- University students experience unique mental health challenges.
- Early detection of depression is crucial for effective intervention.
Purpose of the Study:
- To explore the feasibility of using smartphone sensor data for depression detection in Chinese university students.
- To identify behavioral patterns associated with depressive symptoms using smartphone sensors.
- To assess the accuracy of machine learning models in detecting depression based on sensor data.
Main Methods:
- Collected data from accelerometers, gyroscopes, and light sensors from 12 university students.
- Developed a custom data processing scheme for campus environments.
- Extracted 18 feature sequences, performed feature selection using Pearson correlation, and validated models with leave-one-out cross-validation.
- Utilized common classification algorithms for model training and evaluation.
Main Results:
- Achieved detection accuracy rates ranging from 73.11% to 88.24%.
- Identified significant negative correlations between depression scores (PHQ-9) and dietary regularity, bedtime consistency, and physical activity levels.
- Demonstrated the link between smartphone-derived behavioral data and self-reported depressive symptoms.
Conclusions:
- Smartphone sensors show promise for non-invasive, early depression detection in Chinese higher education settings.
- Behavioral patterns captured by smartphones can serve as indicators of depressive symptoms.
- This approach supports the development of novel, technology-driven mental health support systems for students.
