Machine Learning for Detecting Depression, Anxiety, and Stress in College Students
Author Information
Author(s): Bruno Luis Schaab, Prisla Ücker Calvetti, Sofia Hoffmann, Gabriela Bertoletti Diaz, Maurício Rech, Sílvio César Cazella, Airton Tetelbom Stein, Helena Maria Tannhauser Barros, Pamela Carvalho da Silva, Caroline Tozzi Reppold
Primary Institution: Universidade Federal de Ciências da Saúde de Porto Alegre
Hypothesis
How do machine learning models perform in the detection of depression, anxiety, and stress among undergraduate students?
Conclusion
Machine learning algorithms show promise in detecting mental health issues among university students, but their practical applicability needs further scrutiny.
Supporting Evidence
- Out of 2,304 articles found, 48 studies met the inclusion criteria.
- Accuracy in detecting stress ranged from 63% to 100%, anxiety from 53.69% to 97.9%, and depression from 73.5% to 99.1%.
- 30 out of 33 studies reported accuracy scores above 70%, indicating good performance.
Takeaway
This study looked at how computers can help find out if college students are feeling sad, anxious, or stressed. It found that these computer programs can be pretty good at figuring this out.
Methodology
A systematic review of studies using machine learning to detect mental health issues in undergraduate students, analyzing data from various databases.
Potential Biases
Most studies relied on internal validation, which may overstate performance data.
Limitations
The quality of evidence was assessed as very low, with serious risks of bias and issues in the assessment of outcomes.
Participant Demographics
Participants were predominantly undergraduate students, with ages ranging from 17 to 67 years and a majority being female.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website