Comprehensive Symptom Prediction in Inpatients With Acute Psychiatric Disorders Using Wearable-Based Deep Learning Models: Development and Validation Study
Deep learning models based on wearable sensor data effectively classified symptom deterioration and predicted symptom severity in participants in acute psychiatric wards. Despite lower computational costs, Multi models demonstrated equivalent or superior performance than Single models, suggesting that multitask learning is a promising approach for comprehensive symptom prediction. However, significant variations were observed across wards, which presents a key challenge for developing clinical decision support systems in acute psychiatric wards. Future studies may benefit from recurring local validation or federated learning to address generalizability issues.
Authors
- Yu Sang Lee (ORCID: https://orcid.org/0000-0003-3411-9790)
- Minseok Hong (ORCID: https://orcid.org/0000-0001-6028-8286)
- Yong Min Ahn (ORCID: https://orcid.org/0000-0002-4458-797X)
- Kang Yoon Lee (ORCID: https://orcid.org/0000-0003-3078-6166)
- Sang Jin Rhee (ORCID: https://orcid.org/0000-0001-6221-5953)
- Yong-gyom Kim
- Jeong Hun Yang (ORCID: https://orcid.org/0000-0002-1286-264X)
- Hyunju Lee (ORCID: https://orcid.org/0000-0002-4237-070X)
- Tak Youn (ORCID: https://orcid.org/0000-0002-5239-1993)
- Se Hyun Kim (ORCID: https://orcid.org/0000-0001-7078-2835)
- HongGi Kim
- Ri-Ra Kang
Institutions
- Gachon University (KR)
- Seoul National University (KR)
- Chungnam National University (KR)
- Dongguk University (KR)
- Seoul National University Hospital (KR)
- Silicon Works (South Korea) (KR)
- Yongin Mental Hospital (KR)
Publication Details
- Journal
- Journal of Medical Internet Research
- Published
- 2024-10-20
- DOI
- https://doi.org/10.2196/65994
- Citations
- 7
- Primary Topic
- Mental Health Research Topics
- Type
- article
- Field-Weighted Citation Impact
- 3.55