A highly scalable deep learning language model for common risks prediction among psychiatric inpatients
BMC Medicine, vol. 23, pp. 308
Abstract
BACKGROUND: There is a lack of studies exploring the performance of Transformers-based language models in common risks assessment among psychiatric inpatients. We aim to develop a scalable risk assessment model using multidimensional textualized data and test the stability, robustness, and benefit of this approach. METHODS: In this real-world cohort study, a deep learning language model was developed and validated using first hospitalized cases diagnosed with schizophrenia, bipolar disorder, and depressive disorder between January 2016 and March 2023 in three hospitals. The algorithm was externally validated on an independent testing cohort comprising 1180 patients. A total of 140 features, including first medical records (FMR), laboratory examinations, medical orders, and psychological scales, were assessed for analysis. The outcomes were short- and long-term impulsivity (STI and LTI), risk of suicide (STSS and LTSS), and need of physical restraint (STPR and LTPR) assessed by qualified nurses or clinicians. Analysis was carried out between August 2024 and June 2024. Models with different architectures and input settings were compared with each other. The area under the receiver operating characteristic curve (AUROC) was used to assess the primary performance of models. The clinical utility was determined by the net benefit under Youden's threshold. RESULTS: Of 7451 patients included in this study, 2982 (47.6%) were male, and the median (interquartile range) age was 42 (28-57) years. The overall incidence of outcomes was 635 (8.5%), 728 (10.5%), 659 (8.8%), 803 (10.8%), 588 (7.9%), and 728 (9.8%) for STPR, LTPR, STSS, LTSS, STI, and LTI, respectively. The multitask semi-structured Transformers-based language (SSTL) model showed more promising AUROCs (STPR: 0.915; LTPR: 0.844; STSS: 0.867; LTSS: 0.879; STI: 0.899; LTI: 0.894) in the prediction of these outcomes than single-tasked or multimodal language models and traditional structured data models. Combining FMR with other data from electronic health records led to significant improvements in the performance and clinical utility of SSTL models based on demographic, diagnosis, laboratory tests, treatment, and psychological scales. CONCLUSIONS: The SSTL model shows potential advantages in prognostic evaluation. FMR is a strong predictor for common risks prediction and may benefit other tasks in psychiatry with minimum requirements for data and data processing.
Authors 19
-
Affiliation as printed
School of Medicine, Tongji University, Shanghai, China
-
Affiliation as printed
School of Medicine, Tongji University, Shanghai, China
-
Shanghai Jiao Tong University · Shanghai Mental Health Center · Shanghai Key Laboratory of Psychotic Disorders
Affiliation as printed
Shanghai Key Laboratory of Psychotic Disorders, Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine, Shanghai, 200030, China
-
Affiliation as printed
Clinical Research Center for Mental Disorders, Shanghai Pudong New Area Mental Health Center, School of Medicine, Chinese-German Institute of Mental Health, Tongji University, Shanghai, China
-
Shanghai Changning Mental Health Center
Affiliation as printed
Shanghai Changning Mental Health Center, Changning District, Shanghai, China
-
Shanghai Changning Mental Health Center
Affiliation as printed
Shanghai Changning Mental Health Center, Changning District, Shanghai, China
-
Shanghai Changning Mental Health Center
Affiliation as printed
Shanghai Changning Mental Health Center, Changning District, Shanghai, China
-
Sichuan University · West China Hospital of Sichuan University
Affiliation as printed
Department of Infection Control, West China Hospital, Sichuan University, Chengdu, China
Division of Gastrointestinal Surgery, Department of General Surgery, West China Hospital, Sichuan University, Chengdu, China
-
Affiliation as printed
Shanghai Jinshan District Mental Health Center, Jinshan District, Shanghai, China
-
Affiliation as printed
School of Medicine, Tongji University, Shanghai, China
-
Affiliation as printed
Lakefield College School, Lakefield, ON, Canada
-
Affiliation as printed
Shanghai Putuo Mental Health Center, Putuo District, Shanghai, China
-
Affiliation as printed
Shanghai Putuo Mental Health Center, Putuo District, Shanghai, China
-
Shanghai Hospital Development Center
Affiliation as printed
Shanghai Hospital Development Center, Shanghai, China
-
Affiliation as printed
Clinical Research Center for Mental Disorders, Shanghai Pudong New Area Mental Health Center, School of Medicine, Chinese-German Institute of Mental Health, Tongji University, Shanghai, China
-
East China University of Science and Technology
Affiliation as printed
East China University of Science and Technology, Shanghai, China
-
Jiaojiao Hou Aachen
Universitätsklinikum Aachen · RWTH Aachen University
Affiliation as printed
University Clinic of Child and Adolescent Psychiatry, Psychosomatics and Psychotherapy, RWTH Aachen University, Aachen, 52074, Germany
-
Hui Li corresponding
Shanghai Jiao Tong University · Shanghai Mental Health Center · Shanghai Key Laboratory of Psychotic Disorders
Affiliation as printed
Shanghai Key Laboratory of Psychotic Disorders, Shanghai Mental Health Center, Shanghai Jiaotong University School of Medicine, Shanghai, 200030, China. lihuindyxs@163.com
-
Zisheng Ai corresponding
Affiliation as printed
Department of Medical Statistics, School of Medicine, Tongji University, Shanghai, China. azs1966@126.com
Cited by 7 stored of 7
7 results
No patents citing this paper on Lens.org (checked 2026-10-06).