Psychiatry research · 2026
Digital phenotyping using wearable-determined physical behaviours and machine learning to detect depression and anxiety symptoms in a general population.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Sameh Alireza, Nauha Laura, Seppänen Marjo, Niemelä Maisa, Oussalah Mourad, Korpelainen Raija, Farrahi Vahid
Аннотация
BACKGROUND: Depression and anxiety are widespread mental health disorders, yet their diagnosis remains challenging. Digital phenotyping provides a promising approach for detecting depression and anxiety. This study aims to explore the extent to which physical behaviour metrics can be used as digital phenotypes for identifying individuals with and without depression and anxiety symptoms using machine learning (ML) algorithms.
METHODS: At age 46 years, participants (n = 2,810) from the Northern Finland Birth Cohort 1966 carried wrist- and waist-worn accelerometers for 14 consecutive days. Physical activity, sedentary and sleep behaviours were measured using data from the waist- and wrist-worn devices. A total of 54 physical behaviour metrics were extracted for each participant. Severity of depression and anxiety were assessed using three validated instruments: the Beck Depression Inventory-II, Generalized Anxiety Disorder-7, and the Hopkins Symptom Checklist-25. Five ML algorithms were applied to identify depression and anxiety symptoms. SHapley Additive exPlanations (SHAP) was used to assess the contribution of metrics.
RESULTS: Random forest achieved the best performance with accuracy (66%-72%) and AUC (66%-70%) for all three instruments. Other models showed lower accuracy in predicting depression and anxiety symptoms. In SHAP analysis, wake-up time, time in bed, bedtime, and prolonged sedentary bouts emerged as most important features.
CONCLUSIONS: Physical behaviour metrics using ML models can be utilized with reasonable accuracy to differentiate between participants with and without depression and anxiety symptoms. Our findings support the utility of physical behaviour digital phenotypes for predicting depression and anxiety symptoms in a general population.
METHODS: At age 46 years, participants (n = 2,810) from the Northern Finland Birth Cohort 1966 carried wrist- and waist-worn accelerometers for 14 consecutive days. Physical activity, sedentary and sleep behaviours were measured using data from the waist- and wrist-worn devices. A total of 54 physical behaviour metrics were extracted for each participant. Severity of depression and anxiety were assessed using three validated instruments: the Beck Depression Inventory-II, Generalized Anxiety Disorder-7, and the Hopkins Symptom Checklist-25. Five ML algorithms were applied to identify depression and anxiety symptoms. SHapley Additive exPlanations (SHAP) was used to assess the contribution of metrics.
RESULTS: Random forest achieved the best performance with accuracy (66%-72%) and AUC (66%-70%) for all three instruments. Other models showed lower accuracy in predicting depression and anxiety symptoms. In SHAP analysis, wake-up time, time in bed, bedtime, and prolonged sedentary bouts emerged as most important features.
CONCLUSIONS: Physical behaviour metrics using ML models can be utilized with reasonable accuracy to differentiate between participants with and without depression and anxiety symptoms. Our findings support the utility of physical behaviour digital phenotypes for predicting depression and anxiety symptoms in a general population.