Journal of affective disorders · 2025
A clinical prediction model for risk of depressive episode based on facial behavioral features: A cross-sectional study.
СтатьяНазвание и аннотация: Перевод пока не подготовлен
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.
Qu Tingting, Wang Chennan, Gu Qiwen, Wang Qi, Fan Xiaoxuan, Luo Ranyi, Cao Yuping
Аннотация
BACKGROUND: As an explicit indicator of emotional states, facial behavior features have become an important entry point for depressive episodes because of its objectivity as a biomarker, convenience in collection through non-invasive procedures. This study aimed to develop a clinical prediction model for the risk of depressive episodes using facial behavioral features.
METHODS: A total of 176 hospitalized patients with depressive episodes and 155 healthy controls matched by age and gender were included. Comprehensive assessments including demographic data collection, clinical evaluations, facial video recordings for eye gaze analysis, head movements tracking, and facial action units (AUs) metrics extraction were performed. Independent predictors of depressive episodes were screened by LASSO regression, followed by logistic regression modeling. Areas under the curve (AUC), calibration plot, Hosmer-Lemeshow (H-L) test and Decision Curve Analysis (DCA) were used to validate the discrimination, calibration and clinical usefulness of the model.
RESULTS: Seven predictive factors were identified from 49 variables through LASSO regression, including educational level and standardized AU metrics: AU10_rMean, AU14_rMean, AU26_rMean, AU02_rStd, AU07_rStd, and AU12_rStd. The model displayed moderate predictive ability, with an AUC of 0.863 in the training set and 0.798 in the validation set. The calibration plot and decision curve analysis revealed that this nomogram was in good fitness.
CONCLUSION: This study identified seven potential predictors of depressive episodes and developed a prediction model with stable discriminative performance. It suggested that such facial behavioral features may serve as auxiliary biomarkers for clinical detection, enabling timely interventions to prevent complications associated with depressive episodes.
METHODS: A total of 176 hospitalized patients with depressive episodes and 155 healthy controls matched by age and gender were included. Comprehensive assessments including demographic data collection, clinical evaluations, facial video recordings for eye gaze analysis, head movements tracking, and facial action units (AUs) metrics extraction were performed. Independent predictors of depressive episodes were screened by LASSO regression, followed by logistic regression modeling. Areas under the curve (AUC), calibration plot, Hosmer-Lemeshow (H-L) test and Decision Curve Analysis (DCA) were used to validate the discrimination, calibration and clinical usefulness of the model.
RESULTS: Seven predictive factors were identified from 49 variables through LASSO regression, including educational level and standardized AU metrics: AU10_rMean, AU14_rMean, AU26_rMean, AU02_rStd, AU07_rStd, and AU12_rStd. The model displayed moderate predictive ability, with an AUC of 0.863 in the training set and 0.798 in the validation set. The calibration plot and decision curve analysis revealed that this nomogram was in good fitness.
CONCLUSION: This study identified seven potential predictors of depressive episodes and developed a prediction model with stable discriminative performance. It suggested that such facial behavioral features may serve as auxiliary biomarkers for clinical detection, enabling timely interventions to prevent complications associated with depressive episodes.