JMIR mental health · 2026
Cross-Dataset Evaluation of an Automated Video-Based Model for Detecting Tardive Dyskinesia Using the Clinician's Tardive Inventory: Validation Study.
СтатьяВалидационное исследованиеНазвание 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.
Trosch Richard, Sterns Anthony, Grimm Bradley, Larsen Loren, Talbot Bretton, Ma Fred, Ranjan Rakesh, MacMillan Carlene, Hughes Joel, Friedman Joseph, Muir Owen
Аннотация
BACKGROUND: Tardive dyskinesia (TD) is a common, often underrecognized movement disorder resulting from long-term antipsychotic use, yet its detection in routine mental health care remains inconsistent despite the availability of structured rating scales.
OBJECTIVE: This study evaluated the performance of an artificial intelligence-powered, video-based model for detecting abnormal movements associated with TD using the Clinician's Tardive Inventory (CTI) dataset. We compare automated assessments of videos from the CTI dataset with previously completed clinician-rated Abnormal Involuntary Movement Scale (AIMS) and CTI scores for the dataset's videos to determine the model's reliability and the accuracy of its assessment conclusions relative to expert raters.
METHODS: In total, 69 videos with corresponding AIMS and CTI ratings were analyzed using the visual transformer algorithm model called TDtect reported previously. The dataset included single-video assessments per participant, with varied instructions and movement types. The relationship between automated predictions and clinician ratings was assessed using Pearson correlation, and predictive accuracy was evaluated using area under the curve (AUC) metrics.
RESULTS: The model showed a strong correlation with AIMS total scores (r=0.717) and high diagnostic accuracy (AUC 0.854), which improved further at an optimized threshold (AUC 0.900). Performance differed across anatomical regions, with the tongue, lips, and jaw displaying the highest predictive reliability. Functional CTI components had weaker correlations (r=0.27-0.63), as expected due to the subjective nature of these measures.
CONCLUSIONS: These findings provide preliminary evidence that an artificial intelligence-driven TD detection model can generalize across video protocols, suggesting potential for broader clinical applicability, although further validation is needed. Future refinements and fine-tuning are expected to enhance accuracy, particularly in predicting functional impact.
OBJECTIVE: This study evaluated the performance of an artificial intelligence-powered, video-based model for detecting abnormal movements associated with TD using the Clinician's Tardive Inventory (CTI) dataset. We compare automated assessments of videos from the CTI dataset with previously completed clinician-rated Abnormal Involuntary Movement Scale (AIMS) and CTI scores for the dataset's videos to determine the model's reliability and the accuracy of its assessment conclusions relative to expert raters.
METHODS: In total, 69 videos with corresponding AIMS and CTI ratings were analyzed using the visual transformer algorithm model called TDtect reported previously. The dataset included single-video assessments per participant, with varied instructions and movement types. The relationship between automated predictions and clinician ratings was assessed using Pearson correlation, and predictive accuracy was evaluated using area under the curve (AUC) metrics.
RESULTS: The model showed a strong correlation with AIMS total scores (r=0.717) and high diagnostic accuracy (AUC 0.854), which improved further at an optimized threshold (AUC 0.900). Performance differed across anatomical regions, with the tongue, lips, and jaw displaying the highest predictive reliability. Functional CTI components had weaker correlations (r=0.27-0.63), as expected due to the subjective nature of these measures.
CONCLUSIONS: These findings provide preliminary evidence that an artificial intelligence-driven TD detection model can generalize across video protocols, suggesting potential for broader clinical applicability, although further validation is needed. Future refinements and fine-tuning are expected to enhance accuracy, particularly in predicting functional impact.