MedElevator
Войти в исследования Clinical parkinsonism & related disorders · 2026
A comparative study of video-based and electromyography-based detection of tics.
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.
Lowor Grace, Opoku Melvin Osei, Gomez Julieth, Tvenstrup William, Noy Andrea, Ackon Theresa, Cagle Jackson, Butson Chris, Foote Kelly, Okun Michael S, Gunduz Aysegul
Аннотация
BACKGROUND: Tourette Syndrome (TS) is a complex neurodevelopmental disorder characterized by motor and vocal tics that vary widely across individuals. Current clinical assessments primarily rely on subjective rating scales, limiting the precision of tic evaluation and treatment outcome monitoring. To improve clinical decision-making, objective, reliable detection methods are needed.
OBJECTIVES: This study compared the tic-detection performance of video and electromyography (EMG)-based algorithms. We hypothesized that there would be similar accuracy across modalities and that applying multimodal integration would yield potential benefits.
METHODS: A total of 286 video and EMG sessions were recorded synchronously from 14 participants over three months. Recordings were labeled as "Rest," "Tic," and "Move" using a video object detection algorithm and by applying an EMG double-threshold method. Video-based detection was used as the reference for comparison.
RESULTS: Overall agreement between modalities was 71.8%. EMG achieved 61.7% accuracy for "Rest" (precision: 72.4%, recall: 75.3%, F1-score: 73.8%). Despite an 81.6% accuracy, EMG performed poorly in detecting tics (precision: 15.1%, recall: 5.3%, F1-score: 7.9%) and voluntary movement (accuracy: 72.2%, precision: 14.5%, recall: 21.7%, F1-score: 17.4%). A Pearson correlation of 0.14 was found between tic frequencies captured by both EMG and video detectors, suggesting a weak positive relationship.
CONCLUSION: While EMG shows promise for detecting tics at rest, its overall stand-alone performance for tic identification was limited. These findings highlight the potential benefits of integrating video and EMG modalities to improve the accuracy and reliability of tic detection and monitoring in the setting of TS.
OBJECTIVES: This study compared the tic-detection performance of video and electromyography (EMG)-based algorithms. We hypothesized that there would be similar accuracy across modalities and that applying multimodal integration would yield potential benefits.
METHODS: A total of 286 video and EMG sessions were recorded synchronously from 14 participants over three months. Recordings were labeled as "Rest," "Tic," and "Move" using a video object detection algorithm and by applying an EMG double-threshold method. Video-based detection was used as the reference for comparison.
RESULTS: Overall agreement between modalities was 71.8%. EMG achieved 61.7% accuracy for "Rest" (precision: 72.4%, recall: 75.3%, F1-score: 73.8%). Despite an 81.6% accuracy, EMG performed poorly in detecting tics (precision: 15.1%, recall: 5.3%, F1-score: 7.9%) and voluntary movement (accuracy: 72.2%, precision: 14.5%, recall: 21.7%, F1-score: 17.4%). A Pearson correlation of 0.14 was found between tic frequencies captured by both EMG and video detectors, suggesting a weak positive relationship.
CONCLUSION: While EMG shows promise for detecting tics at rest, its overall stand-alone performance for tic identification was limited. These findings highlight the potential benefits of integrating video and EMG modalities to improve the accuracy and reliability of tic detection and monitoring in the setting of TS.