Clinical neurophysiology : official journal of the International Federation of Clinical Neurophysiology · 2026
Machine learning based EMG analysis of intermuscular coherence and cumulant density in tremor and myoclonus.
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.
van den Brandhof Elina L, van der Stouwe A M Madelein, van der Veen Sterre, Tuitert Inge, Elting Jan W J, Dalenberg Jelle R, Klamer Marrit R, Marapin Ramesh S, Biehl Michael, Tijssen Marina A J
Аннотация
OBJECTIVES: Clinical distinction of essential tremor (ET) and cortical myoclonus (CM) remains challenging due to overlapping symptoms. This study aims to identify characteristics in muscle contraction patterns in individual muscles, as well as intermuscular EMG coupling between agonist antagonist pairs in ET and CM, and to evaluate their diagnostic suitability.
METHODS: We analyzed arm muscle activity in 19 ET and 19 CM patients during two postures with pronated outstretched arms: one with straight wrists and one with extended wrists. We analyzed power spectra, agonist-antagonist coherence, and cumulant density using classical statistical and machine learning methods.
RESULTS: Machine learning analysis achieved high classification performance using engineered features (AUROC: 0.93), power spectra (0.92), and coherence (0.93). Cumulant density-based analysis was less discriminative (0.67), though performance improved with reduced muscle activation. ET was characterized by regular contraction pattern and predominantly alternating bursts depending on the posture, while CM showed irregular, synchronous bursts.
CONCLUSION: The identified EMG signatures - regular alternating bursts in ET and irregular synchronous bursts in CM - demonstrate strong potential to support objective diagnosis.
SIGNIFICANCE: These findings provide statistical evidence supporting clinically observed contraction patterns in ET and CM, which could enhance diagnostic accuracy.
METHODS: We analyzed arm muscle activity in 19 ET and 19 CM patients during two postures with pronated outstretched arms: one with straight wrists and one with extended wrists. We analyzed power spectra, agonist-antagonist coherence, and cumulant density using classical statistical and machine learning methods.
RESULTS: Machine learning analysis achieved high classification performance using engineered features (AUROC: 0.93), power spectra (0.92), and coherence (0.93). Cumulant density-based analysis was less discriminative (0.67), though performance improved with reduced muscle activation. ET was characterized by regular contraction pattern and predominantly alternating bursts depending on the posture, while CM showed irregular, synchronous bursts.
CONCLUSION: The identified EMG signatures - regular alternating bursts in ET and irregular synchronous bursts in CM - demonstrate strong potential to support objective diagnosis.
SIGNIFICANCE: These findings provide statistical evidence supporting clinically observed contraction patterns in ET and CM, which could enhance diagnostic accuracy.