Computers in biology and medicine · 2026
Towards objective identification of myofascial trigger points: A high-density surface electromyography method.
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.
Churproong Seekaow, Ren Guangyuan, Chen Chen, Metcalfe Benjamin, McGuigan Polly, Zhang Dingguo
Аннотация
BACKGROUND: Myofascial trigger points (MTrPs), which arise from muscle overload and subsequent ischemia, contribute to myofascial pain syndrome (MPS). Despite the high prevalence of MPS, diagnosis remains challenging. There is a lack of objective methods for identifying MTrPs using high-density surface electromyography (HD-sEMG), particularly through analysis of EMG activity and decomposed motor unit (MU) characteristics between active and non-MTrPs.
METHODS: EMG and MU activities were compared between active MTrPs (n = 15) and non-MTrPs (n = 22) in the upper and middle trapezius. MU activities were extracted during three 2-s maximal contractions using gradient Convolution Kernel Compensation (gCKC)-based decomposition. Features were classified using five machine-learning models, and global explanation plots were generated to identify important parameters that predicted MTrP classification and to illustrate trends in EMG and MU activity values.
RESULTS: Participants with active MTrP reported severe pain (7 ± 1.25) and contributed 27 EMG samples, while non-MTrPs provided 41 samples. Active MTrPs exhibited lower mean ranks for EMG activity (RMS and MDF) and MU discharge rate (MU_DR) but higher mean ranks for MU number and MU action potential (MUAP) amplitude compared to non-MTrPs (p > 0.05). The artificial neural network (ANN) classifier achieved the highest performance for sample-wise cross-validation using the scaled dataset (82.35%). Although global explanations revealed MU_number as the lowest predictor, active MTrPs tend to show low RMS and MDF but elevated MUAP-amplitude features.
CONCLUSION: HD-sEMG is a promising method for identifying MTrPs. Machine learning methods, such as ANN-based classification, have demonstrated strong performance. However, further investigation is needed to improve classification accuracy for MTrP identification based on EMG decomposition.
METHODS: EMG and MU activities were compared between active MTrPs (n = 15) and non-MTrPs (n = 22) in the upper and middle trapezius. MU activities were extracted during three 2-s maximal contractions using gradient Convolution Kernel Compensation (gCKC)-based decomposition. Features were classified using five machine-learning models, and global explanation plots were generated to identify important parameters that predicted MTrP classification and to illustrate trends in EMG and MU activity values.
RESULTS: Participants with active MTrP reported severe pain (7 ± 1.25) and contributed 27 EMG samples, while non-MTrPs provided 41 samples. Active MTrPs exhibited lower mean ranks for EMG activity (RMS and MDF) and MU discharge rate (MU_DR) but higher mean ranks for MU number and MU action potential (MUAP) amplitude compared to non-MTrPs (p > 0.05). The artificial neural network (ANN) classifier achieved the highest performance for sample-wise cross-validation using the scaled dataset (82.35%). Although global explanations revealed MU_number as the lowest predictor, active MTrPs tend to show low RMS and MDF but elevated MUAP-amplitude features.
CONCLUSION: HD-sEMG is a promising method for identifying MTrPs. Machine learning methods, such as ANN-based classification, have demonstrated strong performance. However, further investigation is needed to improve classification accuracy for MTrP identification based on EMG decomposition.