Event Detection of Muscle Activation Using an Electromyogram

Abstract

In this study, we proposed a precise onset and offset detection algorithm for muscle activation by using an electromyogram (EMG). The adaptive threshold was determined using the constant false alarm rate algorithm. The EMG signal was refined by morphological hole filling, which is used to close up and fill out missing information. By exploiting the EMG amplitude ratio in two channels, we significantly improved the offset detection performance. The proposed method does not require a training process, unlike conventional methods. The experimental results indicated that the estimated errors for both the onset and offset detection are lower than those obtained using two of the conventional methods.

Wavelet-Based Assessment of the Muscle-Activation Frequency Range by EMG Analysis

“…A recent and exciting study proposes an onset and offset detection algorithm for muscle activation validated during hand-close movement of 10 subjects [27]. It is worth noticing that the outcomes of the present study are at least comparable, in terms of MAE ± SD, with those reported in [27], although the algorithm is tested on a different dataset and in a different experimental condition. These results further support the reliability of the present algorithm in detecting muscle onset in time domain.…”

“…Thus, the concomitance of these results allows asserting that the capability of the two algorithms in detecting onset/offset events in simulated sEMG signals is comparable. Moreover, it is worth reminding that the analysis of the algorithm performances in time domain is achieved not to demonstrate that the proposed algorithm is able to outperform the other approaches in assessing onset and offset events in time domain, but just to show that the CWT algorithm is reliable in time domain since it can provide an assessment of onset and offset events at least comparable with reference algorithm reported in the literature [23,27,39,46,47]. The actual aim of the present study is, indeed, to propose an algorithm able to provide further and new information, that is, the frequency content of every single activation detected in time domain, as highlighted in the following section C.…”

“…The choice of this dataset has been driven by the following considerations: it is a large and complete sEMG dataset available for free; it includes a reliable ground truth for onset detection, which is already used and tested in previous studies [19,44,47]; the peculiarities of knee extension and elbow flexion allow a simple and accurate visual detection of the muscle onset and, consequently, reliable identification of the ground truth; results of the application to this dataset of four onset-detection techniques and two filters are available for a direct comparison [44]. A recent and exciting study proposes an onset and offset detection algorithm for muscle activation validated during hand-close movement of 10 subjects [27]. It is worth noticing that the outcomes of the present study are at least comparable, in terms of MAE ± SD, with those reported in [27], although the algorithm is tested on a different dataset and in a different experimental condition.…”

“…DT is even now considered among the most reliable detection techniques; DT is indeed broadly used both in research and in clinics, and it is also implemented in commercial systems [25,26]. A more recent approach adopts an adaptive threshold computed by the constant false alarm rate (CFAR) algorithm, where the EMG signal is refined by morphological hole filling used to close up and fill out missing information [27].…”

Comparison of sEMG Onset Detection Methods for Occupational Exoskeletons on Extensive Close-to-Application Data

“…This characteristic can be used to eliminate the time latencies of the exoskeleton control loop. In contrast to other sEMG analysis methods, onset detection has low demands towards computational power and robustness of measurements, due to its low complexity [4,20,21].…”