A Density-Based Clustering Approach to Motor Unit Potential Characterizations to Support Diagnosis of Neuromuscular Disorders

Artificial intelligence for automatic classification of needle EMG signals: A scoping review

OBJECTIVE

This scoping review provides an overview of artificial intelligence (AI), including machine and deep learning techniques, in the interpretation of clinical needle electromyography (nEMG) signals.

METHODS

A comprehensive search of Medline, Embase and Web of Science was conducted to find peer-reviewed journal articles. All papers published after 2010 were included. The methodological quality of the included studies was assessed with CLAIM (checklist for artificial intelligence in medical imaging).

RESULTS

51 studies were identified that fulfilled the inclusion criteria. 61% used open-source EMGlab data set to develop models to classify nEMG signal in healthy, amyotrophic lateral sclerosis (ALS) and myopathy (25 subjects). Only two articles developed models to classify signals recorded at rest. Most articles reported high performance accuracies, but many were subject to bias and overtraining.

CONCLUSIONS

Current AI-models of nEMG signals are not sufficient for clinical implementation. Suggestions for future research include emphasizing the need for an optimal training and validation approach using large datasets of clinical nEMG data from a diverse patient population.

SIGNIFICANCE

The outcomes of this study and the suggestions made aim to contribute to developing AI-models that can effectively handle signal quality variability and are suitable for daily clinical practice in interpreting nEMG signals.

Can Wavelet Denoising Improve Motor Unit Potential Template Estimation?

Background:

Electromyographic (EMG) signals obtained from a contracted muscle contain valuable information on its activity and health status. Much of this information lies in motor unit potentials (MUPs) of its motor units (MUs), collected during the muscle contraction. Hence, accurate estimation of a MUP template for each MU is crucial.

Objective:

To investigate the possibility of improving MUP template estimation using the wavelet denoising technique.

Material and Methods:

In this analytical study, several MUP template estimators were developed by combining conventional estimation methods and wavelet denoising techniques. A MUP template was initially estimated using conventional methods such as mean, median, median-trimmed mean, or mode. Thereafter, it was post-processed using the wavelet denoising technique.

Results:

Evaluation results of the studied estimators using 40 simulated EMG signals with a true template for each constituent MUP train showed that augmented wavelet- based template estimation methods are more reliable than conventional methods. However, on average, wavelet denoising was not much effective. Around 40 MUPs of a MU is sufficient to estimate its MUP template.

Conclusions:

Although wavelet techniques are effective in EMG signal analysis, here wavelet denoising did not practically improve MUP template estimation. Considering computational simplicity and estimation error, the two methods median and median-trimmed mean are practical estimators that can provide a good estimation of a MUP template for a MU when approximately 40 MUPs are available. Nevertheless, the baseline noise level in the MUP templates estimated using the median-trimmed mean method is slightly lower than that in the templates estimated using the median method.

Transparent Electrophysiological Muscle Classification From EMG Signals Using Fuzzy-Based Multiple Instance Learning

Although a well-established body of literature has examined electrophysiological muscle classification methods and systems, ways to enhance their transparency is still an important challenge and requires further study. In this work, a transparent semi-supervised electrophysiological muscle classification system which uses needle-detected EMG signals to classify muscles as normal, myopathic, or neurogenic is proposed. The electrophysiological muscle classification (EMC) problem is naturally formulated using multiple instance learning (MIL) and needs an adaptation of standard supervised classifiers for the purpose of training and evaluating bags of instances. Here, a novel MIL-based EMC system in which the muscle classifier uses predictions based on motor unit potentials (MUPs) to infer muscle labels is described. This system uses morphological, stability, near fiber and spectral MUP features. Quantitative results obtained from applying the proposed transparent system to four electrophysiologically different groups of muscles, composed of proximal and distal hand and leg muscles, resulted in an average classification accuracy of 95.85%. The findings show the superior and stable performance of the proposed EMC system compared to previous works using other supervised, semi-supervised and unsupervised methods.

The motor unit and quantitative electromyography

Electromyography (EMG) assesses the anatomic motor unit (A-MU), but knowledge of its anatomy, physiology, and changes with pathology is limited. The electrophysiological motor unit (E-MU) and its motor unit potential (E-MUP) represents a fraction of the A-MU. Routine EMG assesses a limited number of E-MUP waveform characteristics (metrics) and their magnitudes qualitatively scaled in a nonlinear manner. Another approach is quantitative EMG (QEMG), whereby 20⁺ E-MUPs are extracted and both basic and derived metrics obtained and values expressed quantitatively. In diseased muscle, many E-MUP metrics may be normal, which complicates diagnostic interpretation. In QEMG, E-MUP metrics can be clustered and statistical analyses performed to assign probabilities that E-MUPs (and the muscle) are normal, neuropathic, or myopathic. In this article we review what is known about the A-MU, the restricted E-MU, E-MUP metrics, and what QEMG offers currently and in the future.

Electrophysiological Muscle Classification Using Multiple Instance Learning and Unsupervised Time and Spectral Domain Analysis

OBJECTIVE

Electrophysiological muscle classification (EMC) is a crucial step in the diagnosis of neuromuscular disorders. Existing quantitative techniques are not sufficiently robust and accurate to be reliably clinically used. Here, EMC is modeled as a multiple instance learning (MIL) problem and a system to infer unsupervised motor unit potential (MUP) labels and create supervised muscle classifications is presented.

METHODS

The system has five main steps: MUP representation using morphological, stability, and near fiber parameters as well as spectral features extracted from wavelet coefficients; MUP feature selection using unsupervised Laplacian scores; MUP clustering using neighborhood distance entropy consistency to find representations of MUP normality and abnormality; muscle representation by embedding its MUP cluster associations in a feature vector; and muscle classification using support vector machines or random forests.

RESULTS

The evaluation data consist of 63, 83, 93, and 84 sets of MUPs recorded in deltoid, vastus medialis, first dorsal interosseous, and tibialis anterior muscles, respectively. The proposed system discovered representations of normal, myopathic, and neurogenic MUPs for each specific muscle type and resulted in an average classification accuracy of 98%, which is higher than in previous works.

CONCLUSION

Modeling EMC as an instance of the MIL solves the traditional problem of characterizing MUPs without full supervision. Furthermore, finding representations of MUP normality and abnormality using morphological, stability, near fiber, and spectral features improve muscle classification.

SIGNIFICANCE

The proposed method is able to characterize MUPs with respect to disease categories, with no a priori information.