Emg Based Diagnosis of Myopathy and Neuropathy Using Machine Learning Techniques (original) (raw)
Related papers
EMG Signal Classification with Effective Features for Diagnosis
Advances in Intelligent Systems and Computing, 2020
Electromyography (EMG) signals are broadly used in various clinical or biomedical applications, prosthesis or rehabilitation devices, Muscle-Computer Interface (MCI), Evolvable Hardware Chip (EHW) development and many other applications. Electromyography (EMG) signal records the myopathy from nonlinear subjects in both time domain and frequency domain. It becomes very difficult to classify these various statuses. In this paper, a feature extraction and classification method of healthy and myopathy EMG signals are proposed where two features have been extracted on both healthy and myopathy EMG. Mean Squared Error (MSE) has been calculated to observe which feature will give better classification result. Then SVM is used to classify the extracted results. To evaluate the proposed model, a standard dataset collected from physionet.org is used where it shows higher accuracy than the conventional methods.
Techniques of EMG signal analysis and classification of Neuromuscular diseases
Artificial intelligence techniques are being used effectively in medical diagnostic tools to increase the diagnostic accuracy and provide additional knowledge. Electromyography (EMG) signals are becoming increasingly important in clinical and biomedical applications. Detection, processing and classification of EMG signals are very desirable because it allows a more standardized evaluation to discriminate between different neuromus-cular diseases. This paper reviews a brief explanation of the different features extraction and classification techniques for classifying EMG signals used in literatures. Wavelet Transform (WT), Principle Component Analysis (PCA), and Independent Component Analysis (ICA) are different feature extraction techniques. Literature presents different techniques to classify EMG data such as probabilistic neural network (PNN), Support Vector Machine (SVM), Artificial Neural Networks (ANN), etc. In this paper neuromuscular disease classification from electromyography (EMG) signals are proposed based on different combination of features extraction methods and types of classifiers. Combination of WT and SVM improved the classification accuracy than other combinations such as DWT with ANN, ICA with MLPN, PCA with ANN and DWT with PNN. Keywords: Electromyogram (EMG), Wavelet Transform (WT), Principle Component Analysis (PCA), Support Vector Machine (SVM), Multilayer Perceptron Neural Network (MLPNN), Probabilistic Neural Network (PNN).
Automatic Classification of Intramuscular EMG to Recognize Pathologies
2019
This paper proposes to assess the relevance of new automated tools for electromyography (EMG) analysis, in order to differentiate neuropathic from myopathic patterns. The challenge is to define the diagnosis with only one iEMG signal per patient. Our proposed method uses the decomposition of the EMG signal to characterize motor unit action potentials (MUAPs). The decomposition of each iEMG signal is carried out with EMGLAB. For each signal, the decomposition provides a code which is used by the automated classification algorithms.We use here the linear Support Vector Machine (SVM) and the Bagging Trees methods. For the learning process we use several EMG signals and in different parts of the muscle. Only one recorded electromyography EMG signal per subject is used for the diagnostic test. We evaluate the k—fold cross-validation and the confusion matrix for both models. The accuracy is 77.3% for the SVM and 68.2% for the Bagging Trees. These are the first developments of this tool to...
Classification of EMG Signals for Assessment of Neuromuscular Disorders
International Journal of Electronics and Electrical Engineering, 2014
An accurate and computationally efficient means of feature extraction of electromyographic (EMG) signal patterns has been the subject of considerable research effort in recent years. Quantitative analysis of EMG signals provides an important source of information for the classification of neuromuscular disorders. The objective of this study is to discriminate between normal (NOR), myopathic (MYO) and neuropathic (NEURO) subjects. The experiment consisted of 22 pathogenic (11 MYO and 11 NEURO) and 12 healthy persons. The signals were recorded at 30% Maximum Voluntary Contraction (MVC) for 5 seconds. Features of MUAPs extracted in time have been quantitatively analysed. We have used binary SVM for classification. Separation of normal subjects from neuromuscular disease subjects has an accuracy of 83.45%, whereas separation of subjects from the two types of subjects (myopathic and neuropathic) has an accuracy of 68.29% which is again high.
Classification and Diagnosis of Myopathy from EMG Signals
We present a methodology to predict the presence of myopathy (muscle disease) from intramuscular electromyography (EMG) signals. By evaluating the shape and frequency of electrical action potentials produced by muscular fibers and captured in EMG measurements, a physician can often detect both the presence and the severity of such disorders. However, EMG measurements can vary significantly across different subjects, different muscles, and according to session-specific characteristics such as muscle fatigue and degree of contraction. By considering fixed-duration (0.5-2 sec) frequency-domain samples of diagnostic regions in EMG signals measured at full muscle contraction, we can automatically detect the presence of myopathies across different subjects and muscles with ~90% accuracy. We argue that our methodology is more generally applicable than existing methods that depend upon accurate segmentation of individual motor unit action potential (MUAP) waveforms. We present a rigorous evaluation of our technique across several different subjects and muscles.
Review on Feature Extraction and Classification of Neuromuscular Disorders
IJMTST, 2019
Electromyography is an efficient tool for the diagnosis of neuromuscular diseases. There are wide variety of neuromuscular diseases that affects the muscles and nervous system, in which the most important are Amyotrophic Lateral Sclerosis (ALS) and Myopathy. These diseases change the shape and characteristic of motor unit action potentials (MUAPs). By analyzing the EMG signals and MUAPs neuromuscular diseases can be diagnosed. This paper gives a brief review of various techniques used in the analysis of EMG signals for the diagnosis of neuromuscular diseases. Various features that are extracted from the signals in time domain, frequency domain and time-frequency domain and different classification techniques and their performance are also studied in this paper.
Characterizing EMG Data using Machine- Learning Tools
Comput Biol Med. , 2014
Effective electromyographic (EMG) signal characterization is critical in the diagnosis of neuromuscular disorders. Machine-learning based pattern classification algorithms are commonly used to produce such characterizations. Several classifiers have been investigated to develop accurate and computationally efficient strategies for EMG signal characterization. This paper provides a critical review of some of the classification methodologies used in EMG characterization, and presents the state-of-the-art accomplishments in this field, emphasizing neuromuscular pathology. The techniques studied are grouped by their methodology, and a summary of the salient findings associated with each method is presented.
Procedia Computer Science, 2017
Electromyography (EMG) signals is usable in order to applications of biomedical, clinical, modern human computer interaction and Evolvable Hardware Chip (EHW) improvement. Advanced methods are needed for perception, disassembly, classification and processing of EMG signals acquired from the muscles. Objective of this article is to show various methods and algorithms in order to analyze an electromyogram signal to ensure effective and efficient ways of understanding signal and its nature. Early diagnosis was indispensable and very important in medical health practice. For this reason, it is important to design accurate diagnostic methods. Today, diagnostic methods include evaluating the patient's story, blood tests, and muscle biopsies. In this article, analysis and Electromyogram signals classification and electromyography are mostly used. System has been successfully implemented utilizing MATLAB software that can distinguish EMG signals from different patients. This article also provides the researcher with a well understanding of electromyogram signalling and analysis processes. This information will auxiliary to improve stronger, more resilient and effective implementations.
Robust Classification of Intramuscular EMG Signals to Aid the Diagnosis of Neuromuscular Disorders
IEEE Open Journal of Engineering in Medicine and Biology
This article presents the design and validation of an accurate automatic diagnostic system to classify intramuscular EMG (iEMG) signals into healthy, myopathy, or neuropathy categories to aid the diagnosis of neuromuscular diseases. Methods: First, an iEMG signal is decimated to produce a set of "disjoint" downsampled signals, which are decomposed by the lifting wavelet transform (LWT). The Higuchi's fractal dimensions (FDs) of LWT coefficients in the subbands are computed. The FDs of LWT subband coefficients are fused with one-dimensional local binary pattern derived from each downsampled signal. Next, a multilayer perceptron neural network (MLPNN) determines the class labels of downsampled signals. Finally, the sequence of class labels is fed to the Boyer-Moore majority vote (BMMV) algorithm, which assigns a class to every iEMG signal. Results: The MLPNN-BMMV classifier was experimented with 250 iEMG signals belonging to three categories. The performance of the classifier was validated in comparison with state-of-the-art approaches. The MLPNN-BMMV has resulted in impressive performance measures (%) using a 10-fold cross-validation-accuracy = 99.87 ± 0.25, sensitivity (normal) = 99.97 ± 0.13, sensitivity (myopathy) = 99.68 ± 0.95, sensitivity (neuropathy) = 99.76 ± 0.66, specificity (normal) = 99.72 ± 0.61, specificity (myopathy) = 99.98 ± 0.10, and specificity (neuropathy) = 99.96 ± 0.14-surpassing the existing approaches. Conclusions: A future research direction is to validate the