Feature extraction for evaluation of Muscular Atrophy

Pal, Pabitra; Mohanty, Niladri Prasad; Kushwaha, Abhishek; Singh, Bikesh Kumar; Mazumdar, Bidyut; Gandhi, Tapan Kumar

doi:10.1109/iccic.2010.5705757

Cited by 16 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Shannon Entropy [62] En(x) = − ∑ N i=1 P i log(P i ), where x: discrete random variable, x i ∈ {x 1 , . .…”

Section: Features Definitionmentioning

confidence: 99%

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance

Hellara,

Barioul,

Sahnoun

et al. 2024

Sensors

View full text Add to dashboard Cite

Effective feature extraction and selection are crucial for the accurate classification and prediction of hand gestures based on electromyographic signals. In this paper, we systematically compare six filter and wrapper feature evaluation methods and investigate their respective impacts on the accuracy of gesture recognition. The investigation is based on several benchmark datasets and one real hand gesture dataset, including 15 hand force exercises collected from 14 healthy subjects using eight commercial sEMG sensors. A total of 37 time- and frequency-domain features were extracted from each sEMG channel. The benchmark dataset revealed that the minimum Redundancy Maximum Relevance (mRMR) feature evaluation method had the poorest performance, resulting in a decrease in classification accuracy. However, the RFE method demonstrated the potential to enhance classification accuracy across most of the datasets. It selected a feature subset comprising 65 features, which led to an accuracy of 97.14%. The Mutual Information (MI) method selected 200 features to reach an accuracy of 97.38%. The Feature Importance (FI) method reached a higher accuracy of 97.62% but selected 140 features. Further investigations have shown that selecting 65 and 75 features with the RFE methods led to an identical accuracy of 97.14%. A thorough examination of the selected features revealed the potential for three additional features from three specific sensors to enhance the classification accuracy to 97.38%. These results highlight the significance of employing an appropriate feature selection method to significantly reduce the number of necessary features while maintaining classification accuracy. They also underscore the necessity for further analysis and refinement to achieve optimal solutions.

show abstract

“…Shannon Entropy [62] En(x) = − ∑ N i=1 P i log(P i ), where x: discrete random variable, x i ∈ {x 1 , . .…”

Section: Features Definitionmentioning

confidence: 99%

Comparative Study of sEMG Feature Evaluation Methods Based on the Hand Gesture Classification Performance

Hellara,

Barioul,

Sahnoun

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…The k-NN classifier was used for classification and 100% classification accuracy was presented by the authors. Pal et al [8] used various features such as spectrogram, rootmean-square, entropy, and kurtosis for detection of the ALS from EMG signals. Merlo et al [9] proposed a fast and reliable method for EMG signal analysis.…”

Section: Introductionmentioning

confidence: 99%

Classification of Amyotrophic Lateral Sclerosis and Healthy Electromyography Signals Based on Transfer Learning

Şengür

Budak

Akbulut

2018

European Journal of Technic

View full text Add to dashboard Cite

This paper investigates the usage of transfer learning in amyotrophic lateral sclerosis (ALS) disease detection. ALS is a dangerous disease which affects the nerve cells in brain and spinal cord. Electromyogram (EMG) is an important measure for analysing of the electrical level of the muscles. EMG based early ALS disease detection system helps the physicians and patients. The proposed work uses EMG signals in discrimination of the ALS and healthy persons. The EMG signals are initially segmented with a overlapped window and each segment is converted to the spectrogram images. The obtained spectrogram images are resized and fed into the pre-trained convolutional neural networks model. The pre-trained model is fine-tuned with the problem at hand. The R002 dataset which is obtained from www.emglab.net is used during the experimental works. Accuracy, sensitivity and specificity measures are used to evaluate the obtained achievement. According to these measures, 97.70% accuracy, 97.97% sensitivity, and 97.29% specificity values are recorded. We further compare the obtained results with some of the existing results that were obtained on the same dataset. The comparisons show that proposed method is outperformed.

show abstract

“…Most of the parameters extracted from EMG signals are based on the time domain, frequency domain, and timefrequency domain. Time domain based features such as root mean square, spectrogram, kurtosis, entropy and power have been used for classification of ALS and healthy signals [9]. Mel-frequency cepstral coefficient based features have been used as input to K-nearest neighbourhood (KNN) classifier for classification of ALS and healthy EMG signals [10].…”

Section: Introductionmentioning

confidence: 99%