Development of an EMG-Based Muscle Health Model for Elbow Trauma Patients

Abstract

Wearable robotic braces have the potential to improve rehabilitative therapies for patients suffering from musculoskeletal (MSK) conditions. Ideally, a quantitative assessment of health would be incorporated into rehabilitative devices to monitor patient recovery. The purpose of this work is to develop a model to distinguish between the healthy and injured arms of elbow trauma patients based on electromyography (EMG) data. Surface EMG recordings were collected from the healthy and injured limbs of 30 elbow trauma patients while performing 10 upper-limb motions. Forty-two features and five feature sets were extracted from the data. Feature selection was performed to improve the class separation and to reduce the computational complexity of the feature sets. The following classifiers were tested: linear discriminant analysis (LDA), support vector machine (SVM), and random forest (RF). The classifiers were used to distinguish between two levels of health: healthy and injured (50% baseline accuracy rate). Maximum fractal length (MFL), myopulse percentage rate (MYOP), power spectrum ratio (PSR) and spike shape analysis features were identified as the best features for classifying elbow muscle health. A majority vote of the LDA classification models provided a cross-validation accuracy of 82.1%. The work described in this paper indicates that it is possible to discern between healthy and injured limbs of patients with MSK elbow injuries. Further assessment and optimization could improve the consistency and accuracy of the classification models. This work is the first of its kind to identify EMG metrics for muscle health assessment by wearable rehabilitative devices.

Modeling the assessment of the upper limb motor function impairment in children with cerebral palsy using sEMG and IMU sensors

“…The results suggest that these features are more efficient to identify muscle health than the more commonly used sEMG features. This is in accordance with (Abel et al, 1996;Farago, 2018;Haddara, 2016;Hogrel, 2005), who reported that features extracted from sEMG can distinguish between normal and abnormal muscle patterns. The proposed assessment method can serve as a potential approach to assess the effectiveness of Robot-AT and of other upper limb conventional therapies.…”

“…It was concluded that Shannon entropy could provide an accurate visual biofeedback for reduction of spasticity in patients with a stroke (Zadnia, Kobravi, Sheikh, Hosseini, & Neuroscience, 2018). Another research study found that implementing an EMG-based model of muscle health in a rehabilitative elbow brace has the potential of assessing patients recovering from Musculoskeletal (MSK) elbow trauma (Farago, 2018). Existing methodologies in pattern recognition suggest that EMG signals could be analyzed to detect muscle strength and motor unit (MU) recruitment as well as measure the effectiveness of rehabilitation therapies.…”

Sensor Fusion in Assistive and Rehabilitation Robotics

“…The fourth paper, by Farago et al [ 4 ], presents an electromyography-based muscle health model for elbow trauma patients. Surface electromyography recordings were collected from healthy and injured limbs of 30 elbow trauma patients during 10 different motions, and multiple classifiers were used to distinguish between healthy and injured states.…”

Wearable Active Electrode for sEMG Monitoring Using Two-Channel Brass Dry Electrodes with Reduced Electronics

“…Surface electromyography (sEMG) allows acquiring the EMG signals of the activity of the muscle fibers in a noninvasive way by placing electrodes in contact with the skin. In spite of the volume, the conductor constitutes an important low-pass filter on the EMG signal that intrinsically reduces the measured information [ 1 ]; these sEMG signals are still useful for medical diagnosis, prosthetics, rehabilitation devices [ 2 ], and sports medicine [ 3 ]. However, the amplitude of these signals, which is in the range of 50 μ V to 30 mV with a frequency bandwidth from 20 Hz to 450 Hz [ 4 ], represents an important concern for developing effective devices aimed at measuring the sEMG signal in long-term recordings without substantial noise and interferences.…”