Identification Of Treatment Efficacy Of Chiropractors using sEMG signal Classification

Abstract

In the domain of Pain management, chiropractic treatment is a developing research field. Although chiropractic intervention-based treatment is a prominent field of study, however, there is a significant gap between its authenticity and efficacy. To address that the suggested approach is directed toward 64-channel High-density surface electromyography (HDSEMG) signals to evaluate changes before and after a single respective session of treatment. The system suggests different types of machine learning technologies, among them being Support Vector Machine (SVM), Random Forest, Decision Tree, and Light Gradient Boosting Machine (LGBM), that help separate control from intervention treatments by their patterns and features used in High-Density Surface Electromyography (HD-SEMG) signals. This requires pre-processing signals to remove unnecessary noise, extracting features using TSFRESH for statistical analysis and signal decomposition, and then classifying them using the various selected machine learning algorithms. This helps to pick out the patterns, and methods to differentiate the two classifications of treatment. The system’s performance is assessed based on metrics such as accuracy, F1 score, precision, and recall. Results show that the proposed methodology correctly identifies treatments, signaling its potential to enhance the credibility of chiropractic treatment. This method has the potential to improve and make good calls on healthcare decision-making processes.