Application of Machine Learning Algorithms for Carpal Tunnel Syndrome

Abstract

Introduction: Median nerve entrapment is commonly related to alterations in the anatomy of the surrounding tissues in the carpal tunnel. Carpal tunnel syndrome (CTS) is the most prevalent form of peripheral entrapment neuropathy. Machine learning (ML) is utilized in a variety of fields. After conferring with a physician, ML enables doctors to conduct the necessary examinations and make an early diagnosis. Methods based on artificial intelligence have the potential to be utilized in clinical practice as a supplementary instrument for accurate evaluation of median nerve entrapment. Despite the rise in ML-based medical research, median nerve entrapment has received less attention. The purpose of this study was to evaluate the performance of classification approaches with ML algorithms in CTS patients utilizing electromyography test data from patients exhibiting varied CTS symptoms and indications. Materials and Methods: Our study includes message and demographic information derived from the electromyography results of 315 individuals. In classification procedures, the logistic regression, support vector machine (SVM), K-nearest neighbor, and naïve Bayes algorithms from ML techniques were utilized. Results: As a result of the classification, performance values for accuracy, precision, sensitivity, specificity, and F1-score were obtained. As a result of our research, the SVM algorithm achieved a 96% success rate. Conclusion: ML algorithms are an emerging method of analysis. The diagnosis and treatment of diseases are gradually gained by clinicians through observation and experience. Therefore, categorization systems can aid in the accurate and timely diagnosis of median nerve entrapment by clinicians.