Deteksi Kelelahan Otot Menggunakan Sinyal Elektromyogram Berbasis Artificial Neural Network

Susila, Stevanno Gamaliel Krisnata Eka (2025) Deteksi Kelelahan Otot Menggunakan Sinyal Elektromyogram Berbasis Artificial Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023211030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (4MB)

Abstract

Kelelahan otot adalah kondisi ketika otot manusia kehilangan kemampuannya untuk menghasilkan gaya maksimal setelah aktivitas fisik berkepanjangan. Kondisi kelelahan otot dapat menyebabkan penurunan performa fisik hingga meningkatkan risiko cedera. Dalam berolahraga, kelelahan otot yang tidak terdeteksi dapat menyebabkan overtraining dan cedera yang serius. Di dunia medis sekalipun, pemantuan kelelahan otot banyak dibutuhkan dalam rehabilitasi dan terapi fisik untuk menyesuaikan beban latihan pasien yang aman dan efektif. Metode pemantauan aktivitas otot biasanya dilakukan menggunakan sinyal elektromiografi (EMG). Sinyal ini merupakan sinyal listrik yang dihasilkan oleh otot rangka. EMG umumnya memiliki bentuk yang sangat kompleks, non-linier dan sangat bervariasi tergantung kondisi fisiologis maupun jenis aktivitas yang dilakukan. Salah satu metode efektif yang dapat membantu proses analisis sinyal EMG adalah metode ekstraksi fitur. Metode ini berfokus untuk menganalisis karakteristik sinyal EMG pada domain waktu dan frekuensi dan dapat mengidentifikasi secara akurat perubahan sinyal otot. Dalam penelitian ini, juga digunakan metode machine learning dengan algoritma Artificial Neural Network (ANN) untuk melakukan pengenalan pola dan klasifikasi sinyal kompleks. Penelitian ini diharapkan mampu membangun sistem deteksi kelelahan otot menggunakan data sEMG yang dianalisis dengan metode ekstraksi fitur dan diolah melalui algoritma ANN. Sistem ini tidak hanya bermanfaat untuk dunia olahraga dan kebugaran, tetapi juga memiliki potensi aplikasi dalam bidang klinis seperti rehabilitasi dan terapi fisik. Didapatkan model dengan nilai akurasi sebesar 85% untuk mendeteksi kondisi otot saat melakukan kontraksi.
======================================================================================================================================
Muscle fatigue is a condition in which human muscles lose their ability to generate maximal force after prolonged physical activity. This fatigue can lead to decreased physical performance and increase the risk of injury. In sports, undetected muscle fatigue can result in overtraining and serious injuries. Even in the medical field, monitoring muscle fatigue is widely needed in rehabilitation and physical therapy to adjust patients’ exercise load safely and effectively. The method commonly used to monitor muscle activity is through electromyography (EMG) signals. These signals are electrical outputs generated by skeletal muscles. EMG signals typically have highly complex, nonlinear patterns that vary greatly depending on physiological conditions and the type of activity being performed. One effective method to aid in EMG signal analysis is feature extraction. This method focuses on analyzing the characteristics of EMG signals in the time and frequency domain and can accurately identify changes in muscle activity. In this study, a machine learning method using an Artificial Neural Network (ANN) algorithm is also employed to perform pattern recognition and classification of these complex signals. This research aims to develop a muscle fatigue detection system using sEMG data, which is analyzed with feature extraction techniques and processed through an ANN algorithm. Such a system would not only be beneficial in sports and fitness but also has potential applications in clinical fields like rehabilitation and physical therapy. The final model reached an accuracy of 85% for detecting muscle condition while doing contraction.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Kelelahan Otot, Sinyal Electromyogram, Ekstraksi Fitur, Artificial Neural Network, Muscle fatigue, Electromyogram Signals, Features Extraction
Subjects:	T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Stevanno Gamaliel Krisnata Eka Susila Susila
Date Deposited:	04 Aug 2025 07:37
Last Modified:	07 Aug 2025 01:18
URI:	http://repository.its.ac.id/id/eprint/127100

Actions (login required)

View Item