Deteksi Kelelahan Otot Menggunakan Elektromiografi Dengan Analisis Continuous Wavelet Transform Dan Convolutional Neural Network

Bakti, Nismat Hajjim Ayu Permata (2025) Deteksi Kelelahan Otot Menggunakan Elektromiografi Dengan Analisis Continuous Wavelet Transform Dan Convolutional Neural Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023211001-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (2MB) | Request a copy

Abstract

Kelelahan otot adalah kondisi fisiologis yang ditandai dengan penurunan kemampuan otot dalam menghasilkan gaya, dan dapat memberikan dampak negatif terhadap performa fisik serta meningkatkan risiko cedera. Kondisi ini dapat terjadi apabila otot bekerja melebihi kapasitas normal, baik dalam aktivitas sehari-hari, olahraga, maupun aktivitas berulang yang membebani jaringan otot. Penelitian mengenai kelemahan otot banyak dikembangkan untuk memahami bagaimana otot bekerja agar dapat meningkatkan kualitas hidup, performa altet, bahkan pemulihan kondisi pasien dalam bidang rehabilitasi medis. Deteksi kelelahan otot yang lebih objektif dapat dilakukan menggunakan sinyal biologis tubuh elektromiografi (EMG). Penelitian ini mengusulkan sistem deteksi kelelahan otot berbasis sinyal EMG menggunakan analisis Continuous Wavelet Transform (CWT) dan klasifikasi dengan model Convolutional Neural Network (CNN) yang digabungkan dengan Multilayer Perceptron (MLP). Data EMG diambil dari otot flexor digitorum profundus (FDP) pada enam subjek sebelum dan setelah melakukan olahraga handgrip. Analisis dilakukan dalam tiga domain, yaitu domain waktu dan domain frekuensi menggunakan ekstraksi fitur, serta pada domain waktu-frekuensi menggunakan CWT. Kemudian, dikembangkan model pembelajaran mesin CNN dan MLP untuk melakukan klasifikasi kelelahan otot. Validasi kelelahan otot dilakukan dengan menggunakan Rate of Perceived Exertion (RPE) dan penurunan kekuatan genggaman tangan sekitar 12,5% yang diukur dengan menggunakan dinamometer. Hasil penelitian menunjukkan bahwa sistem ini mampu mengklasifikasi kondisi kelelahan otot dan kondisi normal dengan akurasi sebesar 83% menggunakan kombinasi gambar dua dimensi CWT dengan skala 50 dan enam ekstraksi fitur dari domain waktu dan domain frekuensi.
==========================================================================================================================================
Muscle fatigue is a physiological condition characterized by a decrease in muscle ability to generate force, which can negatively impact physical performance and increase the risk of injury. This condition can occur when muscles work beyond their normal capacity, whether in daily activities, sports, or repetitive activities that strain muscle tissue. Research on muscle fatigue has been conducted to understand how muscles function and improve quality of life, athlete performance, and medical rehabilitation. Muscle fatigue detection can be done more objectively by electromyography (EMG) on biomedical signals. This research proposes an EMG signal-based muscle fatigue detection system utilizing Continuous Wavelet Transform (CWT) analysis and classification with a Convolutional Neural Network (CNN) model combined with Multilayer Perceptron (MLP). EMG data were taken from the flexor digitorum profundus (FDP) muscle in six subjects before and after performing handgrip exercise. The analysis will be performed in three domains, which are the time domain and frequency domain using feature extraction, as well as in the time-frequency domain using CWT. Then, CNN and MLP machine learning models are developed to perform muscle fatigue classification. Validation of muscle fatigue is done using Rate of Perceived Exertion (RPE) and decreased hand grip strength of around 12.5% measured using a dynamometer. The results show that this system can classify muscle fatigue and normal conditions with an accuracy of 83% using a combination of two-dimensional CWT images and six feature extractions from the time domain and frequency domain.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Continuous Wavelet Transform, Convolutional Neural Network, Elektromiografi, Kelelahan Otot, Electromyography, Muscle Fatigue
Subjects:	T Technology > T Technology (General) > T57.5 Data Processing T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.585 TCP/IP (Computer network protocol)
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Nismat Hajjim Ayu Permata Bakti
Date Deposited:	04 Aug 2025 04:58
Last Modified:	04 Aug 2025 04:58
URI:	http://repository.its.ac.id/id/eprint/125422

Actions (login required)

View Item