Alifa, Ramadhani Putri (2023) Estimasi Kelelahan Berdasarkan Analisa EMG dan Heart Rate Pada Latihan Treadmill Menggunakan Machine Learning. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
07311940000030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2025. Download (2MB) | Request a copy |
Abstract
Kelelahan otot adalah fenomena umum yang membatasi performa atletik dengan terjadi penurunan kekuatan maksimal. Sebanyak 8 dari 10 pelari mengatakan pernah mengalami cedera karena intensitas latihan yang tidak tepat saat berlari. Oleh karena itu, diperlukan sistem penilaian untuk memonitor dan mengukur kelelahan tersebut. Penelitian ini berfokus pada estimasi kelelahan dalam latihan inkremental treadmill dengan menggunakan nilai heart rate reserve (%HRR) saat aktivitas maksimum dan sinyal elektromiografi (EMG) pada quadriceps yang mengalami kelelahan. Integrasi modul EMG Click dan instrumentasi heart rate dilakukan dengan mikrokontroler untuk mengklasifikasikan tingkat kelelahan pada seseorang berdasarkan karakteristik sinyal EMG dan detak jantung. Metode klasifikasi menggunakan Multi-Layer Perceptron Neural Network (MLPNN) dengan input berupa nilai intensitas %HRR dan empat fitur EMG yang berhasil diekstraksi, yaitu Mean Power Frequency (MPF), Root Mean Square (RMS), Variance, dan Energy. Analisis data dari 9 subjek menunjukkan bahwa sebagian besar otot yang mengalami kelelahan memiliki amplitudo yang lebih tinggi dan frekuensi yang lebih rendah. Pemantauan detak jantung menggunakan modul MAX30102 memberikan akurasi yang lebih baik (RMSE = 1,549 bpm) dibandingkan dengan penggunaan Pulsebar pada treadmill Kettler RUN 11 (RMSE = 14,201 bpm). Model ini berhasil mendeteksi kelelahan pada otot dan detak jantung dengan akurasi mencapai 95,45% dan loss sebesar 17,28%.
=================================================================================================================================
Muscle fatigue is a common phenomenon that limits athletic performance, resulting in a decrease in maximum strength. Eight out of ten runners report having experienced injuries caused by improper training intensity while running. Therefore, a monitoring and measurement system is needed to assess this fatigue. This study focuses on estimating fatigue mechanisms during incremental treadmill exercises using the heart rate reserve (%HRR) at maximum activity and electromyography (EMG) signals from fatigued quadriceps muscles. The integration of the EMG Click module and heart rate instrumentation is carried out with a microcontroller to classify fatigue levels in individuals based on the characteristics of the EMG signals and heart rate. The classification method uses a Multi-Layer Perceptron Neural Network (MLPNN) with inputs consisting of the %HRR intensity value and four extracted EMG features: Mean Power Frequency (MPF), Root Mean Square (RMS), Variance, and Energy. Data analysis from nine subjects shows that most fatigued muscles have higher amplitude and lower frequency. Heart rate monitoring using the MAX30102 module yields better accuracy (RMSE = 1.549) compared to using the Pulse bar on the Kettler RUN 11 treadmill (RMSE = 14.201). This model successfully detects muscle fatigue and heart rate with an achieved accuracy of 95.45% and a loss of 17.28%.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | EMG, Heart Rate Reserve, Kelelahan Otot, Latihan Treadmill, Muscle Fatigue, Treadmill Exercise. |
Subjects: | R Medicine > RC Internal medicine > RC1200 Sports Medicine T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7871.674 Detectors. Sensors T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7878 Electronic instruments |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis |
Depositing User: | Ramadhani Putri Alifa |
Date Deposited: | 05 Oct 2023 08:30 |
Last Modified: | 05 Oct 2023 08:30 |
URI: | http://repository.its.ac.id/id/eprint/102813 |
Actions (login required)
View Item |