Pengolahan Sinyal EMG Untuk Menentukan Intensitas Stimulasi Listrik Terapi TENS (Transcutaneous Electrical Nerve Stimulation) Pada Pasien Parkinson

Reawaruw, Timothy Manulus (2026) Pengolahan Sinyal EMG Untuk Menentukan Intensitas Stimulasi Listrik Terapi TENS (Transcutaneous Electrical Nerve Stimulation) Pada Pasien Parkinson. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5049221030-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (8MB) | Request a copy

Abstract

Penyakit Parkinson memiliki gejala tremor istirahat yang sangat mengganggu. Terapi Transcutaneous Electrical Nerve Stimulation terbukti efektif meredakan tremor, namun penentuan intensitasnya masih manual dan statis sehingga berisiko memicu respons kejut dan habituasi saraf. Penelitian ini mengembangkan model komputasional kendali adaptif berbasis sinyal Electromyography untuk menentukan dosis stimulasi secara aman dan responsif. Sinyal dari dataset sekunder diproses menggunakan filter Butterworth dan Discrete Wavelet Transform Level 7. Energi tremor diekstraksi melalui algoritma Moving Root Mean Square berjendela 500 milidetik dan dipetakan ke rentang aman 5 hingga 25 miliampere menggunakan Piecewise Linear Proportional Mapping. Pengujian komputasional pada 10 subjek membuktikan algoritma ini mandiri terhadap variasi pasien berkat penskalaan fitur Min-Max. Komparasi luaran terhadap referensi klinis dan tinjauan pakar membuktikan kepatuhan linieritas mutlak tanpa pergeseran fase, dengan latensi komputasi sangat rendah yakni 505 milidetik, sehingga sistem tervalidasi aman untuk diimplementasikan.
===========================================================================================================================================
Parkinson disease features resting tremors that disrupt daily activities. Transcutaneous Electrical Nerve Stimulation therapy is proven effective in reducing tremor, but intensity settings remain manual and static, risking startle reflexes and nerve habituation. This study develops an Electromyography signal based adaptive control computational model to determine stimulation dosage safely and responsively. Signals from a secondary dataset were processed using Butterworth filters and Level 7 Discrete Wavelet Transform. Tremor energy was extracted via a 500 milliseconds Moving Root Mean Square algorithm and mapped to a safe 5 to 25 milliampere range using Piecewise Linear Proportional Mapping. Computational testing on 10 subjects proved the algorithm is patient independent due to Min-Max feature scaling. Output validation against clinical references and expert judgment demonstrated absolute linear compliance with zero phase shift, achieving a very low computational latency of 505 milliseconds, validating the system as safe for implementation.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Parkinson, tremor, TENS, EMG, model komputasional, validasi in silico, Parkinson's disease, tremor, TENS, EMG, computational model, in silico validation
Subjects:	Q Science > Q Science (General) > Q180.55.M38 Mathematical models
Divisions:	Faculty of medicine and health (MEDICS) > Medical Technology > 11503-(S1) Undergraduate Thesis
Depositing User:	Timothy Manulus Reawaruw
Date Deposited:	04 Aug 2026 07:47
Last Modified:	04 Aug 2026 07:47
URI:	http://repository.its.ac.id/id/eprint/142416

Actions (login required)

View Item