Analisis Ekstraksi Fitur HRV Menggunakan Wearable Continuous Wave Radar Untuk Sistem Deteksi Fibrilasi Atrial Paroksismal

Putri, Vasya Maharani (2025) Analisis Ekstraksi Fitur HRV Menggunakan Wearable Continuous Wave Radar Untuk Sistem Deteksi Fibrilasi Atrial Paroksismal. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 5023211033-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only Download (6MB) | Request a copy

Abstract

Penelitian ini mengembangkan sistem deteksi non-kontak untuk mengenali onset Paroxysmal Atrial Fibrillation (pAF) berdasarkan ekstraksi fitur Heart Rate Variability (HRV)
dari sinyal jantung yang diperoleh menggunakan radar Continuous Wave (CW). Sensor radar K-LC7 digunakan untuk menangkap pergerakan mikro dinding dada yang diproses melalui metode arctangent demodulation dan penyaringan sinyal berbasis Ensemble Empirical Mode Decomposition (EEMD). Deteksi puncak dilakukan menggunakan pendekatan zero-crossing dan threshold adaptif untuk menghasilkan beat-to-beat interval (BBI) sebagai dasar ekstraksi HRV. Sebanyak 12 fitur HRV diekstraksi dari domain spektral, bispektral, dan non-linear, kemudian dibandingkan terhadap fitur dari sinyal ECG untuk menilai kesesuaian karakteristiknya. Hasil menunjukkan bahwa fitur HRV dari radar memiliki tren yang sebanding dengan ECG, dengan rata-rata korelasi SDNN sebesar 53,1% dan RMSSD sebesar 54,2%. Sistem klasifikasi menggunakan Support Vector Machine (SVM) dengan akurasi 80%,
sensitivitas 85,7%, dan spesifisitas 75% dalam membedakan episode normal dan pAF. Meski sistem menunjukkan performa klasifikasi yang menjanjikan, keterbatasan pada akurasi deteksi puncak dan stabilitas performa Raspberry Pi menjadi tantangan yang perlu diperbaiki, termasuk potensi pengembangan dengan pendekatan deteksi frekuensi dan perangkat keras yang lebih efisien untuk mendukung aplikasi wearable.
========================================================================================================================================
This study developed a non-contact system to detect the onset of Paroxysmal Atrial Fibrillation (pAF) based on Heart Rate Variability (HRV) feature extraction from cardiac
signals captured using Continuous Wave (CW) radar. The K-LC7 radar sensor was utilized to detect micro-movements of the chest wall, which were processed using arctangent
demodulation and signal filtering via Ensemble Empirical Mode Decomposition (EEMD). Peak detection was performed using a zero-crossing and adaptive thresholding method to extract the beat-to-beat interval (BBI), forming the basis for HRV analysis. Twelve HRV features were extracted across spectral, bispectral, and non-linear domains and compared to ECG-based HRV features to evaluate their similarity. Results showed that HRV trends from radar signals aligned reasonably with those from ECG, with average correlation values of 53.1% for SDNN and 54.2% for RMSSD. The classification system using a Support Vector Machine (SVM) achieved 80% accuracy, 85.7% sensitivity, and 75% specificity in distinguishing normal and pAF episodes. Despite its promising classification performance, the system still faces
limitations in peak detection accuracy and processing speed, particularly due to computational constraints on the RaspberryPi. Future development should consider frequency-based detection approaches and more efficient embedded platforms to support a real wearable implementation.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Radar Continuous Wave, Heart Rate Variability, Fibrillasi Atrial Paroksismal, Arctangent Demodulation, Ensemble Empirical Mode Decomposition, Support Vector Machine
Subjects:	R Medicine > R Medicine (General) > R856.2 Medical instruments and apparatus. T Technology > T Technology (General) > T59.7 Human-machine systems. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5102.9 Signal processing. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK6564 Radio transmitter-receivers
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Biomedical Engineering > 11410-(S1) Undergraduate Thesis
Depositing User:	Vasya Maharani Putri
Date Deposited:	04 Aug 2025 01:28
Last Modified:	04 Aug 2025 01:28
URI:	http://repository.its.ac.id/id/eprint/125657

Actions (login required)

View Item