Rancang Bangun Sistem Pengukuran Laju Detak Jantung Berbasis Photopletyhsmography Menggunakan Webcam

Gultom, Yonatan Davidson (2021) Rancang Bangun Sistem Pengukuran Laju Detak Jantung Berbasis Photopletyhsmography Menggunakan Webcam. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text 02311740000083-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (5MB) | Request a copy

Abstract

Kondisi tubuh dan lingkungan dapat mempengaruhi laju detak jantung seseorang. Monitoring laju detak jantung secara konsisten dapat membantu mendeteksi berbagai penyakit yang berhubungan dengan jantung. Metode konvensional biasanya melibatkan pengukuran secara kontak seperti ECG, atau pulse oximetry. Pada penelitian ini, dikembangkan sebuah algoritma untuk mengukur laju detak jantung secara non-kontak berbasis photoplethysmography (PPG). Region of Interest yang digunakan adalah daerah wajah penuh, daerah kening dan pipi, serta daerah di bawah kening hingga di atas mulut (2/3 tinggi wajah). Metode ekstraksi PPG yang digunakan adalah Independent Component Analysis (ICA) dan Plane Orthogonal to Skin (POS). Algoritma yang dirancang diujikan terhadap dataset UBFC-RPPG yang tersedia bagi publik untuk keperluan benchmarking. Dibandingkan kondisi subjek lainnya, sistem menunjukkan performansi paling baik pada kondisi subjek diam. Dibandingkan ROI lainnya, ROI 2/3 tinggi wajah menunjukkan nilai paling baik dengan dengan MAE = 6.586 bpm, RMSE = 9.804 bpm, STD = 7.267 bpm, dan r = 0.195. Dibandingkan metode ekstrasi ICA, metode ekstraksi POS menunjukkan nilai paling baik dengan MAE = 4.940, RMSE = 7.446, STD = 5.508, dan r = 0.215.
=====================================================================================================
Body conditions and the environment can affect a person's heart rate. Consistent heart rate monitoring can help detect various heart-related diseases. Conventional methods usually involve contact measurements such as ECG, or pulse oximetry. In this study, an algorithm was developed to measure heart rate non-contact based on photoplethysmography (PPG). The Region of Interest used is the fullface area, the forehead and cheek areas, and the area below the forehead to above the mouth (2/3 face height). The PPG extraction methods used are Independent Component Analysis (ICA) and Plane Orthogonal to Skin (POS). The designed algorithm was tested against the publicly available UBFC-RPPG dataset for benchmarking purposes. Compared to other subject conditions, the system shows the best performance in still subject conditions. Compared to other ROI, ROI of 2/3 face height showed the best value with MAE = 6,586 bpm, RMSE = 9,804 bpm, STD = 7,267 bpm, and r = 0.195. Compared to the ICA extraction method, the POS extraction method showed the best value with MAE = 4,940, RMSE = 7,446, STD = 5,508, and r = 0.215.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Pengukuran Detak Jantung Non-Kontak, Dataset UBFC-RPPG, Independent Component Analysis, Plane-Orthogonal to Skin Non-contact Heart Rate Measurement, UBFC-RPPG Dataset, Independent Component Analysis, Plane-Orthogonal to Skin
Subjects:	R Medicine > RZ Other systems of medicine
Divisions:	Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis
Depositing User:	Yonatan Davidson Gultom
Date Deposited:	01 Sep 2021 01:40
Last Modified:	01 Sep 2021 01:40
URI:	http://repository.its.ac.id/id/eprint/90999

Actions (login required)

View Item