Claim Missing Document
Check
Articles

Found 1 Documents
Search

Deteksi Hipoksia Berdasarkan Detak Jantung, Saturasi Oksigen, Volume Dan Irama Pernafasan Menggunakan Metode K-Nearest Neighbor Leina Alimi Zain; Rizal Maulana; Fitri Utaminingrum
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 1 (2021): Januari 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Symptoms of hypoxia are a condition caused by a lack of oxygen in the cells and tissues of the body and this condition can cause damage to the nerves of the brain, liver and other organs which will lead to death. The use of technology in the medical field has created a system to detect hypoxic symptoms using the K-Nearest Neighbor method. The detection system using the K-Nearest Neighbor method can be carried out in knowing the condition of a person's body without injuring the body or it is called non-invasive. Retrieval of heart rate and oxygen saturation data using the MAX30100 sensor by placing the index finger on the red LED component and the IR photodiode component. It takes 20 seconds and the finger must not move during the take to get the optimal value. In taking the volume and rhythm of breathing is done using a Flex sensor. The hardware used is the Arduino Mega, the MAX30100 sensor and the Flex sensor. The level of accuracy on 10 tests on the MAX30100 sensor is 97.07% and the accuracy level obtained on the Flex sensor is 92.77%. In classifying using the K-Nearest Neighbor method, there is a level of accuracy at the k = 3 value of 90% k = 5 by 80% and k = 7 by 70% and there is a computational average of 3.37 ms in 10 tests.