Analysis of chest X-Ray (CXR) images in COVID-19 patients based on age using the Otsu thresholding segmentation method

Uhty Maesyaroh (Universitas Islam Negeri Walisongo Semarang)
Laelatul Munawaroh (Universitas Islam Negeri Walisongo Semarang)
Heni Sumarti (Universitas Islam Negeri Walisongo Semarang)
Rico Adrial (Universitas Andalas)

Journal of Natural Sciences and Mathematics Research
Vol 7, No 2 (2021): December

The infection with the COVID-19 virus or better known as the Corona virus spread throughout China and other countries around the world until it was designated a pandemic by the World Health Organization (WHO). Detection of patients infected with COVID-19 in the form of RT-PCR, CT-Scan images and Chest X-Ray (CXR). This study aims to analyze CXR images of COVID-19 patients based on age using Otsu Thresholding Segmentation. The image segmentation process uses the Otsu auto-tresholding method to separate objects from the background on the CXR image. The results show that the images of COVID-19 patients have pneumonia spots that are not visible on the original CXR image. The average value of the accuracy of the Otsu Thresholding results is 95.18%. Penunomia spots are mostly found in COVID-19 patients aged 50 to 70 years and over which cause severe lung damage.©2021 JNSMR UIN Walisongo. All rights reserved.