Claim Missing Document
Check
Articles

Found 1 Documents
Search
Journal : International Journal of Engineering and Emerging Technology

A Deep Learning Approach For COVID 19 Detection Via X-Ray Image With Image Correction Method I Gede Totok Suryawan; I Putu Agus Eka Darma Udayana
International Journal of Engineering and Emerging Technology Vol 5 No 2 (2020): July - December
Publisher : Doctorate Program of Engineering Science, Faculty of Engineering, Udayana University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24843/IJEET.2020.v05.i02.p018

Abstract

In the mitigation effort for reducing the spread of the SARS-CoV-2 pandemic in Indonesia, finding, detecting, and containing the suspect be a very crucial step to contain the virus. One of the ways that this can be detected is by thorax x-ray examination by the expert. Transferring the doctor's knowledge to a computer makes the task more scalable and precise. This can be done by building a small artificial intelligence using a simple CNN model to detect COVID biomarkers' presence in x-ray images. As the AI relies heavily on the x-ray dataset as the system's underlying basis has a good quality dataset is very important. However, the x-ray data tend to have a noise problem that will affect their overall system quality. We did a little comparative study with the objective to improve the quality of the dataset with three techniques of image enhancement, namely color denoising, mean denoising, and contrast enhancement, with the mean denoising outperform the other image manipulation method by 4%, which yield the accuracy of the system to 95% with 100 pieces of real-world test data. Hopefully, this study would inspire future studies improving the tech-based pandemic mitigation technology In the future.