DETEKSI PNEUMONIA PADA PASIEN COVID-19 BERDASARKAN CITRA X-RAY DADA MENGGUNAKAN METODE TRANSFER LEARNING
(1) Universitas YARSI Jakarta
(*) Corresponding Author
Abstract
ABSTRACT
The detection of pneumonia in Covid-19 patients has an important role because it can be used to determine the appropriate treatment for patients. This helps ensure that patients receive the highest quality of care possible. The use of a method based on machine learning is one of the approaches that can be taken. At the moment, developing a classification model that has a high level of accuracy is a difficult task. This is due to the fact that the X-ray image of the chest has a complicated structure. The purpose of this research is to evaluate an approach called performance transfer learning in order to determine whether or not a chest x-ray image of a patient with the covid-19 strain contains an infection of pneumonia. The findings indicate that, by utilizing a transfer learning architecture, one can achieve an accuracy of approximately 86.36%.
Keywords : Pneumonia, Detection, Transfer Learning, Covid-19.
Abstrak
Deteksi pneumonia pada pasien Covid-19 memiliki peran penting karena dapat digunakan untuk menentukan jenis perawatan yang tepat bagi pasien. Penggunaan metode deteksi berdasarkan citra x-ray berbebasis machine learning adalah salah satu cara yang dapat dilakukan saat ini, namun pada aplikasinya model machine learning yang dihasilkan belum optimal. Hal ini disebabkan karena citra x-ray memiliki struktur yang kompleks sehingga model machine learning dengan struktur yang sederhana tidak dapat menangkap pola pada citra x-ray dalam mendeteksi infeksi pneumonia pada pasien Covid-19. Riset ini bertujuan untuk mengevaluasi performa arsitektur model machine learning berbasis transfer learning untuk klasifikasi infeksi pneumonia berdasarkan citra x-ray pasien Covid-19. Studi ini menunjukkan bahwa, penggunaan beberapa arsitektur transfer learning menghasilkan tingkat akurasi hingga 84,36%.
Kata Kunci : Pneumonia, Deteksi, Transfer Learning, Covid-19
Keywords
Full Text:
PDFArticle Metrics
Abstract view : 197 timesPDF - 35 times
DOI: https://doi.org/10.26714/me.v15i2.10731
Refbacks
- There are currently no refbacks.
Copyright (c) 2022 MEDIA ELEKTRIKA
Editorial Office of Media Elektrika
Universitas Muhammadiyah Semarang FT-FMIPA Building, 7nd Floor. Phone: 085299398663 |
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.