Peningkatan Performansi Klasifikasi Penderita Covid-19 Berdasarkan Bangkitan Citra X-Ray Paru-paru Menggunakan Metode Siamese Generative Adversarial Network.

Hamzah, Hafidzudin Muhammad (2022) Peningkatan Performansi Klasifikasi Penderita Covid-19 Berdasarkan Bangkitan Citra X-Ray Paru-paru Menggunakan Metode Siamese Generative Adversarial Network. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 06111840007001-Undergraduate_Thesis.pdf Restricted to Repository staff only Download (12MB) | Request a copy

Abstract

Pandemi penyakit Coronavirus 2019 (COVID-19) menjadi perhatian hampir diseluruh dunia dengan ditemukannya beberapa varian baru yang menyebabkan meningkatnya jumlah kematian. Salah satu cara untuk mendeteksi penderita COVID-19 adalah dengan menganalisis citra X-Ray paru-paru menggunakan metode Convolutional Neural Network (CNN). Dalam hal ini, semakin besar data yang digunakan pada pelatihan metode CNN, semakin baik hasil yang didapat. Namun, proses dalam mendapatkan sejumlah data yang besar pada penelitian ini diatasi dengan menggunakan metode Siamese Generative Adversarial Network (SiGAN). Model SiGAN digunakan untuk membangkitkan citra X-Ray paru-paru guna mengatasi ketidakseimbangan data pada dataset. Citra yang dihasilkan kemudian diklasifikasikan menggunakan metode DenseNet. Hasil pengujian menunjukkan bahwa penggunaan model SiGAN mampu meningkatkan akurasi klasifikasi secara signifikan dibandingkan dengan hanya menggunakan data asli. Penelitian ini memberikan kontribusi dalam pengembangan sistem deteksi COVID-19 yang lebih cepat dan akurat berbasis citra medis.
==============================================================================================================================
The COVID-19 pandemic has become a global concern with the discovery of several new variants that have led to an increase in the number of deaths. One way to detect COVID-19 patients is by analyzing chest X-ray images using the Convolutional Neural Network (CNN) method. In this case, the larger the data used in the CNN method training, the better the results obtained. However, the process of obtaining a large amount of data in this research is overcome by using the Siamese Generative Adversarial Network (SiGAN) method. The SiGAN model is used to generate chest X-ray images to overcome data imbalance in the dataset. The resulting images are then classified using the DenseNet method. The test results show that the use of the SiGAN model is able to significantly improve classification accuracy compared to only using the original data. This research contributes to the development of a faster and more accurate COVID-19 detection system based on medical images.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	COVID-19. Siamese Generative Adversarial Network. DenseNet. COVID-19. Siamese Generative Adversarial Network. DenseNet.
Subjects:	Q Science > QA Mathematics
Divisions:	Faculty of Mathematics and Science > Mathematics > 44201-(S1) Undergraduate Thesis
Depositing User:	Mr. Marsudiyana -
Date Deposited:	10 Jun 2026 01:08
Last Modified:	10 Jun 2026 01:08
URI:	http://repository.its.ac.id/id/eprint/133662

Actions (login required)

View Item