KLASIFIKASI TUBERKULOSIS PADA CITRA CHEST X-RAY MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK

SEPTIANI, ANNISA DWI and Rini, Dian Palupi and Miraswan, Kanda Januar (2025) KLASIFIKASI TUBERKULOSIS PADA CITRA CHEST X-RAY MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK. Undergraduate thesis, Sriwijaya University.

[thumbnail of RAMA_55201_09021182126015_COVER.pdf] Image
RAMA_55201_09021182126015_COVER.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (210kB)
[thumbnail of RAMA_55201_09021182126015.pdf] Text
RAMA_55201_09021182126015.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (4MB)
[thumbnail of RAMA_55201_09021182126015_TURNITIN.pdf] Text
RAMA_55201_09021182126015_TURNITIN.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (6MB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_01_front_ref.pdf.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_01_front_ref.pdf.pdf - Accepted Version
Available under License Creative Commons Public Domain Dedication.
Download (855kB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_02.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_02.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (597kB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_03.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_03.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (420kB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_04.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_04.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (1MB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_05.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_05.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (1MB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_06.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_06.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (13kB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_07_ref.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_07_ref.pdf - Bibliography
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (192kB)
[thumbnail of RAMA_55201_09021182126015_0023027804_0009019002_08_lamp.pdf] Text
RAMA_55201_09021182126015_0023027804_0009019002_08_lamp.pdf - Accepted Version
Restricted to Repository staff only
Available under License Creative Commons Public Domain Dedication.
Download (11kB)

Abstract

Tuberculosis is still one of the infectious diseases that causes high mortality rates worldwide. Early detection using Chest X-Ray images is very important in the effort to treat this disease. However, manual interpretation of X-ray results performed by doctors can cause differences in diagnosis. This study aims to classify tuberculosis in Chest X-Ray images by implementing three Convolutional Neural Network (CNN) architectures, namely VGG-19, ResNet-50, and DenseNet-121. This study uses a dataset of 7,000 images consisting of 3,500 normal images and 3,500 tuberculosis images. Experiments were conducted using 12 test scenarios with a combination of hyperparameter learning rate (0.01 and 0.001) and batch size (32 and 64) on each architecture. The results showed that the best model was achieved by VGG-19 architecture with a combination of learning rate 0.001 and batch size 32, which produced the highest performance with accuracy values of 99.57%, precision 99.71%, recall 99.43%, and F1-score 99.57%. This research proves that proper CNN implementation with optimal configuration of hyperparameters can be an effective tool in detecting tuberculosis through Chest X-Ray images.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Tuberculosis Classification, Chest X-Ray, Convolutional Neural Network, VGG-19, ResNet-50, DenseNet-121
Subjects: Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions: 09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User: Annisa Dwi Septiani
Date Deposited: 03 Jun 2025 02:34
Last Modified: 03 Jun 2025 02:34
URI: http://repository.unsri.ac.id/id/eprint/175011

Actions (login required)

View Item View Item