SEGMENTASI DAN KLASIFIKASI PENYAKIT PARU-PARU MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK

ARNALDO, MUHAMMAD and Nurmaini, Siti and Satria, Hadipurnawan (2022) SEGMENTASI DAN KLASIFIKASI PENYAKIT PARU-PARU MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK. Master thesis, Sriwijaya University.

	Text RAMA_55101_09012682024002.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
	Text RAMA_55101_09012682024002_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
Preview	Text RAMA_55101_09012682024002_0002085908_0018048003_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB) | Preview
	Text RAMA_55101_09012682024002_0002085908_0018048003_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (608kB) | Request a copy
	Text RAMA_55101_09012682024002_0002085908_0018048003_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55101_09012682024002_0002085908_0018048003_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55101_09012682024002_0002085908_0018048003_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (177kB) | Request a copy
	Text RAMA_55101_09012682024002_0002085908_0018048003_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (146kB) | Request a copy
	Text RAMA_55101_09012682024002_0002085908_0018048003_07_Lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy

Abstract

Various kinds of lung diseases have become the main cause of death and the cause of many respiratory system complications that result in the death of millions of people every year. The use of medical images along with the deep learning methods can provide faster and more accurate detection of the disease. However, detection using medical images of the lungs also has challenges in the form of variations in the shape, size, and location of the infection area. In addition, multiple lung diseases often have similar symptoms. This study adopted a convolutional neural network (CNN) method with Mask-RCNN architecture to classify multi-class lung diseases. Eight models were employed with different configurations and the best performance was shown by the Mask-RCNN Resnet-50 model, learning rate 10-3, and learning cycle of 100 epochs. The evaluation matrix used shows the results of DSC, MIoU, Precision, recall, and accuracy of 91.98%, 85.25%, 98.84%, 98.86%, 99.47%, respectively. Testing of unseen data is also carried out to test the robustness of the model that has been built.

Item Type:	Thesis (Master)
Uncontrolled Keywords:	Mask-RCNN, Citra Medis, Penyakit Paru-paru, Segmentasi
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 55101-Informatics (S2)
Depositing User:	Muhammad Arnaldo
Date Deposited:	14 Sep 2022 08:06
Last Modified:	14 Sep 2022 08:06
URI:	http://repository.unsri.ac.id/id/eprint/78619

Actions (login required)

View Item