Klasifikasi Citra MRI Tumor Otak Menggunakan Metode Hibrida CNN-ViT

Sukandar, Ivan Christopher (2024) Klasifikasi Citra MRI Tumor Otak Menggunakan Metode Hibrida CNN-ViT. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010086-cover.pdf Download (710kB)
	Text (Bab 1) 20081010086-bab1.pdf Download (241kB)
	Text (Bab 2) 20081010086-bab2.pdf Restricted to Repository staff only until 30 May 2026. Download (972kB)
	Text (Bab 3) 20081010086-bab3.pdf Restricted to Repository staff only until 30 May 2026. Download (605kB)
	Text (Bab 4) 20081010086-bab4.pdf Restricted to Repository staff only until 30 May 2026. Download (2MB)
	Text (Bab 5) 20081010086-bab5.pdf Download (226kB)
	Text (Daftar Pustaka) 20081010086-daftarpustaka.pdf Download (241kB)
	Text (Lampiran) 20081010086-lampiran.pdf Restricted to Repository staff only Download (338kB)

Abstract

A tumor is an abnormal cell that can grow in any part of the human body. This also includes the brain which is the most important organ for humans. Brain tumors can arise because a cell that should grow and die within a certain period of time remains alive and multiplies abnormally. Brain tumors require fast and accurate medical diagnosis because a patient needs to get immediate treatment. The Convolutional Neural Network (CNN) algorithm is one of the most popular algorithms for image data processing and analysis and has been incorporated into medical image classification. In addition, there is also a Vision Transformer algorithm that divides the image into several patches and tokenizes each pixel of the image. Based on this explanation, the author classifies brain tumor MRI images using the CNN-ViT hybrid method. The result of this research is the accuracy and performance of CNN-ViT hybrid compared to CNN and ViT in classifying brain tumors. The best accuracy results were obtained by CNN-ViT with an average test accuracy of 93%, CNN with an average test accuracy of 90.80% and ViT with an average test accuracy of 84.80%. In addition, the classification report results of the best scenario CNN-ViT obtained with data division 80:10:10, Adam optimization and learning rate 0.0001% are accuracy of 94%, precision of 95%, recall of 94% and f1 score of 94%.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Anggraeny, Fetty Tri	19820211 2021212 005	fettyanggraeny.if@upnjatim.ac.id
Thesis advisor	Swari, Made Hanindia Prami	19890205 2018032 001	madehanindia.fik@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ivan Christopher Sukandar

Date Deposited:

24 Jun 2024 07:57

Last Modified:

24 Jun 2024 07:57

URI:

https://repository.upnjatim.ac.id/id/eprint/23589