Klasifikasi MRI Scan Tumor Otak Menggunakan Metode CNN-ELM

Ferdiansyah, Sulthan Ahmad (2024) Klasifikasi MRI Scan Tumor Otak Menggunakan Metode CNN-ELM. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 20081010156.-Cover.pdf Download (1MB)
	Text (Bab 1) 20081010156.-Bab 1.pdf Download (190kB)
	Text (Bab 2) 20081010156.-Bab 2.pdf Restricted to Repository staff only until 30 October 2026. Download (706kB) \| Request a copy
	Text (Bab 3) 20081010156.-Bab 3.pdf Restricted to Repository staff only until 30 October 2026. Download (565kB) \| Request a copy
	Text (Bab 4) 20081010156.-Bab 4.pdf Restricted to Repository staff only until 30 October 2026. Download (1MB) \| Request a copy
	Text (Bab 5) 20081010156.-Bab 5.pdf Download (184kB)
	Text (Daftar Pustaka) 20081010156.-Daftar Pustaka.pdf Download (190kB)

Abstract

This study compares the performance of Convolutional Neural Network (CNN) model and CNN-Extreme Learning Machine (CNN-ELM) hybrid model in brain tumor image classification. The evaluation was conducted on five tests CNN and CNN-ELM with variations in the number of hidden nodes ELM 500, 1000, 2000, 3000, and 4000. The results showed that the CNN with 0,001 learning rate and 16 batch size achieved highest accuracy with 96,57%. CNN-ELM model with 2000 hidden nodes achieved the highest accuracy of 98.1%, slightly higher than CNN which achieved 96.2% accuracy. Although the CNN model showed better consistency with a smaller accuracy difference, the CNN-ELM with 3000 nodes showed the best balance between accuracy and consistency. However, adding or subtracting too many nodes can decrease consistency and affect computation time if the number of nodes is too large. The CNN-ELM model, although providing higher accuracy, requires longer training time because it must go through the CNN training process first for feature extraction. In conclusion, CNN-ELM hybrid can improve accuracy in classification, but the trade-off between the number of hidden nodes, performance, consistency, and training time efficiency needs to be considered.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Rahmat, Basuki	NIDN0023076907	basukirahmat.if@upnjatim.ac.id
Thesis advisor	Anggraeny, Fetty Tri	NIDN0711028201	fettyanggraeny.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Unnamed user with email 20081010156@student.upnjatim.ac.id

Date Deposited:

31 Oct 2024 05:36

Last Modified:

31 Oct 2024 05:36

URI:

https://repository.upnjatim.ac.id/id/eprint/31736