Deep Transfer Learning Networks for Brain Tumor Detection: The Effect of MRI Patient Image Augmentation Methods

The exponential growth of deep learning networks has enabled us to handle difficult tasks, even in the complex field of medicine with small datasets. In the sphere of treatment, they are particularly significant. To identify brain tumors, this research examines how three deep learning networks are a...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:International Journal of Electronics and Communications System Jg. 2; H. 2; S. 39 - 48
Hauptverfasser: Abdalla, Peshraw Ahmed, Qadir, Abdalbasit Mohammed, Rashid, Omed Jamal, Rawf, Karwan M. Hama, Abdulrahman, Ayub O, Mohammed, Bashdar Abdalrahman
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Universitas Islam Negeri Raden Intan Lampung 25.12.2022
Schlagworte:
artificial intelligence
cancer detection
deep learning
densenet201
mobilenetv2
mri
vgg19
ISSN:2798-2610, 2798-2610
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The exponential growth of deep learning networks has enabled us to handle difficult tasks, even in the complex field of medicine with small datasets. In the sphere of treatment, they are particularly significant. To identify brain tumors, this research examines how three deep learning networks are affected by conventional data augmentation methods, including MobileNetV2, VGG19, and DenseNet201. The findings showed that before and after utilizing approaches, picture augmentation schemes significantly affected the networks. The accuracy of MobileNetV2, which was originally 85.33 percent, was then enhanced to 96.88 percent. The accuracy of VGG19, which was 77.33 percent, was then enhanced to 95.31 percent, and DenseNet201, which was originally 82.66 percent, was then enhanced to 93.75 percent. The models' accuracy percentage engagement change is 13.53 percent, 23.25 percent, and 23.25 percent, respectively. Finally, the conclusion showed that applying data augmentation approaches improves performance, producing models far better than those trained from scratch.
ISSN:2798-2610
2798-2610
DOI:10.24042/ijecs.v2i2.14815