Investigation of VGG-16, ResNet-50 and AlexNet Performance for Brain Tumor Detection

Tun Azshafarrah Ton Komar Azaharan, Tun Azshafarrah Ton Komar Azaharan and Abd Kadir Mahamad, Abd Kadir Mahamad and Sharifah Saon, Sharifah Saon and Muladi, Muladi and Sri Wiwoho Mudjanarko, Sri Wiwoho Mudjanarko (2023) Investigation of VGG-16, ResNet-50 and AlexNet Performance for Brain Tumor Detection. Investigation of VGG-16, ResNet-50 and AlexNet Performance for Brain Tumor Detection, 19 (8). pp. 97-109.

Text J16367_86588b65d522261076f05e67e66ec666.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Malignant brain tumours are extremely frequent and deadly, and if they are not found in their early stages, they can shorten a person’s lifespan. After the tumour has been detected, it is essential to classify the tumour in order to develop a successful treatment strategy. This study aims to investigate the three deep learning tools, VGG-16 ResNet50 and AlexNet in order to detect brain tumor using MRI images. The results performance are then evaluated and compared using accuracy, precision and recall criteria. The dataset used contained 155 MRI images which are images with tumors, and 98 of them are nontumors. The AlexNet model perform extremely well on the dataset with 96.10% accuracy, while VGG-16 achieved 94.16% and ResNet-50 achieved 91.56%. The early diagnosis of cancers before they develop physical side effects like paralysis and other problems is positively impacted by these accuracy.

Item Type:	Article
Uncontrolled Keywords:	brain tumor, classification, Vgg-16, Reset-50, AlexNet
Subjects:	T Technology > T Technology (General)
Divisions:	Faculty of Electrical and Electronic Engineering > Department of Electrical Engineering
Depositing User:	Mr. Mohamad Zulkhibri Rahmad
Date Deposited:	03 Jan 2024 01:36
Last Modified:	03 Jan 2024 01:36
URI:	http://eprints.uthm.edu.my/id/eprint/10554

Actions (login required)

View Item