Classification of brain tumors using GoogleNet feature set and machine learning

Subject Areas : BioEngineering

Mahdi Eslami ¹ (Assistant Professor at Department of Electrical and Computer Engineering, Science and Research Branch Islamic Azad University, Tehran, Iran)
SANA golmarziasasl ² (Department of Electrical and Computer Engineering, Science and Research Branch Islamic Azad University, Tehran, Iran)

Received: 2024-02-04 Accepted : 2024-04-12 Published : 2024-07-14

Keywords: neural networks, deep learning, Alex Net, ResNet, VGG16,

Abstract :

In healthcare research, the Internet of Medical Things (IoMT) is transforming how healthcare operates and introducing a new era of the internet. IoMT enables computer-aided diagnosis (CAD) systems, storing health data online and providing patients with valuable information and support. Connected smart devices communicate over the Internet, enabling patients to communicate with medical professionals through IoMT-based the care systems, especially for critical conditions such as brain tumors, which are often precursors to cancer with low survival rates. Are. Early tumor detection and classification is crucial to save human lives, and IoMT-enabled CAD systems are emerging as indispensable solutions. Deep learning, especially Convolutional Neural Net-works (CNN), has gained a lot of interest in this field in recent years. In this research, we classify most common three types of brain tumors, namely, glioma, meningioma, pituitary and use AlexNet, GoogleNet, ResNet18, and VGG16 networks to check their correct diagnosis.