Research indexJournal article
Brain Tumor Classification Using Deep Learning Framework
Dubey, Thosar, Mitra and 2 others
16 Sep 2022
NobleIDark:/48914/w1/20260524/A58A836ADOI10.1109/r10-htc54060.2022.9929922
Brain Tumor Classification Using Deep Learning Framework
Dubey, Thosar, Mitra and 2 others · 2022 · Subscription
NobleIDark:/48914/w1/20260524/A58A836ADOI10.1109/r10-htc54060.2022.9929922
The full text is at the publisher
You can read the abstract here, or open it at the publisher.
Abstract
Brain tumor is the mass or abnormal growth of a cells in the brain. Various types of brain tumors exist. The ability of a radiologist to diagnose a brain tumor early is mostly dependent on his or her experience. As a result, it is vulnerable to human error. It is critical to build an effective diagnostics tool for tumor detection and classification to acquire exact diagnostics and prevent surgery and subjectivity, Although surgery is the most common treatment for brain tumors, tumors that cannot be surgically removed may be treated with radiation or chemotherapy to limit their growth. One of the most popular techniques used to diagnose brain tumors is magnetic resonance imaging (MRI), which produces detailed images of the brain. In this paper, a technique is proposed for classification of brain tumor as glioma, meningioma and pituitary tumor using the framework of convolutional neural network. CNN architecture RESNET with a skip connection is used in this framework. T1 weighted (Axial View) real MR images from hospital and simulated images from Kaggle dataset are used for experimentation.