A benign brain tumor, characterized by the growth of non-cancerous brain cells, has the potential to undergo malignant transformation. Gliomas, the most prevalent primary brain tumors, originate from the development of abnormal glial cells in the brain and spinal cord. Magnetic resonance imaging (MRI), a widely accepted non-invasive imaging method, can detect brain tumors, and offer various tissue contrasts using diverse imaging modalities. Traditionally, the laborious and time-consuming task of manually segmenting and analyzing structural MRI data of brain tumors was the domain of neuroradiologists. However, recent advancements in comprehensive and automated segmentation methods for brain tumors are poised to greatly improve both cancer detection and treatment. BRATS 2015 is a dataset used for the segmentation of brain tumor images, consisting of 200 MRIs depicting high-grade gliomas and 44 MRIs displaying low-grade gliomas. The proposed aims to develop an automated deep learning model for detecting and classifying brain cancers using MRI data. Extensive simulations were carried out on the BRATS dataset to evaluate the enhanced performance in categorizing brain tumors. Thanks to progress in computer-aided design, machine learning, and deep learning techniques, we now have the capability to identify cancerous lesions in medical images.
Identification of Brain Tumors in MR Images Using UNet CNN Model
2023-11-22
304998 byte
Conference paper
Electronic Resource
English
RAO‐UNet: a residual attention and octave UNet for road crack detection via balance loss
DOAJ | 2022
|RAO‐UNet: a residual attention and octave UNet for road crack detection via balance loss
Wiley | 2022
|