Brain lesion tumor segmentation analysis based onMRI images using deep learning

An abnormal accumulation of cells in the brain is called a brain tumor. The skull, which surrounds thebrain, is very strong. Any growth in such a limited space leads to problems .The tumor may becancerous (malignant) or non-cancerous (benign).With the growth of a benign or malignant tumor,intracranial pressure increases .In this case, the brain is damaged, which is fatal. Diagnostic images,including an MRI (or CT scan), are obtained to confirm the presence of a tumor, and, if present, toassess its location, size, and effect on surrounding tissue. Magnetic resonance imaging (MRI) tests arecommonly used to help diagnose brain tumors. Sometimes a dye is injected through a patient's armvein during an MRI study. Brain tumor segmentation is one of the most important tasks in the field ofmedical image processing. Early diagnosis of brain tumors plays an important role in the possibility ofimprovement with treatment and increasing the survival rate of patients. Manual segmentation ofbrain tumors for cancer diagnosis (by humans) from the large number of MRI images generated inroutine medicine is a difficult and time-consuming task .There is a fundamental need for automatedbrain tumor image segmentation .The purpose of this article is to provide a review on MRI-basedbrain tumor image segmentation methods .Recently, the use of deep learning methods for automaticsegmentation has become popular because these methods achieve new and advanced results andcan handle this problem better than other methods. Deep learning methods can also enable efficientprocessing and enable objective and visual evaluation of large volumes of MRI-based image data.

Brain lesion tumor segmentation analysis based onMRI images using deep learning Keywords:

Tumor , MRA , tumor diagnosis , brain damage.