Brain Tumor Detection Using Deep Learning

Abstract

The Using medical imaging to segment brain tumors
is a particularly challenging but crucial process. This
is because of the potential for erroneous prognosis and
diagnosis as a result of manual classification. Working
with vast volumes of data also requires a lot of work.
Because of their similar appearances and wide range
of traits, It might be difficult to tell a brain tumor
originating from healthy tissue in photographs. In this
study, fuzzy C-Means was used to remove brain tumors
from two-dimensional MRI images. CNN and classical
classifiers were then used.The study’s dataset included
tumors of various sizes, shapes, and intensities. KNN,
MLP, LR, SVM, and Naive.The typical classifier of the
scikit-learn module included popular machine learning
methods including Bayes, Random Forest, and others.A
brain tumor is a condition brought on by aberrant brain
cell proliferation. Brain tumors can be classified as either
benign or malignant, with the former being classified as a
cancerous brain tumor. It is difficult to forecast the survival
rate of a patient who has a tumor because of the rarity
and variety of tumor kinds. Fifteen out of one hundred
individuals with brain cancer had a possibility of surviving
for ten years or longer after diagnosis, per a UK cancer
study. The type of tumor, the location of the tumor in
the brain, the abnormalities of the cells, and Additional
elements all influence how a patient with a brain tumor is
treated.