Abstract:
Brain is the most essential organs, which controls a number of complex functions in the human
body. However, brain tumor is the abnormal growth of cells within the brain that affects humans
badly. It is currently one of the leading reasons of increased mortality in both children and
adults. Magnetic resonance imaging (MRI) has been successful in identifying a variety of
diseases related to the brain, particularly tumors. Earlier identification of tumors from brain
MRI images has recently achieved significant importance and is considered a lifesaver for brain
tumor patients. Nevertheless, brain tumor classification is crucial, it is equally important to
know the type of tumors to increase the patient survival rate and suggest proper treatment.
Therefore, manual segmentation and classification of brain tumors from magnetic resonance
(MR) images is a challenging, time and labor-consuming task. Today, MRI is the most popular
medical imaging techniques for treating brain tumors because of its non-invasive (no ionizing
radiation) nature. Moreover, the effort of the research community to come-up with automatic
brain tumor segmentation and classification method has been tremendous. As a result, there are
sample literatures on the area focusing on segmentation and classification using machine learning and deep learning algorithms on public data set. However, there is no related works found
on local dataset. In this study, hence, an attempt is made to construct a classification model
and develop a prototype for brain tumor segmentation and classification using CNN on local
MR images. The artifact was designed using a python module called flask and deployed on a
cloud-based framework called Heroku.
Data was collected from Nisir and Nolot specialized and general clinics at Bahir Dar. The data
was pre-processed to get quality images that are suitable for a deep-learning algorithm to develop a model that predicts tumor type accurately. Out of the to
