The Proposed UNet-based system for segmenting brain tumors in MRI images

It increases the patient's chances of recovery and medical options if a brain tumor is detected and diagnosed early. Brain tumors are detected and diagnosed using magnetic resonance imaging (MRI). In clinical practice, the identification of brain tumors from a large number of MRI images depends entirely on time and experience. Medical diagnosis and treatment recommendations can be facilitated with computer-aided expert systems. There are a variety of frameworks based on machine learning and deep learning that are used to detect brain tumors. Deep learning techniques are used in this paper to develop an efficient framework for segmenting and classifying brain tumors. MRI images were volumetrically segmented using the ۳D-UNet model, and then CNN's were used to classify the tumors. For the purpose of establishing the validity of the models, loss and precision diagrams are presented. Six different experiments with different parameters were performed on the Network, and their results were compared. The results show that the proposed method is more effective than the existing methods.

کلیدواژه ها:

Convolutional neural network ، Magnetic Resonance Images ، U-Net network ، brain tumor ، BraTs ۲۰۱۸