Abstract - In a variety of medical diagnostic applications, automated flaw identification in medical imaging has become a hot topic. Automated tumor detection in MRI is vital because it provides information about abnormal tissues, which is crucial for therapy planning. Human inspection is the traditional approach for detecting defects in magnetic resonance brain imaging. Due to the vast volume of data, this strategy is impractical. As a result, reliable and automatic classification techniques are required to reduce the human fatality rate. Because of the intricacy and variety of tumors, MRI brain tumor identification is a difficult undertaking. In this study, we propose employing machine learning methods to overcome the limitations of traditional classifiers in the detection of tumors in brain MRI.

Key Words: MRI, Brain tumor, Convolution Neural Network, Segmentation

PDF | DOI: 10.17148/IJARCCE.2021.101227

Open chat
Chat with IJARCCE