Investigation of Brain Tumors Detection using Automatic Segmentation Techniques

One of the most crucial steps in the medical imaging is to achieve the accuracy, as relying on the modal’s performance for the patient’s treatment is very risky. The risks go higher in context with the brain tumors, which are also known as intracranial tumors. Apart from the several diagnostic techniques, Magnetic Resonance Images have been used for extraction of the finer details of the brain tumor and have been used till date. In this paper, the analysis of various techniques has been performed in order to conclude for the best technique that can be used for the Content-Based Image Retrieval (CBIR) for brain tumors and its various categories. The analysis will also be useful, in order to remove the semantic gap, between the low-level features and high-level features that are obtained by the model used and the manual diagnostics respectively.