Description

Magnetic Resonance Images are used in detecting and tracking brain tumors. Diagnostic brain MRI scans are usually performed by trained medical technologists who manually prescribe the position and orientation of a scanning volume. T2-weighted scans use a Spin Echo (SE) sequence, with long TE (Echo time) and long TR (repetition time). On a T2-weighted scan, water- and fluid-containing tissues are bright and fat containing tissues are dark. Medical image segmentation plays an important role in medical research field. The self-organizing map (SOM) intrinsically identifies structure and patterns in a high dimensional dataset such as a text corpus, or collection. Algorithmically, the SOM reduces the dimensional space of the collection, which is called quantization, and clusters it accordingly, creating observable relationships. This process is achieved through training the map over much iteration, and while the SOM is algorithmically robust, it lacks the capability to expand, or further cluster, once a map has been generated. In our paper, we propose image segmentation using SELF-ORGANIZING MAP (SOM). Here this process gives better result than other segmentation algorithm.