The human brain is a remarkable part of the human nervous system, which regulates all bodily functions. Brain tumors develop from an abnormal mass of cells within the brain, and timely detection of these tumors often provides a wider range of treatment options. Identifying the infected area in brain tumor, MRI images is a crucial task since brain tumors are several illnesses resulting from an irregular cell growth in the brain. Early detection of tumors enhances patients' chances of recovery. The significance of precise brain tumor detection in brain MR (Magnetic Resonance) images cannot be overstated when medical image analysis and interpretation. Numerous techniques have been discussed in the past, but there remains a necessity to enhance the precision of the outcomes. Hence, an individual's chances of survival are significantly enhanced by early diagnosis and prompt treatment of brain tumors. Thus, this paper introduces automatic brain tumor identification using novel self organizing maps. In the context of medical image segmentation, a method employing the Self-Organizing Map (SOM) algorithm is presented. The SOM algorithm is instrumental in clustering, and the segmentation procedure plays a partitioning the preprocessed input image to extract characteristics like texture, color, shape, and intensity features. The application of the watershed algorithm in segmentation provides precise outline locations. For classification, an unsupervised neural network, called self-organizing maps, is utilized. The evaluation of this technique is based on metrics such as Accuracy, Sensitivity, and Precision. Implementing this approach is expected to yield improved outcomes for brain tumor identification.

Brain, Tumor, MRI (Magnetic Resonance Imaging), Self Organizing Map, Segmentation.