This paper reviews the comparative analysis of medical image segmentation techniques for segmenting tumor from MRI images to clearly determine brain tumours and help the physician make a better diagnosis. Brain tumors are the leading cause of cancer death in children under the age of 20.The brain tumor segmentation is a challenging and complex task. In recent years segmentation problems have been attracted interest and ongoing research. In this paper comparative study of hybrid segmentation techniques like K-means, PSO-K means,Firefly-K means ,Seg-UNet architecture were carried out and found that hybrid Seg-UNet can provide better output of original image and visual observation of tumors. This hybrid method achieves 99.2% segmentation accuracy with sensitivity 96% and specificity 94%. The results showed that hybrid methods provide a significant contribution to help the physicians in Magnetic resonance brain image tumor detection.
.