Brain Tumor segmentation using modified Fuzzy metric based Approach with Adaptive Technique
Dr.D. Sherlin · International Journal of Advanced Trends in Computer Science and Engineering · 2019
Medical image diagnosis field has boomed up to be the key research area and this helps to achieve many advancement in diagnosing and treating patients.One such remarkable area is automatic brain tumour segmentation.Manual identification of tumor from MRI images are challenging, tedious and complex process and they require the support of skilled medical practitioner whereas this automatic segmentation helps in easy diagnosis of the tumor location exactly with less time and modelled to be user friendly.There are numerous automatic segmentation methods used for detecting the position of tumors.But accuracy and time are the major parameters which decide the priority and prominence of the segmentation algorithm.In our proposed modified adaptive K+FCM method, the tumor detection is done based on three important stages, where the first stage is pre-processing which proceed with the denoising process, second stage is segmentation which works on clustering and tumor segmentation and final stage is post-processing which operates on the repair and denoising of segmented tumors.The proposed adaptive filtered based modified K+ Fuzzy corner metric algorithm shows better time precision and accuracy when compared to the existing algorithms like random forest and PCA segmentation methods.The algorithm is analysed with BRATS2013 datasets and results obtained shown that the proposed algorithm shows better segmentation.