Segmentation of Tumor from Brain MRI Using Fuzzy Entropy and Distance Regularised Level Set

Image processing is needed in medical discipline for variety of disease assessments. In this work, Firefly Algorithm (FA)-assisted approach is implemented to extract tumor from brain magnetic resonance image (MRI). MRI is a clinically proven procedure to record and analyze the suspicious regions of vital body parts. The proposed approach is implemented by integrating the fuzzy entropy and Distance Regularized Level Set (DRLS) to mine tumor region from axial, sagittal, and coronal views’ brain MRI dataset. The proposed approach is a three-step procedure, such as skull stripping, FA-assisted fuzzy entropy-based multi-thresholding, and DRLS-based segmentation. After extracting the tumor region, the size of the tumor mass is examined using the 2D Minkowski distance measures, such as area, area density, perimeter, and perimeter density. Further, the vital features from the segmented tumor are extracted using GLCM and Haar wavelet transform. Proposed approach shows an agreeable result in extraction and analysis of brain tumor of chosen MRI dataset.