Brain Tumor Image Segmentation Based On Neighborhood Rough Set And Fuzzy Clustering

Objective: The brain tumor image segmentation system based on neighborhood rough sets and fuzzy clustering is used to solve the problems of high complexity,uneven grayness and partial volume effect in MR brain tumor medical images.It lays the foundation for CAD and provides data support and ideas for early diagnosis and treatment of brain tumor patients.Methods: The neighborhood rough set method is used to simulate the upper and lower approximation of the target and the background area to obtain the edge information of the image.Construct the relationship between the central pixel and the neighborhood.This method uses the membership function in fuzzy C-means clustering to describe the similarity between the pixel value and the cluster center.The membership matrix and clustering center are updated to get the best threshold of brain tumor image segmentation.Results: Aiming at the defects of magnetic resonance(MR)imaging,such as blur and gray unevenness,an image segmentation method based on neighborhood rough set and fuzzy clustering was proposed.The selection of image boundary information by neighborhood rough sets is relatively robust.Combining fuzzy clustering methods can make full use of image spatial information.Experiments use Dice coefficient and Jaccard coefficient to evaluate the accuracy of segmentation results.The results show that this method can effectively segment the tumor area.Its segmentation accuracy is improved compared with the traditional fuzzy clustering segmentation method.Conclusion: In summary,this article has conducted in-depth research on brain tumor image segmentation technology.According to the established neighborhood rough set and fuzzy clustering model,it can obtain ideal segmentation results.This method can accurately and efficiently segment complex brain tumor images in grayscale scenes.It has certain practicability and flexibility.