A new segmentation algorithm for medical volume image based on K-means clustering

K-means algorithm is wildly used in medical image segmentation for its powerful fuzzy information process ability but the algorithm has some shortages such as low efficiency in calculation which limited the usage of the algorithm. Some measures are advanced to overcome the shortages of original K-means algorithm and a new medical volume image segmentation algorithm is presented. Firstly, according to the physical means of the medical data, the volume data field is preprocessed to speed up succeed clustering processing; Secondly, the improved K-means algorithm is deduced and analyzed through improving cluster seed selection method and calculation flow and redesigning pixel processing and operational principle of original K-means algorithm. Finally, the experimental results show that the algorithm has high accuracy when used to segment 3D medical images and can improve calculation speed greatly.