Three Dimensional Blood Vascular Tree Extraction

Over the last few decades,medical image processing has become an invaluable tool in making a diagnosis.The progression from generating two-dimensional X-ray images to the possibility of creating three-dimensional representations of the body has significantly helped when it comes to diagnosis.Medical image segmentation as an important branch of medical images processing,especially in blood vessels image segmentation plays an important role when it comes to diagnosing and developing operations that are more intricate.This paper studied blood vascular tree extraction and mainly focused on three dimensional cerebral vascular tree extraction.During our research,we used MRA medical image sequence,and VKT and ITK open source software.Three methods,region-growing method,level set method and vesselness filter combined with threshold method,were used to the three-dimensional blood vascular tree extraction from the given image sequence.Moreover,the processing effects of the three methods were compared.According to the processing results,level-set method has more advantages than the other two.In addition,we have used the skeletonization method for the extracted vascular tree then performed three dimensional blood vascular tree centerline extraction to lay the foundation for the further works about the vascular disease detection.