Segmentation of 3D magnetic resonance brain vessel images based on level set approaches

Quantitative modeling of brain vasculature is important for diagnosis of vessel pathologies as well as for surgery treatment planning. Magnetic resonance angiography (MRA) provides reliable visualization of vessel tree structure and its organization. Accuracy of vessel segmentation from MRA is an important step in model building; its accuracy influences obtained model quality. This paper presents three level set based segmentation approaches, including one that represents original authors contribution. These methods are combined together with vesselness function estimated for analyzed images. Presented algorithms were applied both for artificial and real brain 3D MR images. Analysis results along with discussion are also included.