ROBUST CEREBROVASCULAR SEGMENTATION IN 4D ASL MRA IMAGES

Author(s)

Phellan, Renzo ; Linder, Thomas ; Helle, Michael ; Falcao, Alexandre X. ; Forkert, Nils D. ; IEEE

• Document type: Journal article
• Source: 2018 IEEE 15TH INTERNATIONAL SYMPOSIUM ON BIOMEDICAL IMAGING (ISBI 2018); v. N/A, p. 4-pg., 2018-01-01.

Abstract

Cerebrovascular diseases are one of the main causes of disability and death in the world. Methods for segmentation of the cerebral blood vessels can help clinicians to visualize the cerebrovascular system, determine age-related normal values, and diagnose and study cerebrovascular diseases. A common problem for these methods is the precise delineation of small vessels, due to different artifacts depending on the medical imaging modality. In this work, we present an automatic segmentation method for four dimensional arterial spin labeling magnetic resonance angiography (4D ASL MRA) images of the brain, which allows an improved segmentation of small vessels, reaching an average Dice similarity coefficient (DSC) of 0.946 based on an evaluation of five datasets from healthy subjects. The results show that the proposed method is able to account for the magnetic labeling decay and flow related artifacts, which considerably reduce the contrast of small vessels present in the image, leading to improved segmentation results. (AU)

• FAPESP's process: 14/12236-1 - AnImaLS: Annotation of Images in Large Scale: what can machines and specialists learn from interaction?
• Grantee: Alexandre Xavier Falcão
• Support Opportunities: Research Projects - Thematic Grants