Meeting Banner
Abstract #4393

Automated 3-D Segmentation of Radiation-induced Cerebral Microbleeds on Susceptibility Weighted Imaging at 3T and 7T

Xiaowei Zou 1 , Wei Bian 2 , Christopher P. Hess 1 , Sarah J. Nelson 1 , and Janine M. Lupo 1

1 Radiology & Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 2 Radiology, Stanford University, Stanford, CA, United States

Manual segmentation of cerebral microbleeds is a lengthy and laborious task due to their small size, finite image resolution, large number, three-dimensional nature, various contrast, and imperfect spherical profile. In this study, we propose an automated 3-D CMB segmentation algorithm applied after automated CMB detection. Our preliminary results show that proposed method can efficiently, robustly and consistently segment CMB regions that the volume distribution of segmented CMB region on susceptibility weighted imaging at 3T and 7T are highly correlated, demonstrating it may be a reliable alternative to manual definitions.

This abstract and the presentation materials are available to members only; a login is required.

Join Here