Share Email Print

Proceedings Paper

A generalizable deep-learning approach to anatomical modeling of brain vasculature (Conference Presentation)
Author(s): Waleed Tahir; Jiabei Zhu; Sreekanth Kura; Xiaojun Cheng; Rafat Damseh; Frederic Lesage; Sava Sakadžic; David A. Boas; Lei Tian

Paper Abstract

Quantitative 3D analysis of brain vasculature is a fundamental problem with important applications, for which vessel segmentation is a first step. Traditional segmentation methods based on parametric models have limited accuracy. More recent techniques based on machine learning have promising results but limited generalization capability. We present a deep-learning based segmentation method that overcomes limitations of existing systems and demonstrates the ability to generalize to various imaging setups, samples including both in-vivo/ex-vivo data, with state-of-the-art results. We achieve so by exploiting several novel methods in deep learning, such as semi-supervised learning. We believe that our work will be another step forward towards improved large-scale neurovascular analysis.

Paper Details

Date Published: 9 March 2020
Proc. SPIE 11226, Neural Imaging and Sensing 2020, 1122611 (9 March 2020); doi: 10.1117/12.2543864
Show Author Affiliations
Waleed Tahir, Boston Univ. (United States)
Jiabei Zhu, Boston Univ. (United States)
Sreekanth Kura, Boston Univ. (United States)
Xiaojun Cheng, Boston Univ. (United States)
Rafat Damseh, Polytechnique Montréal (Canada)
Frederic Lesage, Polytechnique Montréal (Canada)
Sava Sakadžic, Massachusetts General Hospital (United States)
David A. Boas, Boston Univ. (United States)
Lei Tian, Boston Univ. (United States)

Published in SPIE Proceedings Vol. 11226:
Neural Imaging and Sensing 2020
Qingming Luo; Jun Ding; Ling Fu, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?