Abstract #4895

Iterative Refinement of Expert Contours for Improved Ground-truth Quality in Intracranial Vessel Segmentation Neural Network Training

Hanyue Zhou^1,2, Jiayu Xiao², Dan Ruan^1,3, and Zhaoyang Fan^2,4,5

¹Bioengineering, University of California, Los Angeles, Los Angeles, CA, United States, ²Radiology, University of Southern California, Los Angeles, CA, United States, ³Radiation Oncology, University of California, Los Angeles, Los Angeles, CA, United States, ⁴Radiation Oncology, University of Southern California, Los Angeles, CA, United States, ⁵Biomedical Engineering, University of Southern California, Los Angeles, CA, United States

Synopsis

With a typical slice-by-slice labeling fashion, the manual contouring process is subject to large intra-observer variation, especially for small-sized intracranial arteries. We propose an iterative refinement approach for ground truth contours with the help of deep neural networks for intracranial lumen and vessel wall segmentations. We demonstrate that the approach improved the smoothness of the predicted contours and feature quantification, which can potentially boost the robustness of a neural network as a consequence of the reduced uncertainty in expert labels.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.