Meeting Banner
Abstract #4611

Fast and accurate reconstruction of accelerated 7T susceptibility-weighted imaging using multi-scale hybrid CNN-Transformer network

Caohui Duan1, Dong Zhang2, Xiangbing Bian1, Jianxun Qu3, and Xin Lou1
1Department of Radiology, The First Medical Center, Chinese PLA General Hospital, Beijing, China, 2Department of Electrical and Computer Engineering, University of British Columbia, Vancouver, BC, Canada, 3Research Collaboration Team, Siemens Healthineers, Beijing, China

Synopsis

Keywords: Machine Learning/Artificial Intelligence, AI/ML Image Reconstruction

Motivation: Susceptibility-weighted imaging (SWI) at ultra-high field 7T is a powerful tool for evaluating a wide range of pathology, but has long acquisition time.

Goal(s): To investigate the diagnostic performances of highly accelerated 7T SWI using multi-scale hybrid CNN-Transformer Network (MACT-Net).

Approach: The diagnostic performances of the MSCT-Net reconstructed images for identifying swallow tail sign (STS) and distinguishing Parkinson’s disease (PD) patients from healthy controls were evaluated on an independent PD cohort.

Results: With a 6-8-times acceleration, MSCT-Net showed comparable diagnostic performance to the fully sampled 7T SWI for discriminating PD from healthy controls using the bilateral STS evaluation score.

Impact: The potentially reduced scan time with MSCT-Net offers new possibilities for widely adoption of 7T SWI in clinical brain imaging. Furthermore, the approach has the potential to guide design of reconstruction models for other high-resolution 7T MRI data.

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.

Keywords