Abstract #5231

AI-Powered Accurate Lipid Segmentation for Lipid Removal from Brain MRSI Data

Yuanbo Zhang^1,2, Ziwen Ke^1,3,4, Wen Jin^1,2, Ruihao Liu^1,3, Danni Wang^3,5, Wenqi Zhang³, Jingyi Li⁶, Yao Li^3,4, and Zhi-Pei Liang^1,2

¹Beckman Institute for Advanced Science and Technology, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ²Department of Electrical and Computer Engineering, University of Illinois at Urbana-Champaign, Urbana, IL, United States, ³School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China, ⁴Institute of Medical Robotics, Shanghai Jiao Tong University, Shanghai, China, ⁵Neuroimaging Research Branch, National Institute on Drug Abuse, National Institutes of Health, Baltimore, MD, United States, ⁶Department of Statistics, University of Illinois Urbana-Champaign, Urbana, IL, United States

Synopsis

Keywords: Spectroscopy, Segmentation

Motivation: Accurate anatomical constraints are critical for removal of lipid signals from brain MRSI data, but obtaining them through lipid segmentation from anatomical scans remains challenging due to intensity heterogeneity, water-lipid ambiguity, and high-dimensionality of spatial-intensity distribution of lipid-dominating tissues.

Goal(s): To develop an accurate lipid segmentation method for brain MRSI scans.

Approach: We synergistically integrated a deep translation model to facilitate water-lipid separation, a subspace-based probability model to capture the global lipid spatial prior, and a deep diffusion-based lipid support learning to segment lipid-dominating tissues.

Results: The proposed method was validated on experimental data, producing significantly improved lipid segmentation than existing methods.

Impact: The proposed method will significantly improve the robustness of lipid removal and help generate high-quality metabolite maps from brain MRSI data, especially those obtained without lipid suppression.

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.