Meeting Banner
Abstract #2776

Improved MRA Synthesis with KAN-based Network with Global MIP Guidance

Dayang Wang1, Srivathsa Pasumarthi2, Ajit Shankaranarayanan3, and Greg Zaharchuk4
1Subtle Medical Inc., Columbia, SC, United States, 2Subtle Medical Inc., Santa Clara, CA, United States, 3Subtle Medical Inc., Menlo Park, CA, United States, 4Stanford University, Stanford, CA, United States

Synopsis

Keywords: Acquisition Methods, AI/ML Image Reconstruction, MRA Synthesis

Motivation: Time-of-Flight (TOF) Magnetic Resonance Angiography (MRA) is critical for diagnosing vascular diseases such as aneurysms, but long acquisition time and noise susceptibility limit its clinical efficiency.

Goal(s): To avoid the long scan time, we aim to synthesize high-quality TOF-MRA images using faster T1-weighted (T1w) and T2-weighted (T2w) sequences.

Approach: We propose a 2.5D Kolmogorov–Arnold Networks (KAN)-based model, termed sKAN, which integrates KAN with snake convolution.

Results: The Synthesized MRA (Syn-MRA) images closely resemble standard-of-care (SOC) MRA, displaying realistic vessel textures and reduced noise. They can serve as a potential alternative to SOC-TOF-MRA, reducing the need for prolonged scan time.

Impact: The proposed method enables TOF-MRA synthesis from faster T1- and T2-weighted MRI images. The synthesized images closely match standard-of-care (SOC) TOF-MRA, accurately depicting vessels in major central areas, thus eliminating need for lengthy acquisition process of SOC-TOF-MRA in clinical settings.

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