Keywords: AI/ML Image Reconstruction, AI/ML Image Reconstruction
Motivation: Diffusion models in accelerated MRI reconstruction can introduce deep hallucinations and compromise image fidelity. We hypothesise that integrating compressed sensing (CS) with diffusion models can recover structural details and improve reconstruction fidelity.
Goal(s): This study develops a hybrid approach – combining generative AI with CS-MRI – to achieve faithful accelerated MRI reconstructions whilst minimising deep hallucinations.
Approach: We introduce an integration of a diffusion model with CS-MRI – leveraging both Fourier and wavelet domain – to enhance reconstruction quality and detail preservation.
Results: Our method recovers details lost in diffusion model reconstruction, surpassing GRAPPA and diffusion model reconstruction alone in PSNR and SSIM.
Impact: By leveraging the strengths of both generative AI and compressed sensing (CS), the proposed method faithfully restores images from undersampled measurements, achieves high-fidelity MRI reconstructions, and does not obfuscate the diagnostic quality of the scan or downstream AI tasks.
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.
Keywords