Keywords: Analysis/Processing, AI/ML Software
Motivation: AIF in DCE MRI is often compromised by partial volume effects and lacks reproducibility across different radiologists. Also, different brain regions may require different optimal AIFs, but one global AIF is used for most DCE MRI.
Goal(s): To develop a method to automatically generate optimal, brain region-specific AIFs to achieve more accurate pharmacokinetic mapping in DCE-MRI.
Approach: We utilized physics informed neural network (PINN) using a 1D CNN architecture and trained it using only simulation-based datasets.
Results: The proposed method reduced noise, stabilized Ktrans variation, and enhanced highlighting of suspected tumor regions, as MCA influence decreased toward peripheral slices.
Impact: In this study, we proposed a PINN method for end-to-end automated extraction of local AIFs from multiple tissue response functions. The proposed method can also overcome partial volume effects with high reproducibility, potentially improving routine clinical DCE studies.
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