Abstract #2106

Optimization of IVIM-DKI MR Imaging model using Automatic Differentiation Methods: Enhancing the Computation Efficacy by 100 Fold

Hisham Mohammed Hanif¹, Himansu Maurya¹, Esha Baidya Kayal², Ganeshkumar M², Rithvik Hariprasad², Raju Sharma³, Kedar Khare^1,4, Devasenathipathy Kandasamy³, and Amit Mehndiratta^2,5,6

¹Department of Physics, Indian Institute of Technology - Delhi, New Delhi, India, ²Centre for Biomedical Engineering, Indian Institute of Technology - Delhi, New Delhi, India, ³Department of Radiodiagnosis, All India Institute of Medical Sciences, New Delhi, India, ⁴Optics and Photonics Centre, Indian Institute of Technology - Delhi, New Delhi, India, ⁵Faculty of Medicine and Health, University of New South Wales (UNSW), Sydney, Australia, ⁶Yardi School of Artificial Intelligence, Indian Institute of Technology - Delhi, New Delhi, India

Synopsis

Keywords: Quantitative Imaging, IVIM, IVIM, IVIM-DKI, Diffusion MRI, Perfusion.

Motivation: Spatially constrained IDTV method have shown improved IVIM-DKI parameter estimation, although suffers from high time complexity preventing it to be used in clinical setup in real-time.

Goal(s): This study aims to improve the computational time and efficiency of IVIM-DKI model optimization and proposes a novel spatially constrained IDIV-AD method using automatic differentiation framework.

Approach: The proposed novel IDTV-AD method demonstrated at least 100 times faster computation time than the IDTV method maintaining the robustness and quality of parameter estimation.

Results: Proposed novel IDTV-AD method demonstrated a time-efficient implementation and robust parameters estimation of IVIM-DKI parameters in pancreatic tumors and healthy pancreatic tissue.

Impact: Novel spatially constrained IDTV-AD method using automatic differentiation demonstrated significantly fast (at least 100 times faster than the state-of-the-art method) and accurate IVIM-DKI parameter estimation, tested in clinical dataset of pancreatic cancer cohort.

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.