Deep learning–based synthesis of hyperpolarized gas MRI ventilation from 3D multi-inflation proton MRI

Joshua Russell Astley, Alberto M Biancardi, Helen Marshall, Laurie J Smith, Paul JC Hughes, Guilhem J Collier, Matthew Q Hatton, Jim M Wild, Bilal Tahir

Keywords: Deep learning, image synthesis, proton MRI, functional imaging

TL;DR: We demonstrate that a combination of proton MRI ventilation modeling and multi-inflation proton MRI using a deep learning approach can generate realistic synthetic hyperpolarized gas MRI ventilation scans.

Abstract: Hyperpolarized (HP) gas MRI allows visualization and quantification of regional lung ventilation; however, there is limited clinical uptake due to the requirement for highly specialized equipment and exogenous contrast agents. Alternative, non-contrast, model-based proton ($^1$H)-MRI surrogates of ventilation, which correlate moderately with HP gas MRI, have been proposed. Recently, deep learning (DL)-based methods have been used for the synthesis of HP gas MRI from free-breathing $^1$H-MRI for a single 2D section. Here, we developed and evaluated a multi-channel 3D DL method that combines modeling and data-driven approaches to synthesize HP gas MRI ventilation scans from multi-inflation $^1$H-MRI.

Registration: I acknowledge that acceptance of this work at MIDL requires at least one of the authors to register and present the work during the conference.

Authorship: I confirm that I am the author of this work and that it has not been submitted to another publication before.

Paper Type: novel methodological ideas without extensive validation

Primary Subject Area: Image Synthesis

Secondary Subject Area: Transfer Learning and Domain Adaptation

Confidentiality And Author Instructions: I read the call for papers and author instructions. I acknowledge that exceeding the page limit and/or altering the latex template can result in desk rejection.