A multi-channel deep learning approach for lung cavity estimation using hyperpolarized gas and proton MRI
Keywords: Deep learning, segmentation, multi-channel, multi-modal
TL;DR: We develop and validate a multi-channel, multi-modality deep learning approach, integrating functional and structural imaging, to generate accurate segmentations of the lung cavity.
Abstract: Hyperpolarized (HP) gas MRI enables quantification of regional lung ventilation via clinical biomarkers such as the ventilation defect percentage (VDP). VDP is computed from segmentations derived from spatially co-registered functional HP gas MRI and structural proton ($^1$H)-MRI; although these scans are acquired at similar inflation levels, misalignments are frequent, requiring a lung cavity estimation (LCE). Here, we propose a multi-channel deep learning method for generating LCEs using HP gas and $^1$H-MRI. We compare the performance of the proposed method to single-channel alternatives.
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: Segmentation
Secondary Subject Area: Validation Study
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.