Abstract #2284

Automatic Segmentation of Hyperpolarized Gas MRI via Deep Learning

Joshua R Astley^1,2, Alberto M Biancardi¹, Paul JC Hughes¹, Laurie J Smith¹, Helen Marshall¹, Grace T Mussell¹, James Eaden¹, Nicholas D Weatherley¹, Guilhem J Collier¹, Jim M Wild¹, and Bilal A Tahir^1,2

¹POLARIS, Department of Infection, Immunity & Cardiovascular Disease, University of Sheffield, Sheffield, United Kingdom, ²Department of Oncology and Metabolism, University of Sheffield, Sheffield, United Kingdom

Deep learning (DL)-based segmentation was conducted on a total of 431 ³He and ¹²⁹Xe 3D ventilation images using several training paradigms. Combined ³He and ¹²⁹Xe training showed a significant improvement over all other DL methods. In the majority of DL models, no significant difference was observed between ³He and ¹²⁹Xe testing data. Results suggest that ³He and ¹²⁹Xe images share important features that allow combined ³He and ¹²⁹Xe DL models to provide superior segmentations to singular gas models. In addition, it was shown that DL generates faster segmentations without the requirement of proton MRI compared to state-of-the-art model-based solutions.

This abstract and the presentation materials are available to members only; a login is required.

Join Here