Meeting Banner
Abstract #0107

Deep learning for improved workflow in MRgFUS treatment planning

Pan Su1,2, Sijia Guo1,3, Florian Maier4, Steven Roys1,3, Himanshu Bhat2, Elias R. Melhem1, Dheeraj Gandhi1, Rao P. Gullapalli1,3, and Jiachen Zhuo1,3
1Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, Baltimore, MD, United States, 2Siemens Medical Solutions USA Inc, Malvern, PA, United States, 3Center for Metabolic Imaging and Therapeutics (CMIT), University of Maryland Medical Center, Baltimore, MD, United States, 4Siemens Healthcare GmbH, Erlangen, Germany

Transcranial MRI-guided focused ultrasound (tcMRgFUS) is a promising technique to treat multiple diseases. Here we examined the feasibility of leveraging deep-learning to convert MRI dual echo UTE images directly to synthesized CT skull images. We demonstrated that the derived model is capable of not only segmenting the UTE images to generate synthetic CT skull masks that are highly comparable to true CT skull masks, but is also able to reliably predict the CT skull intensities in Hounsfield units. Furthermore, we demonstrated that synthetic CT skull can be reliably used for skull-density-ratio (SDR) determination and predicting target temperature rise in tcMRgFUS.

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

Join Here