Abstract #3101

Deep learning-based B0 inhomogeneities mapping using sparse CEST spectral data

Yiran Li¹, Danfeng Xie¹, Hanlu Yang¹, Li Bai¹, Guanshu Liu², and Ze Wang³

¹Department of Electrical and Computer Engineering, Temple University, PHILADELPHIA, PA, United States, ²Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine, Baltimore, MD, United States, ³Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine, Baltimore, MD, United States

Chemical Exchange Saturation Transfer (CEST) is an MR based imaging method that can image compounds containing protons exhibiting a suitable exchange rate with bulk water. One of the crucial technical hurdles in CEST MRI is, as CEST signal highly depends on the saturation frequency, how to accurately correct the B0 inhomogeneity in each voxel. We proposed two deep learning (DL) based methods for estimating B0 inhomogeneities to accelerate CEST imaging using spare samples. While only a small sample size was used, our study shows the potential of DL-based B0 mapping, which can greatly reduce the total CEST acquisition time.

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

Join Here