Abstract #5112

Accelerated CEST Imaging with Parallel Deep Convolutional Neural Networks

Huajun She¹, Shu Zhang¹, Xinzeng Wang¹, Ece Ercan¹, Jochen Keupp², Anath Madhuranthakam^1,3, Ivan Dimitrov^1,4, Robert Lenkinski^1,3, and Elena Vinogradov^1,3

¹Radiology, UT Southwestern Medical Center, Dallas, TX, United States, ²Philips Research, Hamburg, Germany, ³Advanced Imaging Research Center, UT Southwestern Medical Center, Dallas, TX, United States, ⁴Philips Healthcare, Gainesville, FL, United States

CEST is a new contrast mechanism in MRI. However, a successful application of CEST is hampered by its slow acquisition. This work investigates accelerating CEST imaging using parallel convolutional neural networks (PCNN). We extend the Cascade-CNN into a multi-channel model and train the network establish a mapping from the multi-coil input to multi-coil output. This work is the first try to apply deep learning and convolutional neural networks technique in accelerating CEST imaging. The in vivo brain results show that the proposed method demonstrates a high quality reconstruction of the MTR_asym maps with different saturation pulses at R=4.

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

Join Here