Simulation and Synthesis in Medical Imaging

Patch Based Synthesis of Whole Head MR Images: Application To EPI Distortion Correction

Simulation and Synthesis in Medical Imaging

Simulation and synthesis in medical imaging : first International Workshop, SASHIMI 2016, held in conjunction with MICCAI 2016, Athens, Greece, October 21, 2016, Proceedings. SASHIMI (Workshop) (1st : 2016 : Athens, Greece), October 2016

28367541

978-3-31-946629-3, 978-3-31-946630-9

Snehashis Roy, Yi-Yu Chou, Amod Jog, John A. Butman, Dzung L. Pham

Different magnetic resonance imaging pulse sequences are used to generate image contrasts based on physical properties of tissues, which provide different and often complementary information about them. Therefore multiple image contrasts are useful for multimodal analysis of medical images. Often, medical image processing algorithms are optimized for particular image contrasts. If a desirable contrast is unavailable, contrast synthesis (or modality synthesis) methods try to "synthesize" the unavailable constrasts from the available ones. Most of the recent image synthesis methods generate synthetic brain images, while whole head magnetic resonance (MR) images can also be useful for many applications. We propose an atlas based patch matching algorithm to synthesize T2-w whole head (including brain, skull, eyes etc) images from T1-w images for the purpose of distortion correction of diffusion weighted MR images. The geometric distortion in diffusion MR images due to in-homogeneous B0 magnetic field are often corrected by non-linearly registering the corresponding b = 0 image with zero diffusion gradient to an undistorted T2-w image. We show that our synthetic T2-w images can be used as a template in absence of a real T2-w image. Our patch based method requires multiple atlases with T1 and T2 to be registeLowRes to a given target T1. Then for every patch on the target, multiple similar looking matching patches are found on the atlas T1 images and corresponding patches on the atlas T2 images are combined to generate a synthetic T2 of the target. We experimented on image data obtained from 44 patients with traumatic brain injury (TBI), and showed that our synthesized T2 images produce more accurate distortion correction than a state-of-the-art registration based image synthesis method.