Abstract #1857

Diffusion MRI spherical mean improves deep cerebellar nuclei segmentation

Jon Haitz Legarreta¹, Zhou Lan^1,2, Yuqian Chen¹, Fan Zhang³, Edward Yeterian^4,5, Nikos Makris⁶, Jarrett Rushmore^4,6, Yogesh Rathi⁶, and Lauren J. O’Donnell¹

¹Radiology, Brigham and Women's Hospital, Mass General Brigham/Harvard Medical School, Boston, MA, United States, ²Center for Clinical Investigation, Brigham and Women’s Hospital, Mass General Brigham, Boston, MA, United States, ³School of Information and Communication Engineering, University of Electronic Science and Technology of China, Chengdu, China, ⁴Anatomy and Neurobiology, Boston University Chobanian & Avedisian School of Medicine, Boston, MA, United States, ⁵Psychology, Colby College, Waterville, ME, United States, ⁶Psychiatry, Brigham and Women’s Hospital, Mass General Brigham/Harvard Medical School, Boston, MA, United States

Synopsis

Keywords: Analysis/Processing, Segmentation, brain, cerebellum, deep learning, diffusion MRI, neural networks

Motivation: Structural MRI data enables cerebellar nuclei delineation, but small errors in registration to diffusion MRI data can hinder cerebellar brain connectivity analysis.

Goal(s): Our goal is to provide an accurate deep cerebellar nuclei segmentation method employing diffusion MRI data.

Approach: We employ several MRI contrasts to train a deep neural network to segment the deep cerebellar nuclei and assess their performance.

Results: Diffusion MRI spherical mean data enable to segment the deep cerebellar nuclei with improved precision.

Impact: Diffusion MRI spherical mean should be considered as a relevant image contrast for cerebellar structure segmentation using deep neural networks towards an increasingly accurate connectivity analysis.

How to access this content:

For one year after publication, abstracts and videos are only open to registrants of this annual meeting. Registrants should use their existing login information. Non-registrant access can be purchased via the ISMRM E-Library.

After one year, current ISMRM & ISMRT members get free access to both the abstracts and videos. Non-members and non-registrants must purchase access via the ISMRM E-Library.

After two years, the meeting proceedings (abstracts) are opened to the public and require no login information. Videos remain behind password for access by members, registrants and E-Library customers.

Click here for more information on becoming a member.