Meeting Banner
Abstract #1010

Deep learning 3D white matter fiber orientation from 2D histology: pulling 3D rabbits out of 2D hats

Kurt G Schilling1, Vishwesh Nath2, Samuel Remedios2, Roza G Bayrak3, Yurui Gao1, Justin A Blaber4, Adam W Anderson1,5, and Bennett A Landman4,6

1Vanderbilt University Institute of Imaging Science, Nashville, TN, United States, 2Electrical Engineering and Computer Science, Vanderbilt University, Nashville, TN, United States, 3Computer Science, Vanderbilt University, Nashville, TN, United States, 4Electrical Engineering, Vanderbilt University, Nashville, TN, United States, 5Biomedical Engineering, Vanderbilt University, Nashville, TN, United States, 6Vanderbilt University Institue of Imaging Science, Nashville, TN, United States

Most histological analysis of tissue microstructure is inherently 2D. In this work, we implement a deep learning approach to extract 3D microstructural measures from 2D microscopy images. Specifically, we train a neural network to estimate 3D fiber orientation distributions from myelin-stained micrographs. We apply this technique to an entirely unseen brain, which suggests the potential to use this methodology on consecutive 2D slices to investigate 3D structural connectivity using “myelin-stained-tractography” at resolutions much higher than possible with current diffusion MRI practices. There is potential to use similar techniques to estimate a number of 3D metrics from common 2D histological contrasts.

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.

Keywords