Premium
The Brain Atlas Glow Up: Utilizing Autofluorescence to Enhance Contrast Between White and Gray Matter
Author(s) -
Shearer Jake,
Stabio Maureen,
Spitzer Victor
Publication year - 2020
Publication title -
the faseb journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.709
H-Index - 277
eISSN - 1530-6860
pISSN - 0892-6638
DOI - 10.1096/fasebj.2020.34.s1.02396
Subject(s) - white matter , brain atlas , myelin , contrast (vision) , autofluorescence , computer science , biomedical engineering , computer vision , high resolution , artificial intelligence , magnetic resonance imaging , medicine , optics , physics , radiology , neuroscience , geology , biology , fluorescence , remote sensing , central nervous system
Myelin stained brain atlases are essential tools used to develop an understanding of brain organization and wiring. However, in a medical setting, accurate diagnoses require providers to be able to mentally translate the information gained from atlases to their patient’s MRI data. Myelin stained atlases provide excellent contrast and detail but are routinely constructed with a centimeter or more between images. This poor volumetric resolution poses a challenge when comparing anatomy from myelin stained atlases to MRI datasets as there are often significant gaps of data between subsequent images. Students must learn how to interpolate the missing data between the myelin stained sections. This study was designed to test the hypothesis that an ultrahigh‐resolution, high‐contrast, volumetric atlas of the brain could be re‐constructed from images obtained from blockface milling of whole brains while taking advantage of the autofluorescent properties of myelin to enhance contrast. Using custom‐built 470nm LED panels in conjunction with a 511/20nm bandpass gel filter, we were able to capture images with enhanced contrast between white and gray matter with sufficient detail to identify individual nuclei of the brain. Formalin‐fixed brains were cryopreserved, frozen, and serially imaged at 50μm intervals. At each interval, we imaged the brain using both full color and fluorescent photography, then converted the data into image volumes. The high volumetric resolution of the visual data allows any viewing plane to be generated and directly compared with the equivalent plane in an MRI volume. The volumes were then coregistered and efficiently segmented using the high contrast between white and gray matter to generate full‐color, 3D models of gray matter nuclei and white matter fiber tracts. We also created an accompanying 2D atlas using traditional atlas sections. This atlas will enhance the literacy of MRI anatomy by allowing both students and providers to visualize the information gaps in current myelin stained atlases. Support or Funding Information University of Colorado: Modern Human Anatomy Program, Touch of Life Technologies, Colorado State Anatomical Board, Inworks