Learning-based long-range axon tracing in dense scenes

Mark Hernandez, Adam Brewster, Larry Thul, Brian A. Telfer, Arjun Majumdar, Heejin Choi, Taeyun Ku, Kwanghun Chung, Laura J. Brattain

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

Recent advances in intact brain imaging, such as CLARITY and MAP (Magnified Analysis of the Proteome), enable volumetric imaging of brain tissue at subcellular resolution. Currently there is no effective tool for automated axon tracing at single fiber resolution in densely populated imagery. To address this need, we developed a machine learning-based high performance computing pipeline: 1) a convolutional neural network detects axon fiber voxels, 2) morphological operations extract fiber centerlines, 3) tracking logic connects fiber segments across low-intensity gaps and unresolved fiber crossings. The pipeline was implemented on a CPU cluster and tested on a 250 gigabyte volume of SMI-312 densely labeled axon fibers, imaged from parts of the hippocampus and cortex of a MAP-processed mouse brain. The pipeline automatically traced 221,298 fibers across gray and white matter in 10 hours. Of the traced fibers, 104 exceeded 1 mm, with the longest being 2.16 mm. An accuracy of 84% was reported based on manual evaluation of the 200 longest fibers. While there is room to improve accuracy, this pipeline offers a significant speed-up and increased efficiency over tracing one neuron at a time or manual fiber tracing. As the pipeline is scaled up to evaluate larger regions of the brain, connectivity patterns derived from these automatically traced long-range axons can potentially provide insights into brain function.

Original languageEnglish
Title of host publication2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018
PublisherIEEE Computer Society
Pages1578-1582
Number of pages5
ISBN (Electronic)9781538636367
DOIs
Publication statusPublished - 2018 May 23
Event15th IEEE International Symposium on Biomedical Imaging, ISBI 2018 - Washington, United States
Duration: 2018 Apr 42018 Apr 7

Publication series

NameProceedings - International Symposium on Biomedical Imaging
Volume2018-April
ISSN (Print)1945-7928
ISSN (Electronic)1945-8452

Other

Other15th IEEE International Symposium on Biomedical Imaging, ISBI 2018
CountryUnited States
CityWashington
Period18/4/418/4/7

All Science Journal Classification (ASJC) codes

  • Biomedical Engineering
  • Radiology Nuclear Medicine and imaging

Fingerprint Dive into the research topics of 'Learning-based long-range axon tracing in dense scenes'. Together they form a unique fingerprint.

  • Cite this

    Hernandez, M., Brewster, A., Thul, L., Telfer, B. A., Majumdar, A., Choi, H., Ku, T., Chung, K., & Brattain, L. J. (2018). Learning-based long-range axon tracing in dense scenes. In 2018 IEEE 15th International Symposium on Biomedical Imaging, ISBI 2018 (pp. 1578-1582). (Proceedings - International Symposium on Biomedical Imaging; Vol. 2018-April). IEEE Computer Society. https://doi.org/10.1109/ISBI.2018.8363875