We implement deep neural nets for instance-based segmentation of axons in TEM images. Our approach studying segmentation of unmyelinated fibers recently published in [1].

Data associated with this study, were collected as part of the Stimulating Peripheral Activity to Relieve Conditions (SPARC) program and are available through the SPARC Portal (RRID: SCR_017041) under a CC-BY 4.0 license.

All training, testing, and evaluation scripts are hosted in this GitHub repository. Due to large file sizes pretrained models cannot be included in the GitHub package and have to be downloaded from the following link. 

Pretrained Models

Citation:

Plebani, Emanuele; Biscola, Natalia P; Havton, Leif A; Rajwa, Bartek; Shemonti, Abida Sanjana; Jaffey, Deborah; Powley, Terry; Keast, Janet R; Lu, Kun-Han; Dundar, M Murat; "High-throughput segmentation of unmyelinated axons by deep learning", Scientific Reports 12 1 16-Jan 2022, Nature Publishing Group
DOI: https://doi.org/10.1038/s41598-022-04854-3

Acknowledgement:

The segmentation research reported in this publication was supported by the NIH Common Fund under SPARC OT2 (Stimulating Peripheral Activity to Relieve Conditions) award OD023847.

Contact:

Murat Dundar
mdundar at iupui dot edu