Related Experiment Video
Updated: Jul 12, 2026
06:17
Investigating Mammalian Axon Regeneration: In Vivo Electroporation of Adult Mouse Dorsal Root Ganglion
Published on: September 1, 2018
Deep learning-based hybrid analysis for axonal regeneration and myelination in rat sciatic nerve
Jun Hong Won1, Chawon Yun1,2, So Young Lee1
1Department of Orthopedic Surgery, Korea University Guro Hospital, Korea University College of Medicine, 148, Gurodong-ro, Guro-gu, Seoul, Korea.
Scientific Reports
|July 10, 2026
Summary
Automated analysis of peripheral nerve regeneration significantly speeds up histomorphometry. A hybrid deep learning approach offers the best balance of speed and accuracy for axon and myelin analysis.
Area of Science:
- Neuroscience
- Biomedical Engineering
- Computational Biology
Background:
- Quantitative histomorphometry of peripheral nerves is crucial for studying axonal regeneration and remyelination.
- Manual analysis is time-consuming and not feasible for large-scale research.
Purpose of the Study:
- To compare the efficiency and reliability of three automated morphometric methods against manual analysis for peripheral nerve studies.
- To identify an optimal automated workflow for quantitative histomorphometry.
Main Methods:
- Six rat sciatic nerves (3 naive, 3 regenerating) were analyzed using semithin transverse sections.
- Manual analysis by three observers served as the reference.
- Automated analyses included Trainable Weka Segmentation, AxonDeepSeg, and a refined AxonDeepSeg approach.
- Key parameters: axon count, diameter, area, g-ratio, myelin thickness, and analysis time.
Main Results:
- Automated methods reduced analysis time by over 50% compared to manual analysis.
- AxonDeepSeg was the fastest automated method.
- Fully automated methods showed variable agreement with manual analysis, especially in regenerating nerves.
- A hybrid deep learning approach with manual refinement improved accuracy and reduced bias.
Conclusions:
- Automated histomorphometry, particularly hybrid deep learning methods, offers a more efficient and reliable alternative to manual analysis for peripheral nerve studies.
- This advancement supports large-scale investigations into axonal regeneration and remyelination.
