Repulsive force based snake model to segment and track neuronal axons in 3D microscopy image stacks

Area of Science:

• Neuroscience
• Computational Biology
• Image Analysis

Background:

• Axonal and dendritic branching patterns are crucial for synaptic connectivity.
• Current 3D imaging allows visualization of neuronal processes but lacks robust methods for individual axon tracing.

Purpose of the Study:

• To develop a robust method for segmenting and tracking individual axons in 3D image stacks.
• To enable detailed analysis of axonal branching patterns and their impact on neural connectivity.

Main Methods:

• A repulsive force-based snake model was developed for axonal profile segmentation.
• The method utilizes results from adjacent 2D image slices to guide 3D segmentation.
• Individual axons are extracted based on the segmentation of the entire 3D image stack.

Main Results:

• The snake model successfully segments all axonal profiles within 2D images.
• Segmentation is robustly extended across hundreds of images in a 3D stack, connecting axonal profiles.
• The method allows for the extraction of individual axons from complex 3D datasets.

Conclusions:

• The repulsive force snake model provides a robust solution for tracing individual axons in 3D microscopy data.
• This method facilitates the study of axonal branching and its role in synaptic connectivity.
• The approach is validated using 3D images of fluorescently labeled axons in transgenic mice.