Multi-Domain Data Aggregation for Axon and Myelin Segmentation in Histology Images

TL;DR

The paper addresses the lack of generalizable axon and myelin segmentation across histology modalities, species, and pathologies. It proposes a publicly accessible multi-domain segmentation model built on a durable nn-UNet backbone with a resolution-ignorant design, trained via extensive data aggregation across TEM, SEM, BF, and CARS datasets. The results show that aggregated multi-domain training yields equal or better performance than single-domain models and enhances out-of-distribution generalization, with the FULL_AGG generalist outperforming dedicated baselines on both in-distribution and OoD data. The work provides an open-source tool integrated into AxonDeepSeg, promoting durable, easy-to-maintain segmentation workflows for neuroscience researchers.

Abstract

Quantifying axon and myelin properties (e.g., axon diameter, myelin thickness, g-ratio) in histology images can provide useful information about microstructural changes caused by neurodegenerative diseases. Automatic tissue segmentation is an important tool for these datasets, as a single stained section can contain up to thousands of axons. Advances in deep learning have made this task quick and reliable with minimal overhead, but a deep learning model trained by one research group will hardly ever be usable by other groups due to differences in their histology training data. This is partly due to subject diversity (different body parts, species, genetics, pathologies) and also to the range of modern microscopy imaging techniques resulting in a wide variability of image features (i.e., contrast, resolution). There is a pressing need to make AI accessible to neuroscience researchers to facilitate and accelerate their workflow, but publicly available models are scarce and poorly maintained. Our approach is to aggregate data from multiple imaging modalities (bright field, electron microscopy, Raman spectroscopy) and species (mouse, rat, rabbit, human), to create an open-source, durable tool for axon and myelin segmentation. Our generalist model makes it easier for researchers to process their data and can be fine-tuned for better performance on specific domains. We study the benefits of different aggregation schemes. This multi-domain segmentation model performs better than single-modality dedicated learners (p=0.03077), generalizes better on out-of-distribution data and is easier to use and maintain. Importantly, we package the segmentation tool into a well-maintained open-source software ecosystem (see https://github.com/axondeepseg/axondeepseg).

Figures (8)

• Figure 1: Dataset previews
• Figure 2: Intra-Modality Aggregation Results: Performance of dedicated and generalist models on all BF datasets
• Figure 3: Inter-Modality Aggregation Results: Performance of dedicated and generalist models on all imaging modalities.
• Figure A: Visual description of experiments
• Figure B: Dataset-wise results of individual folds for intra-modality.