High-throughput segmentation of unmyelinated axons by deep learning

Abstract

Axonal characterizations of connectomes in healthy and disease phenotypes are surprisingly incomplete and biased because unmyelinated axons, the most prevalent type of fibers in the nervous system, have largely been ignored as their quantitative assessment quickly becomes unmanageable as the number of axons increases. Herein, we introduce the first prototype of a high-throughput processing pipeline for automated segmentation of unmyelinated fibers. Our team has used transmission electron microscopy images of vagus and pelvic nerves in rats. All unmyelinated axons in these images are individually annotated and used as labeled data to train and validate a deep instance segmentation network. We investigate the effect of different training strategies on the overall segmentation accuracy of the network. We extensively validate the segmentation algorithm as a stand-alone segmentation tool as well as in an expert-in-the-loop hybrid segmentation setting with preliminary, albeit remarkably encouraging results. Our algorithm achieves an instance-level $$F_1$$ F 1 score of between 0.7 and 0.9 on various test images in the stand-alone mode and reduces expert annotation labor by 80% in the hybrid setting. We hope that this new high-throughput segmentation pipeline will enable quick and accurate characterization of unmyelinated fibers at scale and become instrumental in significantly advancing our understanding of connectomes in both the peripheral and the central nervous systems.

Organ- and function-specific anatomical organization of the vagus nerve supports fascicular vagus nerve stimulation

“…Deep-learning based algorithms have recently received attention in anatomically guided, medical image segmentation [44]. With regard to segmentation of anatomical features of nerves, deep-learning algorithms have been used on EM images [45], with regard to single fibers, and on ultrasound [46] and histological images [47], with regard to fascicles. To the best of our knowledge, this is the first use of convolutional neural networks, in particular mask-RCNN, on micro-CT and IHC data, to automatically segment and extract anatomical features at the fascicle and single fiber level.…”

A novel statistical methodology for quantifying the spatial arrangements of axons in peripheral nerves

“…A list of the TEM images used in this study is shown in Table 1 . The protocols and techniques followed for nerve sample collection, processing, and imaging are documented in our previous work (Plebani et al, 2022 ). The data is publicly available via NIH-supported SPARC Pennsieve database (Havton et al, 2022 ).…”

“…The data is publicly available via NIH-supported SPARC Pennsieve database (Havton et al, 2022 ). Briefly, the unmyelinated axons in some of these TEM images were manually annotated and used as labeled data to train, validate, test, and evaluate an automated segmentation model based on the U-Net architecture (Ronneberger et al, 2015 ; Plebani et al, 2022 ). The segmentation model is a U-Net with four stages: the convolutional layers have a batch normalization layer followed by a ReLU activation layer, and the bottleneck stage has extra dropout layers between convolutions (Plebani et al, 2022 ).…”

“…We have included in our study additional TEM images from the rat pelvic nerve, another multi-functional major nerve of the autonomic nervous system that supplies sensory and motor axons to the urogenital organs and lower bowel. Both TEM data sets have been published through the SPARC Portal RRID:SCR_017041 under a CC-BY 4.0 license (Plebani et al, 2022 ).…”

“…Cross-sections of large peripheral nerves reveal a variety of components (myelinated and unmyelinated axons, Schwann cells), high-order structures (fascicles and Remak bundles), and raise questions regarding the spatial arrangement of these components, similarities between multiple arrangements, and their relationship to various biological factors including age, sex, and diseases. This study focuses on the unmyelinated axons segmented utilizing our high-throughput deep learning model (Plebani et al, 2022 ). We aim to define a notion of similarity (or dissimilarity) between the spatial arrangements of the unmyelinated axons and quantify the distances between them.…”