A distribution-aware semi-supervised pipeline for cost-effective neuron segmentation

TL;DR

This paper introduces a new semi-supervised method for segmenting neurons in electron microscopy images, which reduces the need for extensive manual labeling.

Contribution

The novel pipeline addresses distribution mismatch by using unsupervised selection and mixed-view consistency regularization for better segmentation.

Findings

The proposed pipeline improves segmentation by selecting representative sub-volumes for annotation.

Mixed-view consistency regularization enhances model performance across diverse EM datasets.

The method reduces proofreading efforts in large-scale connectomic reconstructions.

Abstract

Semi-supervised learning offers a cost-effective approach for neuron segmentation in electron microscopy (EM) volumes. This technique leverages unlabeled data to regularize supervised training for robust neuron boundary prediction. However, distribution mismatch between labeled and unlabeled data, caused by limited annotations and diverse neuronal structures, limits model generalization. In this study, we develop a distribution-aware pipeline to address the inherent mismatch issue and enhance semi-supervised neuron segmentation in EM volumes. At the data level, we select representative sub-volumes for annotation using an unsupervised measure of distributional similarity, ensuring broad coverage of neuronal structures. At the model level, we encourage consistent predictions across mixed views of labeled and unlabeled data. This design prompts the network to align feature distributions and learn shared semantics. Experiments on diverse EM datasets demonstrate the effectiveness of our method, which holds the potential to reduce proofreading demands and accelerate large-scale connectomic reconstruction efforts.

•Addressing the distribution mismatch inherent in semi-supervised neuron segmentation•An unsupervised heuristic for selecting representative sub-volumes in EM data•A novel semi-supervised framework using mixed-view consistency regularization•Extensive experiments across diverse EM imaging modalities and voxel resolutions

Addressing the distribution mismatch inherent in semi-supervised neuron segmentation

An unsupervised heuristic for selecting representative sub-volumes in EM data

A novel semi-supervised framework using mixed-view consistency regularization

Extensive experiments across diverse EM imaging modalities and voxel resolutions

Health sciences

Figures