A Distribution-aware Semi-Supervised Pipeline for Cost-effective Neuron Segmentation in Volume Electron Microscopy

Semi-supervised learning offers a cost-effective approach for neuron segmentation in electron microscopy (EM) volumes. This technique leverages extensive unlabeled data to regularize supervised training for more robust predictions of neuron boundaries. However, the distribution mismatch between labeled and unlabeled data, due to limited annotations and diverse neuronal structures, can cause unreliable predictions on unlabeled data, thus limiting the generalization of semi-supervised models. In this paper, 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 propose an unsupervised heuristic to select valuable sub-volumes as labeled data based on distribution similarity in a pretrained feature space, ensuring a representative coverage of neuronal structures. At the model level, we introduce an axial-through mixing strategy into anisotropic neuron segmentation and integrate it into a semi-supervised framework. Building on this, we establish cross-view consistency constraints through intra- and inter-mixing of labeled and unlabeled data, which facilitates the learning of shared semantics across distributions. Extensive experiments on public EM datasets from multiple species, imaging modalities, and resolutions demonstrate the effectiveness of our method. Codes and demos are available at https://github.com/yanchaoz/SL-SSNS.