Automatic 3d segmentation of dendrites in brain images
A Fully Automatic Pipeline for 3D Dendrite Instance Segmentation in SBF-SEM
Computer Vision and Pattern Recognition
Summary
Studying tiny tree-like branches in brain cells called dendrites helps researchers understand brain changes. The authors created a fully automatic method to find and separate these dendrites in 3D from electron microscope images. Their system combines different machine learning steps to work without human help and can capture fine details. This makes it easier to study brain structures and potentially understand diseases better.
What this means in practice
- •For neuroscience imaging teams: Automatically segment and reconstruct individual dendrites in 3D brain microscopy images to analyze structural plasticity and pathology.
- •For biomedical image analysis developers: Integrate a multi-step pipeline combining detection, mask refinement, and 3D linking to improve dendrite segmentation accuracy on volumetric datasets.
Authors
Zewen Zhuo, Ilya Belevich, Eija Jokitalo, Alejandra Sierra, Jussi Tohka
Abstract
Accurate three-dimensional (3D) reconstruction of individual dendrites in serial block-face scanning electron microscopy (SBF-SEM) is essential for quantifying structural plasticity in the brain, yet manual annotation at scale is infeasible. We present a fully automatic pipeline for 3D dendrite instance segmentation that unifies YOLOv6-guided Segment Anything Model (SAM) prompting on downsampled slices, iterative two-dimensional mask refinement, random forest 3D instance linking, and instance-aware high-resolution refinement using nnU-Net at native resolution into a single system requiring no manual prompting at inference. Applied to hippocampal CA1 SBF-SEM datasets from a control rat and a pilocarpine- induced epileptic rat, our pipeline reconstructs coherent, well- separated dendrites with high semantic accuracy (Dice 0.93 and 0.91) and strong instance-level performance on control tissue, while analysis of the more challenging epileptic tissue identifies instance recognition in dense regions as the principal remaining limitation. The high-resolution refinement stage recovers thin dendritic protrusions, providing a basis for downstream spine- level analysis. Code is available at https://github.com/ ZE-WEN/dendrite-3d-instance-seg.