NEURD offers automated proofreading and feature extraction for connectomics

We are in the era of millimetre-scale electron microscopy volumes collected at nanometre resolution<sup>1,2</sup>. Dense reconstruction of cellular compartments in these electron microscopy volumes has been enabled by recent advances in machine learning<sup>3-6</sup>. Automated segmentation methods produce exceptionally accurate reconstructions of cells, but post hoc proofreading is still required to generate large connectomes that are free of merge and split errors. The elaborate 3D meshes of neurons in these volumes contain detailed morphological information at multiple scales, from the diameter, shape and branching patterns of axons and dendrites, down to the fine-scale structure of dendritic spines. However, extracting these features can require substantial effort to piece together existing tools into custom workflows. Here, building on existing open source software for mesh manipulation, we present Neural Decomposition (NEURD), a software package that decomposes meshed neurons into compact and extensively annotated graph representations. With these feature-rich graphs, we automate a variety of tasks such as state-of-the-art automated proofreading of merge errors, cell classification, spine detection, axonal-dendritic proximities and other annotations. These features enable many downstream analyses of neural morphology and connectivity, making these massive and complex datasets more accessible to neuroscience researchers.

Authors

Institutions

Publication Details

Journal
Nature
Published
2025-04-09
DOI
https://doi.org/10.1038/s41586-025-08660-5
Citations
19
Primary Topic
Modular Robots and Swarm Intelligence
Type
article
Field-Weighted Citation Impact
12.13

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

NEURD offers automated proofreading and feature extraction for connectomics

Ágnes L. Bodor, Victoria A. Rose, Lindsey Kitchell, R. Clay Reid et al.
19 citations
Nature
Modular Robots and Swarm Intelligence
12.13
article

NEURD offers automated proofreading and feature extraction for connectomics

Ágnes L. Bodor, Victoria A. Rose, Lindsey Kitchell, R. Clay Reid, Alexander B. Kunin, Sergiy Popovych, JoAnn Buchanan, Szi-chieh Yu, Stelios Papadopoulos, Nuno Maçarico da Costa, Thomas Macrina, Paul G. Fahey, William S. Wong, Zhen Jia, Russel Torres, Sven Dorkenwald, Shanka Subhra Mondal, Emmanouil Froudarakis, Manuel Castro, Leila Elabbady, Fabian H. Sinz, Zhuokun Ding, Nicholas L. Turner, Caitlyn Bishop, H. Sebastian Seung, William Silversmith, Nico Kemnitz, Jingpeng Wu, Edgar Y. Walker, Patricia K. Rivlin, Sam Kinn, Marc Takeno, Daniel J. Bumbarger, Andreas S. Tolias, Shang Mu, Xaq Pitkow, Akhilesh Halageri, Brendan Celii, Daniel Xenes, Jacob Reimer, Casey M Schneider-Mizell, Brock A. Wester, Eric Mitchell, Gayathri Mahalingam, Dan Kapner, Forrest Collman, Barak Nehoran, Erick Cobos, Derrick Brittain, Saumil S. Patel, Kai Li, Eric Wang, J. Alexander Bae, Christos Papadopoulos, Kisuk Lee, Ran Lu, Wenjing Yin, Chris Jordan
article en
19 citations

Abstract

We are in the era of millimetre-scale electron microscopy volumes collected at nanometre resolution<sup>1,2</sup>. Dense reconstruction of cellular compartments in these electron microscopy volumes has been enabled by recent advances in machine learning<sup>3-6</sup>. Automated segmentation methods produce exceptionally accurate reconstructions of cells, but post hoc proofreading is still required to generate large connectomes that are free of merge and split errors. The elaborate 3D meshes of neurons in these volumes contain detailed morphological information at multiple scales, from the diameter, shape and branching patterns of axons and dendrites, down to the fine-scale structure of dendritic spines. However, extracting these features can require substantial effort to piece together existing tools into custom workflows. Here, building on existing open source software for mesh manipulation, we present Neural Decomposition (NEURD), a software package that decomposes meshed neurons into compact and extensively annotated graph representations. With these feature-rich graphs, we automate a variety of tasks such as state-of-the-art automated proofreading of merge errors, cell classification, spine detection, axonal-dendritic proximities and other annotations. These features enable many downstream analyses of neural morphology and connectivity, making these massive and complex datasets more accessible to neuroscience researchers.

NatureVol. 640(8058)
Creighton University (US), Allen Institute for Brain Science (US), Baylor College of Medicine (US), Neurosciences Institute (US), Princeton University (US), University of Washington (US), Johns Hopkins University Applied Physics Laboratory (US), Allen Institute (US), Artificial Intelligence in Medicine (Canada) (CA), Massachusetts Institute of Technology (US), University of Göttingen (DE), Rice University (US), Carnegie Mellon University (US), Foundation for Research and Technology Hellas (GR), Stanford University (US)
National Science Foundation, Intelligence Advanced Research Projects Activity, Interior Business Center, National Institute of Mental Health
Openalex Percentile: Top 1%
Modular Robots and Swarm Intelligence
12.13
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.