{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"institution":[{"name":"bioRxiv"}],"indexed":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T16:49:13Z","timestamp":1786121353735,"version":"build-2736575974"},"posted":{"date-parts":[[2026,8,4]]},"group-title":"Neuroscience","reference-count":23,"publisher":"openRxiv","license":[{"start":{"date-parts":[[2026,8,4]],"date-time":"2026-08-04T00:00:00Z","timestamp":1785801600000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by-nc-nd\/4.0\/"}],"funder":[{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"crossref","award":["R01 NS121761"],"award-info":[{"award-number":["R01 NS121761"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000002","name":"National Institutes of Health","doi-asserted-by":"crossref","award":["UM1 NS132173"],"award-info":[{"award-number":["UM1 NS132173"]}],"id":[{"id":"10.13039\/100000002","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"accepted":{"date-parts":[[2026,8,4]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                <jats:p>Reconstructing neuronal morphology from large 3D microscopy volumes is essential for quantitative neuro-science but remains challenging due to noise, low contrast, and complex branching geometry. We present an automated neuron tracing method that formulates reconstruction as a sequential decision-making problem and trains a 3D convolutional policy via imitation learning. The key technical contribution is an online expert-action retrieval scheme that derives locally valid continuation directions directly from a gold-standard SWC tree, including at bifurcations. Combined with DAgger-style dataset aggregation, the policy is trained under its own state distribution without requiring handcrafted reward functions or a separate segmentation-to-skeleton postprocessing stage. We package the approach into an end-to-end tracing pipeline with optional path refinement operations and tools for quantitative evaluation and interactive editing. On the BigNeuron Gold166 benchmark, our method achieves state-of-the-art reconstruction accuracy, outperforming all competing challenge methods under both morphology-based aggregate metrics and geometric distance comparisons.<\/jats:p>","DOI":"10.64898\/2026.07.29.741595","type":"posted-content","created":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T07:25:16Z","timestamp":1785914716000},"source":"Crossref","is-referenced-by-count":0,"title":["Automated Neuron Tracing with Imitation Learning"],"prefix":"10.64898","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8857-3492","authenticated-orcid":false,"given":"Bryson","family":"Gray","sequence":"first","affiliation":[{"name":"University of California, Los Angeles"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4607-6807","authenticated-orcid":false,"given":"Daniel","family":"Tward","sequence":"additional","affiliation":[{"name":"University of California, Los Angeles"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"54368","reference":[{"key":"2026080621403637000_2026.07.29.741595v1.1","doi-asserted-by":"publisher","DOI":"10.1523\/JNEUROSCI.2055-07.2007"},{"key":"2026080621403637000_2026.07.29.741595v1.2","doi-asserted-by":"crossref","unstructured":"Balaram, S. , Arulkumaran, K. , Dai, T. , Bharath, A.A. , 2019. A maximum entropy deep reinforcement learning neural tracker, in: Machine Learning in Medical Imaging, Springer International Publishing. pp. 400\u2013408.","DOI":"10.1007\/978-3-030-32692-0_46"},{"key":"2026080621403637000_2026.07.29.741595v1.3","doi-asserted-by":"crossref","unstructured":"Banerjee, S. , Stam, C. , Tward, D.J. , Savoia, S. , Wang, Y. , Mitra, P.P. , 2025. Skeletonization of neuronal processes using discrete morse techniques from computational topology. ArXiv, arXiv\u20132505.","DOI":"10.1101\/2025.05.12.653477"},{"key":"2026080621403637000_2026.07.29.741595v1.4","doi-asserted-by":"publisher","DOI":"10.1016\/j.neuroimage.2006.09.018"},{"key":"2026080621403637000_2026.07.29.741595v1.5","unstructured":"Cai, H. , Xu, X. , Lu, J. , Lichtman, J. , Yung, W.H. , 2006. Neuron tracing using shortest path graph, in: MICCAI."},{"key":"2026080621403637000_2026.07.29.741595v1.6","doi-asserted-by":"crossref","first-page":"568","DOI":"10.1016\/j.neuroimage.2008.05.022","article-title":"Quantitative tractography metrics of white matter integrity in diffusion-tensor MRI","volume":"42","year":"2008","journal-title":"Neuroimage"},{"key":"2026080621403637000_2026.07.29.741595v1.7","unstructured":"Dai, X.e.a. , 2019. Deep reinforcement learning for subpixel neural tracking, in: MICCAI Workshop on Machine Learning in Medical Imaging."},{"key":"2026080621403637000_2026.07.29.741595v1.8","unstructured":"Han, Y. , Chen, Z. , Zhao, Y. , Xu, C. , Shao, Y. , Peng, Y. , Mu, Y. , Lian, W. , 2026. DexHiL: A human-in-the-loop framework for vision-language-action model post-training in dexterous manipulation. arXiv [cs.RO]."},{"key":"2026080621403637000_2026.07.29.741595v1.9","doi-asserted-by":"crossref","unstructured":"He, K. , Zhang, X. , Ren, S. , Sun, J. , 2015. Delving deep into rectifiers: Surpassing human-level performance on imagenet classification, in: Proceedings of the IEEE international conference on computer vision, pp. 1026\u20131034.","DOI":"10.1109\/ICCV.2015.123"},{"key":"2026080621403637000_2026.07.29.741595v1.10","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA.2019.8793698"},{"key":"2026080621403637000_2026.07.29.741595v1.11","doi-asserted-by":"crossref","first-page":"1549","DOI":"10.1038\/s41467-019-09515-0","article-title":"Precise segmentation of densely interweaving neuron clusters using g-cut","volume":"10","year":"2019","journal-title":"Nature communications"},{"key":"2026080621403637000_2026.07.29.741595v1.12","unstructured":"Li, R.e.a. , 2017. Deepneuron: An open deep learning toolbox for neuron tracing. Brain Informatics."},{"key":"2026080621403637000_2026.07.29.741595v1.13","doi-asserted-by":"crossref","first-page":"824","DOI":"10.1038\/s41592-023-01848-5","article-title":"BigNeuron: a resource to benchmark and predict performance of algorithms for automated tracing of neurons in light microscopy datasets","volume":"20","year":"2023","journal-title":"Nat. Methods"},{"key":"2026080621403637000_2026.07.29.741595v1.14","first-page":"11","article-title":"A review of automated neuron reconstruction techniques","volume":"13","year":"2015","journal-title":"Neuroinformatics"},{"key":"2026080621403637000_2026.07.29.741595v1.15","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btr237"},{"key":"2026080621403637000_2026.07.29.741595v1.16","doi-asserted-by":"crossref","unstructured":"Ronneberger, O. , Fischer, P. , Brox, T. , 2015. U-net: Convolutional networks for biomedical image segmentation, in: MICCAI.","DOI":"10.1007\/978-3-319-24574-4_28"},{"key":"2026080621403637000_2026.07.29.741595v1.17","unstructured":"Ross, S. , Gordon, G. , Bagnell, D. , 2011. A reduction of imitation learning and structured prediction to no-regret online learning, in: AISTATS."},{"key":"2026080621403637000_2026.07.29.741595v1.18","doi-asserted-by":"publisher","DOI":"10.1038\/nprot.2008.51"},{"key":"2026080621403637000_2026.07.29.741595v1.19","doi-asserted-by":"publisher","DOI":"10.1002\/ima.22005"},{"key":"2026080621403637000_2026.07.29.741595v1.20","doi-asserted-by":"publisher","DOI":"10.1016\/j.cell.2019.07.042"},{"key":"2026080621403637000_2026.07.29.741595v1.21","doi-asserted-by":"crossref","unstructured":"Xu, X. , Hou, Y. , Xin, C. , Liu, Z. , Song, S. , 2025. Compliant residual DAgger: Improving real-world contact-rich manipulation with human corrections. arXiv [cs.RO].","DOI":"10.52202\/085713-4660"},{"key":"2026080621403637000_2026.07.29.741595v1.22","doi-asserted-by":"crossref","first-page":"1936","DOI":"10.1038\/s41592-024-02401-8","article-title":"Collaborative augmented reconstruction of 3d neuron morphology in mouse and human brains","volume":"21","year":"2024","journal-title":"Nature Methods"},{"key":"2026080621403637000_2026.07.29.741595v1.23","unstructured":"Zhang, Y.e.a. , 2010. Rivulet: 3d neuron morphology tracing with iterative backtracking. Neuroinformatics."}],"container-title":[],"original-title":[],"link":[{"URL":"https:\/\/syndication.highwire.org\/content\/doi\/10.64898\/2026.07.29.741595","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,7]],"date-time":"2026-08-07T04:41:35Z","timestamp":1786077695000},"score":1,"resource":{"primary":{"URL":"http:\/\/biorxiv.org\/lookup\/doi\/10.64898\/2026.07.29.741595"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,4]]},"references-count":23,"URL":"https:\/\/doi.org\/10.64898\/2026.07.29.741595","relation":{},"subject":[],"published":{"date-parts":[[2026,8,4]]},"subtype":"preprint"}}