Result: mEMbrain: an interactive deep learning MATLAB tool for connectomic segmentation on commodity desktops.

Title:
mEMbrain: an interactive deep learning MATLAB tool for connectomic segmentation on commodity desktops.
Authors:
Pavarino EC; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Yang E; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Dhanyasi N; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Wang MD; Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada.; Department of Molecular Genetics, University of Toronto, Toronto, ON, Canada., Bidel F; Department of Neurobiology, Silberman Institute of Life Sciences, The Hebrew University, Jerusalem, Israel., Lu X; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Yang F; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Francisco Park C; Department of Physics, Harvard University, Cambridge, MA, United States., Bangalore Renuka M; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Drescher B; Department of Biology, University of Massachusetts Amherst, Amherst, MA, United States., Samuel ADT; Department of Physics, Harvard University, Cambridge, MA, United States., Hochner B; Department of Neurobiology, Silberman Institute of Life Sciences, The Hebrew University, Jerusalem, Israel., Katz PS; Department of Biology, University of Massachusetts Amherst, Amherst, MA, United States., Zhen M; Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada., Lichtman JW; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States., Meirovitch Y; Department of Cellular and Molecular Biology, Harvard University, Cambridge, MA, United States.
Source:
Frontiers in neural circuits [Front Neural Circuits] 2023 Jun 15; Vol. 17, pp. 952921. Date of Electronic Publication: 2023 Jun 15 (Print Publication: 2023).
Publication Type:
Journal Article; Research Support, N.I.H., Extramural; Research Support, Non-U.S. Gov't
Language:
English
Journal Info:
Publisher: Frontiers Research Foundation Country of Publication: Switzerland NLM ID: 101477940 Publication Model: eCollection Cited Medium: Internet ISSN: 1662-5110 (Electronic) Linking ISSN: 16625110 NLM ISO Abbreviation: Front Neural Circuits Subsets: MEDLINE
Imprint Name(s):
Original Publication: [Lausanne, Switzerland] : Frontiers Research Foundation, c2007-
MeSH Terms:
Connectome*/methods , Deep Learning*, Animals ; Image Processing, Computer-Assisted/methods ; Software ; Algorithms
Comments:
Update of: bioRxiv. 2023 Apr 17:2023.04.17.537196. doi: 10.1101/2023.04.17.537196.. (PMID: 37131600)
References:
Science. 2023 Mar 10;379(6636):eadd9330. (PMID: 36893230)
Nat Methods. 2018 Aug;15(8):605-610. (PMID: 30013046)
Proc Natl Acad Sci U S A. 2012 Mar 13;109(11):E622; author reply E623. (PMID: 22343282)
Front Neural Circuits. 2018 Nov 13;12:101. (PMID: 30483068)
Nat Methods. 2017 Jul;14(7):691-694. (PMID: 28604722)
Cell. 2022 Mar 17;185(6):1082-1100.e24. (PMID: 35216674)
Neuron. 2015 Sep 23;87(6):1193-1206. (PMID: 26402603)
Front Neural Circuits. 2018 Oct 16;12:88. (PMID: 30386216)
Nature. 2013 Aug 8;500(7461):168-74. (PMID: 23925239)
Commun Integr Biol. 2011 Jan;4(1):26-9. (PMID: 21509172)
Curr Biol. 2008 Mar 11;18(5):337-42. (PMID: 18328706)
Front Comput Neurosci. 2014 Nov 03;8:137. (PMID: 25404913)
Nat Commun. 2021 May 13;12(1):2785. (PMID: 33986261)
Elife. 2020 Dec 14;9:. (PMID: 33315010)
J Exp Biol. 2018 Apr 11;221(Pt 7):. (PMID: 29511068)
Elife. 2020 Sep 07;9:. (PMID: 32880371)
Neural Comput. 2010 Feb;22(2):511-38. (PMID: 19922289)
Cell. 2015 Jul 30;162(3):648-61. (PMID: 26232230)
Sci Rep. 2019 Dec 19;9(1):19413. (PMID: 31857624)
Nature. 2011 Mar 10;471(7337):177-82. (PMID: 21390124)
Nat Neurosci. 2014 Nov;17(11):1448-54. (PMID: 25349911)
J Microsc. 2005 Apr;218(Pt 1):52-61. (PMID: 15817063)
Nat Methods. 2019 Dec;16(12):1226-1232. (PMID: 31570887)
Dev Biol. 1975 Oct;46(2):326-42. (PMID: 1183723)
Cell. 2018 Jul 26;174(3):730-743.e22. (PMID: 30033368)
Nat Methods. 2012 Jun 28;9(7):676-82. (PMID: 22743772)
Nat Methods. 2013 Jun;10(6):494-500. (PMID: 23722208)
Curr Biol. 2022 Jan 10;32(1):176-189.e5. (PMID: 34822765)
Nat Rev Neurosci. 2012 Feb 22;13(5):351-8. (PMID: 22353782)
Nature. 2024 Mar;627(8003):367-373. (PMID: 38383788)
Biol Psychiatry. 2023 Aug 15;94(4):352-360. (PMID: 36740206)
Nat Neurosci. 2011 Jul 10;14(8):1081-8. (PMID: 21743472)
Nature. 2022 Jan;601(7891):98-104. (PMID: 34912123)
Elife. 2023 Jul 06;12:. (PMID: 37410519)
Philos Trans R Soc Lond B Biol Sci. 1986 Nov 12;314(1165):1-340. (PMID: 22462104)
Nat Neurosci. 2018 Sep;21(9):1281-1289. (PMID: 30127430)
Nat Methods. 2021 Jan;18(1):100-106. (PMID: 33318659)
Nature. 2016 Apr 21;532(7599):370-4. (PMID: 27018655)
Curr Biol. 2012 Oct 23;22(20):R887-92. (PMID: 23098601)
Nat Methods. 2015 Apr;12(4):319-22. (PMID: 25686390)
Nat Neurosci. 2020 Nov;23(11):1433-1443. (PMID: 32958923)
Nature. 2021 Aug;596(7871):257-261. (PMID: 34349261)
Front Neuroanat. 2021 Nov 05;15:732520. (PMID: 34819841)
Elife. 2017 May 16;6:. (PMID: 28508746)
Science. 2011 Nov 4;334(6056):618-23. (PMID: 22053041)
Neuron. 2012 Jun 7;74(5):816-29. (PMID: 22681687)
Nat Methods. 2012 Jul;9(7):671-5. (PMID: 22930834)
Front Neural Circuits. 2014 Jun 27;8:68. (PMID: 25018701)
Cell. 2020 Sep 17;182(6):1372-1376. (PMID: 32946777)
Nat Rev Neurosci. 2008 Jun;9(6):417-22. (PMID: 18446160)
Cell Rep. 2019 Nov 26;29(9):2849-2861.e6. (PMID: 31775050)
Grant Information:
P50 MH094271 United States MH NIMH NIH HHS; R01 NS113119 United States NS NINDS NIH HHS; U01 NS108637 United States NS NINDS NIH HHS; U19 NS104653 United States NS NINDS NIH HHS; R01 NS133654 United States NS NINDS NIH HHS; U01 NS123972 United States NS NINDS NIH HHS; K99 MH128891 United States MH NIMH NIH HHS
Contributed Indexing:
Keywords: MATLAB; VAST; affordable connectomics; deep learning; lightweight software; segmentation; semi-automatic neural circuit reconstruction; volume electron microscopy
Entry Date(s):
Date Created: 20230703 Date Completed: 20230705 Latest Revision: 20250723
Update Code:
20250723
PubMed Central ID:
PMC10309043
DOI:
10.3389/fncir.2023.952921
PMID:
37396399
Database:
MEDLINE

Further Information

Connectomics is fundamental in propelling our understanding of the nervous system's organization, unearthing cells and wiring diagrams reconstructed from volume electron microscopy (EM) datasets. Such reconstructions, on the one hand, have benefited from ever more precise automatic segmentation methods, which leverage sophisticated deep learning architectures and advanced machine learning algorithms. On the other hand, the field of neuroscience at large, and of image processing in particular, has manifested a need for user-friendly and open source tools which enable the community to carry out advanced analyses. In line with this second vein, here we propose mEMbrain, an interactive MATLAB-based software which wraps algorithms and functions that enable labeling and segmentation of electron microscopy datasets in a user-friendly user interface compatible with Linux and Windows. Through its integration as an API to the volume annotation and segmentation tool VAST, mEMbrain encompasses functions for ground truth generation, image preprocessing, training of deep neural networks, and on-the-fly predictions for proofreading and evaluation. The final goals of our tool are to expedite manual labeling efforts and to harness MATLAB users with an array of semi-automatic approaches for instance segmentation. We tested our tool on a variety of datasets that span different species at various scales, regions of the nervous system and developmental stages. To further expedite research in connectomics, we provide an EM resource of ground truth annotation from four different animals and five datasets, amounting to around 180 h of expert annotations, yielding more than 1.2 GB of annotated EM images. In addition, we provide a set of four pre-trained networks for said datasets. All tools are available from https://lichtman.rc.fas.harvard.edu/mEMbrain/. With our software, our hope is to provide a solution for lab-based neural reconstructions which does not require coding by the user, thus paving the way to affordable connectomics.
(Copyright © 2023 Pavarino, Yang, Dhanyasi, Wang, Bidel, Lu, Yang, Francisco Park, Bangalore Renuka, Drescher, Samuel, Hochner, Katz, Zhen, Lichtman and Meirovitch.)

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.