Overview
The source code is publicly available on GitHub.
Paper
The paper is available on bioRxiv.
Trained models
This section contains the links to model with the best performance reported in the paper: the 3D U-Net trained with a positive-unlabeled learning ratio of 1-64.
Datasets
We provide the links to the datasets used in the paper below. We also provide a very small subset of training data that can be downloaded from here and a subset of the testing data that can be downloaded from here.
To train model on the mSCT videos with different PU ratios, refer to the documentation in the GitHub repository.
Individual Files
We also provide the list of individual of the testing dataset only.
| File Name | Download |
|---|
Acknowledgments
We acknowledge the Neuronal Cell Culture Platform of the CERVO Brain Research Center for the preparation of the dissociated hippocampal cultures. We thank Annette Schwerdtfeger for careful proofreading of the manuscript. Funding was provided by grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) (RGPIN-06704-2019 to F.L.C. and RGPIN-2017-06171 to P.D.K.), Fonds de Recherche Nature et Technologie (FRQNT) Team Grant (2021-PR-284335 to F.L.C, C.G. and P.D.K.), the Canadian Institute for Health Research (CIHR) (471107 to F.L.C. and P.D.K.), a New Frontiers in Research Fund Exploration Grant (NFRFE-2020-00933 to F.L.C., C.G. and R.H.), and the Neuronex Initiative (National Science Foundation 2014862, Fonds de recherche du Québec - Santé 295824 to F.L.C. and P.D.K.). R.~H. acknowledges support from CIFAR, the Azrieli and Alfred. P. Sloan foundations and the Connaught Fund. C.G. is a CIFAR AI-Chair and F.L.C. is a Canada Research Chair Tier II. A.B. was supported by a scholarship from NSERC. F.B. was supported by scholarships from FRQNT, NeuroQuébec, and the Québec Bio-Imaging Network. A.B., F.B., and T.W. were awarded an excellence scholarship from the FRQNT strategic cluster UNIQUE.
