A novel retinal ganglion cell quantification tool based on deep learning.


Journal

Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288

Informations de publication

Date de publication:
12 01 2021
Historique:
received: 31 07 2020
accepted: 15 12 2020
entrez: 13 1 2021
pubmed: 14 1 2021
medline: 11 8 2021
Statut: epublish

Résumé

Glaucoma is a disease associated with the loss of retinal ganglion cells (RGCs), and remains one of the primary causes of blindness worldwide. Major research efforts are presently directed towards the understanding of disease pathogenesis and the development of new therapies, with the help of rodent models as an important preclinical research tool. The ultimate goal is reaching neuroprotection of the RGCs, which requires a tool to reliably quantify RGC survival. Hence, we demonstrate a novel deep learning pipeline that enables fully automated RGC quantification in the entire murine retina. This software, called RGCode (Retinal Ganglion Cell quantification based On DEep learning), provides a user-friendly interface that requires the input of RBPMS-immunostained flatmounts and returns the total RGC count, retinal area and density, together with output images showing the computed counts and isodensity maps. The counting model was trained on RBPMS-stained healthy and glaucomatous retinas, obtained from mice subjected to microbead-induced ocular hypertension and optic nerve crush injury paradigms. RGCode demonstrates excellent performance in RGC quantification as compared to manual counts. Furthermore, we convincingly show that RGCode has potential for wider application, by retraining the model with a minimal set of training data to count FluoroGold-traced RGCs.

Identifiants

pubmed: 33436866
doi: 10.1038/s41598-020-80308-y
pii: 10.1038/s41598-020-80308-y
pmc: PMC7804414
doi:

Types de publication

Journal Article Research Support, Non-U.S. Gov't

Langues

eng

Sous-ensembles de citation

IM

Pagination

702

Références

Mead, B. & Tomarev, S. Evaluating retinal ganglion cell loss and dysfunction. Exp. Eye Res. 151, 96–106 (2016).
pubmed: 27523467 pmcid: 5045805 doi: 10.1016/j.exer.2016.08.006
Guo, L. & Cordeiro, M. F. Assessment of neuroprotection in the retina with DARC. Prog. Brain Res. 173, 437–450 (2008).
pubmed: 18929126 pmcid: 2603274 doi: 10.1016/S0079-6123(08)01130-8
Köbbert, C. et al. Current concepts in neuroanatomical tracing. Prog. Neurobiol. 62, 327–351 (2000).
pubmed: 10856608 doi: 10.1016/S0301-0082(00)00019-8
Abdel-Majid, R. M., Archibald, M. L., Tremblay, F. & Baldridge, W. H. Tracer coupling of neurons in the rat retina inner nuclear layer labeled by Fluorogold. Brain Res. 1063, 114–120 (2005).
pubmed: 16263096 doi: 10.1016/j.brainres.2005.09.046
Peinado-Ramon, P., Salvador, M., Villegas-Perez, M. P. & Vidal-Sanz, M. Effects of axotomy and intraocular administration of NT-4, NT-3, and brain-derived neurotrophic factor on the survival of adult rat retinal ganglion cells. A quantitative in vivo study. Invest. Ophthalmol. Vis. Sci. 37, 489–500 (1996).
pubmed: 8595949
Nadal-Nicolás, F. M. et al. Brn3a as a marker of retinal ganglion cells: qualitative and quantitative time course studies in naive and optic nerve-injured retinas. Invest. Ophthalmol. Vis. Sci. 50, 3860–3868 (2009).
pubmed: 19264888 doi: 10.1167/iovs.08-3267
Nadal-Nicolas, F. M. et al. Whole number, distribution and co-expression of Brn3 transcription factors in retinal ganglion cells of adult albino and pigmented rats. PLoS ONE 7, e49830 (2012).
pubmed: 23166779 pmcid: 3500320 doi: 10.1371/journal.pone.0049830
Rodriguez, A. R., de Sevilla Müller, L. P. & Brecha, N. C. The RNA binding protein RBPMS is a selective marker of ganglion cells in the mammalian retina. J. Comp. Neurol.522, 1411–1443 (2014).
Tran, N. M. et al. Single-cell profiles of retinal ganglion cells differing in resilience to injury reveal neuroprotective genes. Neuron 104, 1039–1055 (2019).
pubmed: 31784286 pmcid: 6923571 doi: 10.1016/j.neuron.2019.11.006
Rheaume, B. A. et al. Single cell transcriptome profiling of retinal ganglion cells identifies cellular subtypes. Nat. Commun. 9, 2759 (2018).
pubmed: 30018341 pmcid: 6050223 doi: 10.1038/s41467-018-05134-3
Kwong, J. M., Caprioli, J. & Piri, N. RNA binding protein with multiple splicing: a new marker for retinal ganglion cells. Invest. Ophthalmol. Vis. Sci. 51, 1052–1058 (2010).
pubmed: 19737887 pmcid: 3979483 doi: 10.1167/iovs.09-4098
Baden, T. et al. The functional diversity of retinal ganglion cells in the mouse. Nature 529, 345–350 (2016).
pubmed: 26735013 pmcid: 4724341 doi: 10.1038/nature16468
Bae, J. A. et al. Digital museum of retinal ganglion cells with dense anatomy and physiology. Cell 173, 1293–1306 (2018).
pubmed: 29775596 pmcid: 6556895 doi: 10.1016/j.cell.2018.04.040
Zhao, D. et al. Characterization of the circumlimbal suture model of chronic iop elevation in mice and assessment of changes in gene expression of stretch sensitive channels. Front. Neurosci. 11, 41 (2017).
pubmed: 28239332 pmcid: 5301305 doi: 10.3389/fnins.2017.00041
Liu, R., Wang, Y., Pu, M. & Gao, J. Effect of alpha lipoic acid on retinal ganglion cell survival in an optic nerve crush model. Mol. Vis. 22, 1122–1136 (2016).
pubmed: 27703307 pmcid: 5040455
Wang, W. et al. Programmed cell death-1 is expressed in large retinal ganglion cells and is upregulated after optic nerve crush. Exp. Eye Res. 140, 1–9 (2015).
pubmed: 26277582 pmcid: 5420326 doi: 10.1016/j.exer.2015.08.008
Guymer, C. et al. Software for quantifying and batch processing images of brn3a and RBPMS immunolabelled retinal ganglion cells in retinal wholemounts. Transl. Vis. Sci. Technol. 9, 28 (2020).
pubmed: 32821525 pmcid: 7409150 doi: 10.1167/tvst.9.6.28
Moen, E. et al. Deep learning for cellular image analysis. Nat. Methods 16, 1233–1246 (2019).
pubmed: 31133758 doi: 10.1038/s41592-019-0403-1 pmcid: 8759575
Ronneberger, O., Fischer, P. & Brox, T. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-assisted Intervention, 234–241 (Springer, 2015).
De Fauw, J. et al. Clinically applicable deep learning for diagnosis and referral in retinal disease. Nat. Med. 24, 1342–1350 (2018).
pubmed: 30104768 doi: 10.1038/s41591-018-0107-6
Ritch, M. D. et al. Axonet: a deep learning-based tool to count retinal ganglion cell axons. Sci. Rep. 10, 8034 (2020).
pubmed: 32415269 pmcid: 7228952 doi: 10.1038/s41598-020-64898-1
Falk, T. et al. U-Net: deep learning for cell counting, detection, and morphometry. Nat. Methods 16, 67–70 (2019).
pubmed: 30559429 doi: 10.1038/s41592-018-0261-2
Shermin, T., Murshed, M., Lu, G. & Teng, S. W. Transfer learning using classification layer features of CNN. arXiv:1811.07459 (2018).
Hofbauer, A. & Dräger, U. C. Depth segregation of retinal ganglion cells projecting to mouse superior colliculus. J. Comp. Neurol. 234, 465–474 (1985).
pubmed: 3988995 doi: 10.1002/cne.902340405
Petros, T. J., Rebsam, A. & Mason, C. A. Retinal axon growth at the optic chiasm: to cross or not to cross. Annu. Rev. Neurosci. 31, 295–315 (2008).
pubmed: 18558857 doi: 10.1146/annurev.neuro.31.060407.125609
Campbell, C. G., Ting, D. S., Keane, P. A. & Foster, P. J. The potential application of artificial intelligence for diagnosis and management of glaucoma in adults. Br. Med. Bull. 134, 21–33 (2020).
pubmed: 32518944 doi: 10.1093/bmb/ldaa012
Balyen, L. & Peto, T. Promising artificial intelligence-machine learning-deep learning algorithms in ophthalmology. Asia Pac. J. Ophthal. 8, 264–272 (2019).
Zheng, C., Johnson, T. V., Garg, A. & Boland, M. V. Artificial intelligence in glaucoma. Curr. Opin. Ophthalmol. 30, 97–103 (2019).
pubmed: 30562242 doi: 10.1097/ICU.0000000000000552
Martin, K. R. et al. Use of machine learning on contact lens sensor-derived parameters for the diagnosis of primary open-angle glaucoma. Am. J. Ophthalmol. 194, 46–53 (2018).
pubmed: 30053471 doi: 10.1016/j.ajo.2018.07.005
Normando, E. M. et al. A CNN-aided method to predict glaucoma progression using DARC (detection of apoptosing retinal cells). Expert Rev. Mol. Diagn.1–12 (2020).
Li, W. et al. Automatic anterior chamber angle measurement for ultrasound biomicroscopy using deep learning. J. Glaucoma 29, 81–85 (2020).
pubmed: 31790065 doi: 10.1097/IJG.0000000000001411
Xu, B. Y. et al. Deep learning classifiers for automated detection of gonioscopic angle closure based on anterior segment OCT images. Am. J. Ophthalmol. 208, 273–280 (2019).
pubmed: 31445003 pmcid: 6888901 doi: 10.1016/j.ajo.2019.08.004
Hedberg-Buenz, A. et al. Quantitative measurement of retinal ganglion cell populations via histology-based random forest classification. Exp. Eye Res. 146, 370–385 (2016).
pubmed: 26474494 doi: 10.1016/j.exer.2015.09.011
Alahmari, S. S. et al. Automated cell counts on tissue sections by deep learning and unbiased stereology. J. Chem. Neuroanat. 96, 94–101 (2019).
pubmed: 30594529 doi: 10.1016/j.jchemneu.2018.12.010
Phoulady, H. A., Goldgof, D., Hall, L. O. & Mouton, P. R. Automatic ground truth for deep learning stereology of immunostained neurons and microglia in mouse neocortex. J. Chem. Neuroanat. 98, 1–7 (2019).
doi: 10.1016/j.jchemneu.2019.02.006
Pham, B. et al. Cell counting and segmentation of immunohistochemical images in the spinal cord: Comparing deep learning and traditional approaches. In 2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 842–845 (IEEE, 2018).
Penttinen, A.-M. et al. Implementation of deep neural networks to count dopamine neurons in substantia nigra. Eur. J. Neurosci. 48, 2354–2361 (2018).
pubmed: 30144349 pmcid: 6585833 doi: 10.1111/ejn.14129
Kyriazis, A. D. et al. An end-to-end system for automatic characterization of iba1 immunopositive microglia in whole slide imaging. Neuroinformatics 17, 373–389 (2019).
pubmed: 30406865 doi: 10.1007/s12021-018-9405-x
Suleymanova, I. et al. A deep convolutional neural network approach for astrocyte detection. Sci. Rep. 8, 1–7 (2018).
doi: 10.1038/s41598-018-31284-x
Dibas, A., Millar, C., Al-Farra, A. & Yorio, T. Neuroprotective effects of psalmotoxin-1, an acid-sensing ion channel (ASIC) inhibitor, in ischemia reperfusion in mouse eyes. Curr. Eye Res. 43, 921–933 (2018).
pubmed: 29595330 doi: 10.1080/02713683.2018.1454478
Stankowska, D. L. et al. Hybrid compound SA-2 is neuroprotective in animal models of retinal ganglion cell death. Invest. Ophthalmol. Vis. Sci. 60, 3064–3073 (2019).
pubmed: 31348824 doi: 10.1167/iovs.18-25999
Cone, F. E., Gelman, S. E., Son, J. L., Pease, M. E. & Quigley, H. A. Differential susceptibility to experimental glaucoma among 3 mouse strains using bead and viscoelastic injection. Exp. Eye Res. 91, 415–424 (2010).
pubmed: 20599961 pmcid: 2954410 doi: 10.1016/j.exer.2010.06.018
Frankfort, B. J. et al. Elevated intraocular pressure causes inner retinal dysfunction before cell loss in a mouse model of experimental glaucoma. Invest. Ophthalmol. Vis. Sci. 54, 762–770 (2013).
pubmed: 23221072 pmcid: 3562118 doi: 10.1167/iovs.12-10581
Mukai, R. et al. Mouse model of ocular hypertension with retinal ganglion cell degeneration. PLoS ONE 14, e0208713 (2019).
pubmed: 30640920 pmcid: 6331128 doi: 10.1371/journal.pone.0208713
Blandford, S. N. et al. Retinal characterization of the Thy1-GCaMP3 transgenic mouse line after optic nerve transection. Invest. Ophthalmol. Vis. Sci. 60, 183–191 (2019).
pubmed: 30640971 doi: 10.1167/iovs.18-25861
Omodaka, K. et al. Neuroprotective effect against axonal damage-induced retinal ganglion cell death in apolipoprotein e-deficient mice through the suppression of kainate receptor signaling. Brain Res. 1586, 203–212 (2014).
pubmed: 25160129 doi: 10.1016/j.brainres.2014.08.053
Yamamoto, K. et al. The novel Rho kinase (ROCK) inhibitor K-115: a new candidate drug for neuroprotective treatment in glaucoma. Invest. Ophthalmol. Vis. Sci. 55, 7126–7136 (2014).
pubmed: 25277230 doi: 10.1167/iovs.13-13842
Ryu, M. et al. Critical role of calpain in axonal damage-induced retinal ganglion cell death. J. Neurosci. 90, 802–815 (2012).
Tsuda, S. et al. Real-time imaging of RGC death with a cell-impermeable nucleic acid dyeing compound after optic nerve crush in a murine model. Exp. Eye Res. 146, 179–188 (2016).
pubmed: 27013099 doi: 10.1016/j.exer.2016.03.017
Xia, X. et al. Protection of pattern electroretinogram and retinal ganglion cells by oncostatin m after optic nerve injury. PLoS ONE 9, e108524 (2014).
pubmed: 25243471 pmcid: 4171539 doi: 10.1371/journal.pone.0108524
Himori, N. et al. Critical neuroprotective roles of heme oxygenase-1 induction against axonal injury-induced retinal ganglion cell death. J. Neurosci. 92, 1134–1142 (2014).
De Groef, L. et al. Differential visual system organization and susceptibility to experimental models of optic neuropathies in three commonly used mouse strains. Exp. Eye Res. 145, 235–247 (2016).
pubmed: 26791081 doi: 10.1016/j.exer.2016.01.006
Dräger, U. & Olsen, J. F. Ganglion cell distribution in the retina of the mouse. Invest. Ophthalmol. Vis. Sci. 20, 285–293 (1981).
pubmed: 6162818
Jeon, C.-J., Strettoi, E. & Masland, R. H. The major cell populations of the mouse retina. J. Neurosci. 18, 8936–8946 (1998).
pubmed: 9786999 pmcid: 6793518 doi: 10.1523/JNEUROSCI.18-21-08936.1998
Salinas-Navarro, M. et al. Retinal ganglion cell population in adult albino and pigmented mice: a computerized analysis of the entire population and its spatial distribution. Vis. Res. 49, 637–647 (2009).
pubmed: 19948111 doi: 10.1016/j.visres.2009.01.010
Jakobs, T. C., Libby, R. T., Ben, Y., John, S. W. & Masland, R. H. Retinal ganglion cell degeneration is topological but not cell type specific in DBA/2J mice. J. Cell Biol. 171, 313–325 (2005).
pubmed: 16247030 pmcid: 2171185 doi: 10.1083/jcb.200506099
Geeraerts, E. et al. A freely available semi-automated method for quantifying retinal ganglion cells in entire retinal flatmounts. Exp. Eye Res. 147, 105–113 (2016).
pubmed: 27107795 doi: 10.1016/j.exer.2016.04.010
Guo, C. et al. A murine glaucoma model induced by rapid in vivo photopolymerization of hyaluronic acid glycidyl methacrylate. PLoS ONE 13, e0196529 (2018).
pubmed: 29949582 pmcid: 6021085 doi: 10.1371/journal.pone.0196529
Geeraerts, E. et al. Optogenetic stimulation of the superior colliculus confers retinal neuroprotection in a mouse glaucoma model. J. Neurosci. 39, 2313–2325 (2019).
pubmed: 30655352 pmcid: 6433760 doi: 10.1523/JNEUROSCI.0872-18.2018
Valiente-Soriano, F. J. et al. Effects of ocular hypertension in the visual system of pigmented mice. PLoS ONE 10, e0121134 (2015).
pubmed: 25811653 pmcid: 4374934 doi: 10.1371/journal.pone.0121134
Schlamp, C. L., Li, Y., Dietz, J. A., Janssen, K. T. & Nickells, R. W. Progressive ganglion cell loss and optic nerve degeneration in DBA/2J mice is variable and asymmetric. BMC Neurosci. 7, 66 (2006).
pubmed: 17018142 pmcid: 1621073 doi: 10.1186/1471-2202-7-66
Soto, I. et al. Retinal ganglion cells downregulate gene expression and lose their axons within the optic nerve head in a mouse glaucoma model. J. Neurosci. 28, 548–561 (2008).
pubmed: 18184797 pmcid: 6670511 doi: 10.1523/JNEUROSCI.3714-07.2008
DellaSantina, L., Inman, D. M., Lupien, C. B., Horner, P. J. & Wong, R. O. Differential progression of structural and functional alterations in distinct retinal ganglion cell types in a mouse model of glaucoma. J. Neurosci. 33, 17444–17457 (2013).
doi: 10.1523/JNEUROSCI.5461-12.2013
Galindo-Romero, C. et al. Axotomy-induced retinal ganglion cell death in adult mice: quantitative and topographic time course analyses. Exp. Eye Res. 92, 377–387 (2011).
pubmed: 21354138 doi: 10.1016/j.exer.2011.02.008
Pérez de Lara, M. . J. et al. Assessment of inner retina dysfunction and progressive ganglion cell loss in a mouse model of glaucoma. Exp. Eye Res. 122, 40–49 (2014).
pubmed: 24631335 doi: 10.1016/j.exer.2014.02.022
Ito, Y. A., Belforte, N., Vargas, J. L. C. & Di Polo, A. A magnetic microbead occlusion model to induce ocular hypertension-dependent glaucoma in mice. J. Vis. Exp. 109, e53731 (2016).
Nadal-Nicolás, F. M., Salinas-Navarro, M., Vidal-Sanz, M. & Agudo-Barriuso, M. Two methods to trace retinal ganglion cells with fluorogold: from the intact optic nerve or by stereotactic injection into the optic tract. Exp. Eye Res. 131, 12–19 (2015).
pubmed: 25482219 doi: 10.1016/j.exer.2014.12.005
Schindelin, J. et al. Fiji: an open-source platform for biological-image analysis. Nat. Methods 9, 676–682 (2012).
pubmed: 22743772 doi: 10.1038/nmeth.2019
R Core Team. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2019).

Auteurs

Luca Masin (L)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Marie Claes (M)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Steven Bergmans (S)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Lien Cools (L)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Lien Andries (L)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Benjamin M Davis (BM)

Glaucoma and Retinal Neurodegenerative Disease Research Group, Institute of Ophthalmology, University College London, London, UK.
Central Laser Facility, Science and Technologies Facilities Council, UK Research and Innovation, Didcot, Oxfordshire, UK.

Lieve Moons (L)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium.

Lies De Groef (L)

Department of Biology, Neural Circuit Development and Regeneration Research Group, KU Leuven, Leuven, Belgium. lies.degroef@kuleuven.be.

Articles similaires

Robotic Surgical Procedures Animals Humans Telemedicine Models, Animal

Odour generalisation and detection dog training.

Lyn Caldicott, Thomas W Pike, Helen E Zulch et al.
1.00
Animals Odorants Dogs Generalization, Psychological Smell

Selecting optimal software code descriptors-The case of Java.

Yegor Bugayenko, Zamira Kholmatova, Artem Kruglov et al.
1.00
Software Algorithms Programming Languages
Animals TOR Serine-Threonine Kinases Colorectal Neoplasms Colitis Mice

Classifications MeSH