Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Jul 30.
Published in final edited form as: Proc SPIE Int Soc Opt Eng. 2017 Mar 3;10060:100601I. doi: 10.1117/12.2267477

Synchrotron microCT imaging of soft tissue in juvenile zebrafish reveals retinotectal projections

Xuying Xin 1,2,3, Darin Clark 4, Khai Chung Ang 1,2,3, Damian B van Rossum 1,2,3, Jean Copper 1,2,3, Xianghui Xiao 5, Patrick J La Riviere 6, Keith C Cheng 1,2,3,*
PMCID: PMC7392147  NIHMSID: NIHMS1611833  PMID: 32733117

Abstract

Biomedical research and clinical diagnosis would benefit greatly from full volume determinations of anatomical phenotype. Comprehensive tools for morphological phenotyping are central for the emerging field of phenomics, which requires high-throughput, systematic, accurate, and reproducible data collection from organisms affected by genetic, disease, or environmental variables. Theoretically, complete anatomical phenotyping requires the assessment of every cell type in the whole organism, but this ideal is presently untenable due to the lack of an unbiased 3D imaging method that allows histopathological assessment of any cell type despite optical opacity. Histopathology, the current clinical standard for diagnostic phenotyping, involves the microscopic study of tissue sections to assess qualitative aspects of tissue architecture, disease mechanisms, and physiological state. However, quantitative features of tissue architecture such as cellular composition and cell counting in tissue volumes can only be approximated due to characteristics of tissue sectioning, including incomplete sampling and the constraints of 2D imaging of 5 micron thick tissue slabs. We have used a small, vertebrate organism, the zebrafish, to test the potential of microCT for systematic macroscopic and microscopic morphological phenotyping. While cell resolution is routinely achieved using methods such as light sheet fluorescence microscopy and optical tomography, these methods do not provide the pancellular perspective characteristic of histology, and are constrained by the limited penetration of visible light through pigmented and opaque specimens, as characterizes zebrafish juveniles. Here, we provide an example of neuroanatomy that can be studied by microCT of stained soft tissue at 1.43 micron isotropic voxel resolution. We conclude that synchrotron microCT is a form of 3D imaging that may potentially be adopted towards more reproducible, large-scale, morphological phenotyping of optically opaque tissues. Further development of soft tissue microCT, visualization and quantitative tools will enhance its utility.

Keywords: zebrafish, phenomics, 3D imaging, image visualization, soft tissue imaging, optic nerves, retinotectal projections, synchrotron microCT

1. INTRODUCTION

1.1. Large-scale phenotyping and phenomics.

Phenotype, including anatomical phenotype, is determined by genes and the environment. The systematic study of phenotype across all organisms and across all genes and environmental variables comprises the nascent field of phenomics. Since a single genetic or environmental variable can affect multiple organ systems, any systematic, -omic level attempt to define complete anatomical phenotypes will require a rapid way to define anatomical phenotype across all cell types and tissues at once, in whole model organisms, without requiring gene-specific assays. How do we approach the problem of completely defining complete phenotypes across genes and environmental variables within vertebrates? Since we cannot experiment on humans, what model system and methodologies can we use?

1.2. Zebrafish as a model system.

The zebrafish is a popular model organism for the study of vertebrate biology, being well suited for studies related to vertebrate development1,2, genetic analysis3,4, disease5, 6, toxicology7,8, evolution9,10, and cancer11,12. Its genome has been sequenced13,14 and well-characterized mutant strains are readily available15, 16. In addition, zebrafish share many orthologous genes with humans. Their small size, high fecundity and sophisticated genetic toolbox (e.g., genetic screens17,18, morpholino knockdowns, knockouts19,20 through genome editing and fluorescently tagging proteins in transgenic animals21,22) make the zebrafish an ideal model for studying the functions of vertebrate genes. The international zebrafish community wishes to address the functions of ~20,000 vertebrate genes through the characterization of morphological, physiological and behavioral phenotypes of zebrafish with specific gene deficiencies. Morphological phenotyping of whole vertebrate organisms such as the zebrafish would allow us to capture the temporal effects of single gene deficiencies across organ systems. Indeed, the ability to image across developmental stages is key because many single gene mutations and diseases show clinically relevant phenotypes at older, rather than early development stages (for example, sexual differentiation is only complete at a juvenile life stage).

1.3. Current approaches and challenges for morphological phenotyping.

Imaging at cell resolution allows one to make useful assessments of physiological state and pathophysiological defects across organisms, since all living things are comprised of cell(s)23,24, and disease involves cellular processes that are formally defined through histological studies25. In zebrafish mutations in many, and potentially most genes, affect more than one organ system26. These complexities emphasize the importance of finding a way to image and to objectively phenotype the whole organism at resolutions that allow histological phenotyping of the full range of tissue types.

Currently, morphological phenotypes of zebrafish are captured predominantly by imaging in the visible light spectrum, which includes conventional light microscopy27, fluorescent microscopy28 , 29 and optical projection tomography30. Conventional light microscopy generates color images with high spatial resolution and large fields of view, making it very useful for morphological studies. It is used as the foundation for tissue diagnosis in medicine. However, histology requires tissue sectioning, which is time-consuming and is associated with loss of sample integrity and loss of tissue (i.e., a lack of complete sampling). Furthermore, the reconstruction of 3-dimensional (3D) volumes of tissue through the alignment of separately imaged 2D sections has poor z-axis resolution and is associated with significant distortion and a loss of cellular detail. Fluorescent microscopy captures spatial data with 3D imaging and employs fluorescence to reveal ultrastructural features and patterns of protein deposition. While fluorescence-based 3D imaging at cellular resolution is now routine in zebrafish embryos31 and larvae32, it is limited to superficial structures for juvenile zebrafish and adult zebrafish due to tissue opacity and diffraction-based resolution loss associated with thicker tissues. In addition, 3D fluorescent microscopic images are commonly non-isotropic, making accurate rotational representations of true anatomy impossible. Image quality is degraded beyond embryonic life stages (especially beyond five days post fertilization for zebrafish) due to increasing pigmentation, thickness, and opacity. Since fluorescent microscopy is based on fluorescence, the provision of anatomical context by nonfluorescent cells is not possible. Optical projection tomography has been successfully used in semitransparent zebrafish larvae for high throughput morphological phenotyping at a voxel resolution of ~5 microns30, but does not allow most cells types to be distinguished or characterized histologically. Histological assessments require the diagnostician to visualize cellular features such as nuclear chromocenters and cytoplasmic vacuoles that are on the order of 2–3 microns in diameter, so lower resolution images are inadequate for accurate disease detection or diagnosis. In short, no existing methods fulfil the desired characteristics for complete anatomical phenotyping of whole organisms in 3D.

1.4. MicroCT as a potential tool for morphological phenotyping.

We have considered a range of 3D imaging modalities (including ultrasound, magnetic resonance imaging, and electron microscopy) for whole organism cell resolution morphological phenotyping33. Based on this survey and other considerations, we have proposed that synchrotron-based micro-computed tomography (microCT) may be well-suited to generate whole organism zebrafish images at all life stages and at cell resolution for accurate and systematic morphological phenotyping. Among other benefits, we reasoned that the high flux X-rays of synchrotron sources would produce images with high signal-to-noise ratio and without the problems of tube sources that have “spot” sizes that limit resolution. In the present study we use synchrotron microCT to image a prominent part of the nervous system of juvenile zebrafish, the visual system, to illustrate some of what can be seen in soft tissue using this imaging modality.

2. SAMPLES AND METHODS

2.1. Zebrafish samples.

Zebrafish specimens were anesthetized in 10x Finquel (MS-222 or tricaine) solution (Argent Chemical Laboratories (Redmond, WA) buffered in 1% Phosphate-buffered saline (PBS), and fixed in 10% Neutral Buffered Formalin (NBF) (Fisher Scientific, Allentown, PA). Samples were then stained with Uranium Acetate and Osmium Tetroxide (UaOs) (1% Ua and 1% Os) or 0.3% Phosphotungstic Acid (PTA) for 24 hours, infiltrated and embedded in Glycol Methacrylate (Polysciences, Inc., Warrington, PA) or EMBed-812 (Electron Microscopy Sciences, Hatfield, PA) in kapton tubing (Small Parts, Inc., Logansport, IN). Zebrafish were maintained under standard laboratory conditions and staged according to ZFIN13. All procedures on live animals were approved by the Institutional Animal Care and Use Committee (IACUC) at the Pennsylvania State University.

2.2. Synchrotron MicroCT Imaging was performed at the 2-BM beamline at the Advanced Photon Source (APS) of Argonne National Lab. The synchrotron radiations are an unfocused beam of 25 mm (horizontal) by 4 mm (vertical) with energy resolution (ΔE/E) of 1×10−2 , flux (photons/sec) of 1×1012 @17 keV, and energy range of 0.5–33 keV which can be tuned using a double multilayer monochromator. Diagrams and detailed description of synchrotron microCT imaging can be found elsewhere34,35.

2.3. Image Processing.

2D longitudinal projection images were directly obtained from the synchrotron microCT scanning system and used for reconstruction of the 3D volume. We have applied VGStudio Max 2.1 (Volume Graphics, Heidelberg, Germany) to the microCT data for 3D visualizations. It handles large data in raw, tiff and other formats.

3. RESULTS AND DISCUSSION

3.1. The visual system of juvenile zebrafish as a case study for microCT.

To enhance use of the zebrafish as a model for the systematic and complete study of morphological phenotypes, we have been working towards the ideal of soft tissue microCT imaging of sufficient resolution and of sufficient pancellular nature that would satisfy the pathologist. We have chosen the zebrafish as a model for developing this capability because all the cell types of a whole vertebrate organism fit within the width of the fields of view characteristic of synchrotron microCT. The juveniles used here were about 1 cm in length and less than 3 mm in width. In these early experiments, we focused on study of nerve tracts associated with vision.

The zebrafish is a frequently used model organism for vision science36, 37due to its transparency, the availability of transgenic tools for fluorescent tagging of its anatomical components38,39 and its genetic similarities to humans. While spectacular and highly informative fluorescent imaging is possible, existing 3D reconstructions of the nervous system are lacking in highly pigmented and opaque organismal and tissue specimens such as juvenile zebrafish. Therefore, the field would benefit from the ability to image nerve tracts in optically opaque samples such as juvenile zebrafish in which high-resolution fluorescent imaging is a challenge. Further benefits derive from the addition of context provided by cell types supporting and surrounding the nervous system, including interstitial and structural tissue components such as bone and vasculature. We have imaged metal-stained wild type juvenile zebrafish by synchrotron microCT, and rendered the images with the Maximum Intensity Projection (MIP)40 algorithm of the volume rendering program VGStudio Max (see Methods and Figure 1). Since Osmium tetroxide stains myelin in fixed tissue and absorbs X-rays, we have been able to resolve soft tissue elements of the visual system including optic nerves, optic chiasm, and retinotectal projections (i.e., nerve fibers connecting the retina with the optic nerve and the optic tectum of the midbrain). The identification of other myelinated components of the nervous system of juvenile zebrafish is being pursued (Xin and Cheng, unpublished). Note that the parts of the optical system of juvenile adults, including the optic radiations, had not yet been visualized in juvenile zebrafish in this detail, due to optical opacity and pigmented cells that interfere with the use of fluorescence methods to study neuronal circuits.

Figure 1, Anterior view of the optic nerves, optic chiasm, and retinotectal projections by synchrotron microCT imaging.

Figure 1,

Open in a new tab

(A) left lateral external view of a 48 dpf, wild type, fixed and UaOs stained juvenile zebrafish; (B) left lateral view of a slab of the same fish, with visualizations adjusted to emphasize the nervous system. Vertical lines demarcate the boundaries of the digital slab shown in panel C, which shows an anterior view of the optic chiasm (OC) in which the optic nerves decussate. Part of the left optic nerve is seen to cross in front of the left optic nerve. The optic tracts (OTs) extend from the optic chiasm to the optic radiations (ORs), ending in the primary visual cortex (PVC). L, (left) lens; R, (right) retina. These panels were generated using VGStudio Max 2.1, rendered with the MIP algorithm.

3.2. Volume rendering of nervous tissue.

The visualizations shown here involved transforming tomographic microCT volume reconstructions into 3D images that illustrate macroscopic anatomical structures. This study highlights some of our progress enhancing synchrotron microCT data with visualization techniques such as MIP (maximum intensity projection). The MIP algorithm projects the darkest voxels in each ray to the image surface allowing 3D structures with darker voxels to stand out from surrounding structures with lighter voxels. This visualization facilitates the evaluation of anatomic structures at multiple scales within optically opaque juvenile zebrafish. These images were extracted from a single tomographic reconstruction of a fixed and stained juvenile zebrafish.

3.3. The potential of soft tissue microCT.

The next step in this work will be to refine and extend of our visualization techniques to other organ systems such as the brain vascular network and cranial nerves, to be reported in more complete form elsewhere. The images in Figure 1 were made possible by: (i) the high resolution and field of view afforded by the flux, parallel-beam nature, and optics of the synchrotron microCT system at 2-BM, (ii) the affinity of soft tissue to metal stains, and (iii) digital visualization techniques developed through software applications. Other imaging modalities as presently used do not appear capable of capturing the full organism of this size at such high resolutions in a way that allows the study of all cell types. We noted in the study of these Ua/Os-stained samples that the strength of the signal in the retina is so strong as to make the general study of all cell types suboptimal. The use of alternative stains and the study of parts of the body at higher resolution will be presented elsewhere.

3.4. Computational demands.

The large file sizes involved (~100 GB per fish) require a high level of computational resources in terms of storage, processing power, and RAM (at least double the file size), as well as powerful but expensive 3D image processing software such as VGStudio Max. These computational demands mean that larger scale applications of high-resolution synchrotron microCT that require public access will be enhanced by shifting the bulk of image processing computation to high-performance computing resources on the server side, and piping the output to web-based applications as the primary means of data access for the public and larger scientific communities.

3.5. Other logistical issues.

Synchrotron resources are expensive to set up, and the creation of larger scale imaging capabilities for biological samples remains an unsolved problem. The small number of suitable beamlines, whose North American facilities include Argonne National Laboratory’s Advanced Photon Source near Chicago, IL and Brookhaven National Laboratory’s NSLS-II on Long Island, NY, means that access to synchrotron microCT resources is limited. Based on our work in this endeavour and the work of others in the microCT field, this lack of access is a barrier to any large-scale basic science or translational application to biological samples such as soft tissue microCT as shown here. Further exploration of the exciting potential benefit of soft tissue microCT will serve the interests of the scientific community and society at large, and require the creation of additional synchrotron based imaging instrumentation for biology, and be accelerated by the development of alternative high-flux and coherent x-ray sources.

4. CONCLUSIONS

Our microCT images of fixed and stained juvenile zebrafish reveal that soft tissue microCT can be used to image the vertebrate visual system despite the presence of optically opaque and pigmented tissues. This is in contrast to fluorescence microscopy that is typically limited to small (<400 micron thick), transparent samples that are nearly devoid of pigmentation. In contrast, synchrotron microCT provides high resolution (~1.43 micron isotropic voxel resolution) in optically opaque and pigmented samples over large fields of view (~3 mm). This potential suggests synchrotron microCT as a potential tool for systematic morphological phenotyping of small, whole organisms. Image-visualization techniques such as that shown here have the potential for automation to increase the throughput of phenotyping. The degree to which these image-processing techniques are extendable to other sample types, other resolutions, other imaging modalities, and other fields will be of substantial interest to determine.

ACKNOWLEDGEMENTS

We acknowledge support from NIH grants R24 RR017441, R24 OD011152, and 1R24 OD18559 to KCC, R01CA242956, R01AR052535 and RO1CA13468 to PLR, and funds from the Jake Gittlen Memorial Golf Tournament, The State of Pennsylvania Department of Health for a Tobacco CURE Cancer Award, and from the Pennsylvania State University Institute for Cyberscience seed grant in support of the Consortium for Interdisciplinary Image Informatics and Visualization (CI3V) to KCC. The Pennsylvania Department of Health specifically disclaims responsibility for any analyses, interpretations or conclusions. We acknowledge help from Tiffany Foster, Alex Lin, and Steve Peckins in the Cheng lab, F. De Carlo of the Advanced Photon Source at the Argonne National Laboratory in Illinois, and Jackie Webb (Rhode Island).

REFERENCES

  • [1].Lawson ND, Wolfe SA, “Forward and Reverse Genetic Approaches for the Analysis of Vertebrate Development in the Zebrafish,” Developmental Cell. 21, 48–64 (2011). [DOI] [PubMed] [Google Scholar]
  • [2].Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF, “Stages of embryonic development of the zebrafish.” Dev Dyn. 203(3), 253–310 (1995). [DOI] [PubMed] [Google Scholar]
  • [3].Driever W, Stemple DL, Schier AF and Solnica-Krezel L, “Zebrafish: genetic tools for studying vertebrate development,” TIG. 10, 152–159 (1994). [DOI] [PubMed] [Google Scholar]
  • [4].Muto A, Orger MB, Wehman AM, Smear MC, Kay JN, Page-McCaw PS, Gahtan E, Xiao T, Nevin LM, Gosse NJ, Staub W, Finger-Baier K, and Baier H, “Forward Genetic Analysis of Visual Behavior in Zebrafish,” PLoS Genet. 1(5), 66 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Ingham PW, “The power of the zebrafish for disease analysis,” Hum Mol Genet. 18(1), 107–112 (2009). [DOI] [PubMed] [Google Scholar]
  • [6].Lieschke GJ and Currie PD, “Animal models of human disease: zebrafish swim into view,” Nat. Rev. Genet 8, 353–367 (2007). [DOI] [PubMed] [Google Scholar]
  • [7].MacRae CA and Peterson RT, “Zebrafish as tools for drug discovery,” Nat. Rev. Drug Discov 14, 721–731 (2015). [DOI] [PubMed] [Google Scholar]
  • [8].Rubinstein AL, “Zebrafish: From disease modeling to drug discovery,” Curr Opin Drug Discov Devel, 6(2), 218–223 (2003). [PubMed] [Google Scholar]
  • [9].Iwanami N, “Zebrafish as a model for understanding the evolution of the vertebrate immune system and human primary immunodeficiency,” Exp Hematol. 42(8), 697–70 (2014). [DOI] [PubMed] [Google Scholar]
  • [10].Furutani-Seikia M and Wittbrodt J, “Medaka and zebrafish, an evolutionary twin study,” Mech Dev. 121, 629–637 (2004). [DOI] [PubMed] [Google Scholar]
  • [11].Dang M, Fogley R, Zon LI, “Identifying Novel Cancer Therapies Using Chemical Genetics and Zebrafish,” Adv Exp Med Biol. 916, 103–24 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [12].Feitsma H and Cuppen E, “Zebrafish as a Cancer Model,” Mol Cancer Res 6, 685–694 (2008). [DOI] [PubMed] [Google Scholar]
  • [13].Howe K, et al. , “The zebrafish reference genome sequence and its relationship to the human genome,” Nature 496(7446), 498–503 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Woods IG, Kelly PD, Chu F, Ngo-Hazelett P, Yan YL, Huang H, Postlethwait JH, Talbot WS, “A comparative map of the zebrafish genome,” Genome Res. 10(12), 1903–14 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Howe DG, Bradford YM, Conlin T, Eagle AE, Fashena D, Frazer K, Knight J, Mani P, Martin R, Moxon SA, Paddock H, Pich C, Ramachandran S, Ruef BJ, Ruzicka L, Schaper K, Shao X, Singer A, Sprunger B, Van Slyke CE, Westerfield M, “ZFIN, the Zebrafish Model Organism Database: increased support for mutants and transgenics,” Nucleic Acids Res, (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [16].Grunwald DJ and Eisen JS, “Headwaters of the zebrafish —emergence of a new model vertebrate,” Nat Rev Genet. 3(9), 717 (2002). [DOI] [PubMed] [Google Scholar]
  • [17].Hartwell LH, Hood L, Goldberg ML, Reynolds AE, Silver LM, Veres RC., “Genetics: from genes to genomes,” McGraw-Hill Higher Education, Boston: (2008). [Google Scholar]
  • [18].Patton EE, Zon LI “The art and design of genetic screens: zebrafish,” Nat Rev Genet. 2(12), 956–66 (2001). [DOI] [PubMed] [Google Scholar]
  • [19].Bill BR, Petzold AM, and Clark KJ, Schimmenti LA, and Ekker SC, “A Primer for Morpholino Use in Zebrafish,” Zebrafish 6(1), 69–77 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [20].Heasman J “Morpholino oligos: making sense of antisense?” Dev Biol. 243(2), 209–14 (2002). [DOI] [PubMed] [Google Scholar]
  • [21].Jontes JD, Emond MR, “Fluorescence imaging of transgenic zebrafish embryos,” Cold Spring Harb Protoc. 5, 604–608 (2012). [DOI] [PubMed] [Google Scholar]
  • [22].Crivat G and Taraska JW, “Imaging proteins inside cells with fluorescent tags,” Trends Biotechnol. 30(1), 8–16 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [23].Inwood S, “The man who knew too much: the strange and inventive life of Robert Hooke,” London: Pan, 1635–1703 (2003). [Google Scholar]
  • [24].Karling JS, “Schleiden’s Contribution to the Cell Theory,” Am. Nat 73(749), 517–37 (1939). [Google Scholar]
  • [25].Virchow R, “Cellular Pathology as Based upon Physiological and Pathological Histology,” University of Berlin, 1–566 (1858). [DOI] [PubMed] [Google Scholar]
  • [26].Cheng KC, Katz SR, Lin AY, Xin X, Ding Y “Whole-Organism Cellular Pathology: A Systems Approach to Phenomics,” Adv Genet. 95, 89–115 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].Cole ES, “Conventional light microscopy,” Curr. Protoc. Essential Lab. Tech 12(9), 1–29 (2016). [Google Scholar]
  • [28].Ignatius MS, Langenau DM, “Fluorescent imaging of cancer in zebrafish,” Methods Cell Biol. 105, 437–59 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Ritter K, Rising M, “2 Americans, 1 German win chemistry Nobel,” Associated Press, (2014). [Google Scholar]
  • [30].Pardo-Martin C, Allalou A, Medina J, Eimon PM., Wählby C and Yanik MF, “High-trhoughput hyperdimensional vertebrate phenotyping,” Nat. Commun 4, 1467 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [31].Keller PJ, “In vivo imaging of zebrafish embryogenesis,” Methods 62(3), 268–278 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [32].Ahrens MB, Orger MB, Robson DN, Li JM and Keller PJ, “Whole-brain functional imaging at cellular resolution using light-sheet microscopy,” Nature Methods 10, p.413–420 (2013). [DOI] [PubMed] [Google Scholar]
  • [33].Cheng KC, Xin X, “Whole-animal imaging, gene function, and the Zebrafish Phenome Project,” Curr Opin Genetics Dev. 21, 620–629 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [34].La Riviere PJ, Clark D, Rojek A, Vargas P, Xiao X, DeCarlo F, Kindlmann G, Cheng K, “Optimizing synchrotron micro-CT for high-throughput phenotyping of zebrafish,” Developments in X-Ray Tomography VII 7804, 1–8 (2010). [Google Scholar]
  • [35].De Carlo F, Xiao X, and Tieman B, “X-Ray Tomography System, Automation and Remote Access at Beamline 2-BM of the Advanced Photon Source,” Proc. of SPIE 6318, 1605–7422 (2006). [Google Scholar]
  • [36].Gestri G, Link BA, Neuhauss SC, “The Visual System of Zebrafish and its Use to Model Human Ocular Diseases,” Dev Neurobiol. 72 (3), 302–307 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Bilottaa J, Saszik S, “The zebrafish as a model visual system,” Int. J. Devl Neuroscience 19:7, p. 621–629, (2001). [DOI] [PubMed] [Google Scholar]
  • [38].Antinucci P and Hindges R, “A crystal-clear zebrafish for in vivo imaging,” Sci Rep. 6, 1–10 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [39].Antinucci P et al. , “Neural mechanisms generating orientation selectivity in the retina,” Curr Biol. 26(14), 1802–1815 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [40].Wallis JW, Miller TR, Lerner CA, Kleerup EC, “Three-dimensional display in nuclear medicine,” IEEE Trans Med Imaging 8(4), 297–303 (1989). [DOI] [PubMed] [Google Scholar]

RESOURCES