Search

Purpose To design a robust and automated estimation method for measuring the retinal nerve fiber layer (RNFL) thickness using spectral domain optical coherence tomography (SD-OCT). Methods We developed a deep learning–based image segmentation network for automated segmentation of the RNFL in SD-OCT B-scans of mouse eyes. In total, 5500 SD-OCT B-scans (5200 B-scans were used as training data with the remaining 300 B-scans used as testing data) were used to develop this segmentation network. Postprocessing operations were then applied on the segmentation results to fill any discontinuities or remove any speckles in the RNFL. Subsequently, a three-dimensional retina thickness map was generated by z-stacking 100 segmentation processed thickness B-scan images together. Finally, the average absolute difference between algorithm predicted RNFL thickness compared to the ground truth manual human segmentation was calculated. Results The proposed method achieves an average dice similarity coefficient of 0.929 in the SD-OCT segmentation task and an average absolute difference of 0.0009 mm in thickness estimation task on the basis of the testing dataset. We also evaluated our segmentation algorithm on another biological dataset with SD-OCT volumes for RNFL thickness after the optic nerve crush injury. Results were shown to be comparable between the predicted and manually measured retina thickness values. Conclusions Experimental results demonstrate that our automated segmentation algorithm reliably predicts the RNFL thickness in SD-OCT volumes of mouse eyes compared to laborious and more subjective manual SD-OCT RNFL segmentation. Translational Relevance Automated segmentation using a deep learning–based algorithm for murine eye OCT effectively and rapidly produced nerve fiber layer thicknesses comparable to manual segmentation.

Purpose To create a mouse traumatic optic neuropathy (TON) model that is reproducible, reliable, and easy to manipulate with high specificity to retinal ganglion cell (RGC) layer and no mortality. The model will be useful for understanding the pathophysiology of retinal ganglion cell death and for testing neuroprotective therapeutics. Methods An Nd:YAG laser was used to generate focal photodisruptive retinal damage. Noninvasive in vivo ophthalmologic imaging technologies such as optical coherence tomography (OCT) and confocal laser scanning ophthalmoscopy (CSLO) were used to longitudinally track the retinal nerve fiber layer (RNFL) thickness and RGC number change, respectively. Immunostaining and pattern electroretinography (PERG) were also used to evaluate structure and functional change after laser injury. Results Our ND:YAG laser generates a concussive photodisruptive laser shockwave force which induces focal RGC death in the targeted area. We observed a correlative decrease in RGCs number, RNFL, and PERG function of RGC in the laser zone. The pattern of RNFL thinning and RGC soma loss correlates with the pattern and amount of fluorescence loss on OCT and CSLO images, respectively. The ND:YAG laser does not cause any damage to other layers in the retina nor any side effects including changes in intraocular pressure, corneal edema, and calcification or mortality (which has been observed in other TON models). Conclusions We have created a new and novel RGC TON death model that confers no mortality and produces a quantifiable decrease in RGC number and function. The laser targeted regions of the retina correlate with both in vivo imaging by OCT and CSLO and histologically with regions of RGC loss without ophthalmic side effects. Translational relevance This laser-based TON injury model is simple to implement, is reproducible, and is useful for determining the molecular and cellular pathophysiology of TON and RGC death and for testing neuroprotective therapeutics.

Author(s): Moussa, Kareem; Feinstein, Max; Pekmezci, Melike; Lee, Jun Hui; Bloomer, Michele; Oldenburg, Catherine; Sun, Zhimin; Lee, Richard K; Ying, Gui-Shuang; Han, Ying | Abstract: PurposeTo compare the macroscopic and microscopic histologic changes in eyes treated with micropulse transscleral cyclophotocoagulation (MP-TCP) versus continuous wave transscleral cyclophotocoagulation (CW-TCP).MethodsTwelve halves of globes from three pairs of adult cadaveric eyes were randomly assigned to nontreated control, CW-TCP, single MP-TCP treatment, or double MP-TCP treatments, and then sectioned for histologic analysis. Presence or absence of the following four unique histologic changes was recorded: splitting within the ciliary process epithelium (splitting), separation of the pigmented ciliary process epithelium from the stroma (separation), coagulation of collagen and destruction of ciliary process stroma (coagulation), and full-thickness destruction of ciliary process epithelium (destruction).ResultsA total of 498 slides were analyzed, and laser scars in all treated specimens were located in the pars plana. Logistic regression analysis showed that compared with controls, CW-TCP-treated specimens were significantly more likely to experience separation (odds ratio [OR] = 11.1, P = 0.02), coagulation (OR = 24.3, P = 0.002), and destruction (OR = 11.1, P = 0.03). Destruction of the ciliary process epithelium was observed exclusively in CW-TCP-treated sections. No significant differences in histologic features were observed between controls and MP-TCP.ConclusionsMP-TCP does not produce significant histologic changes in cadaveric eyes, whereas CW-TCP treatment does.Translational relevanceThese findings improve understanding of the mechanism of MP-TCP, help explain the increased rates of adverse effects following CW-TCP treatment compared with MP-TCP, and describe effects of MP-TCP at various doses.

PURPOSE: To identify genetic risk factors contributing to central corneal thickness (CCT) in individuals from South India, a population with a high prevalence of ocular disorders. METHODS: One hundred ninety-five individuals from 15 large South Indian pedigrees were genotyped using the Omni2.5 bead array. Family-based association for CCT was conducted using the score test in MERLIN. RESULTS: Genome-wide association study (GWAS) identified strongest association for single nucleotide polymorphisms (SNPs) in the first intron of WNT7B and CCT (top SNP rs9330813; β = −0.57, 95% confidence interval CI: −0.78 to −0.36; P = 1.7 × 10−7). We further investigated rs9330813 in a Latino cohort and four independent European cohorts. A meta-analysis of these data sets demonstrated statistically significant association between rs9330813 and CCT (β = −3.94, 95% CI: −5.23 to −2.66; P = 1.7 × 10−9). WNT7B SNPs located in the same genomic region that includes rs9330813 have previously been associated with CCT in Latinos but with other ocular quantitative traits related to myopia (corneal curvature and axial length) in a Japanese population (rs10453441 and rs200329677). To evaluate the specificity of the observed WNT7B association with CCT in the South Indian families, we completed an ocular phenome-wide association study (PheWAS) for the top WNT7B SNPs using 45 ocular traits measured in these same families including corneal curvature and axial length. The ocular PheWAS results indicate that in the South Indian families WNT7B SNPs are primarily associated with CCT. CONCLUSIONS: The results indicate robust evidence for association between WNT7B SNPs and CCT in South Indian pedigrees, and suggest that WNT7B SNPs can have population-specific effects on ocular quantitative traits.

Purpose To estimate visual field (VF) sensitivity at which retinal nerve fiber layer (RNFL) thinning reaches the measurement floor and at which RNFL stops thinning (change points), the dynamic range of RNFL thickness, and the number of steps from normal to RNFL floor among three optical coherence tomography (OCT) devices. Methods Glaucomatous patients (n = 58) and healthy subjects (n = 55-60) prospectively underwent VF testing and RNFL thickness measurement with Cirrus, Spectralis, and RTVue. Change points and corresponding RNFL thicknesses were estimated with simple linear regression (SLR) and Bayesian change point (BCP) analyses. The dynamic range and number of steps to RNFL floor were determined. Results The average VF change points and corresponding residual thickness at the time RNFL stopped thinning were -22.2 dB and 57.0 μm (Cirrus), -25.3 dB and 49.2 μm (Spectralis), and -24.6 dB and 64.7 μm (RTVue). The RNFL dynamic ranges derived from SLR values were wider on Spectralis (52.6 μm) than on Cirrus (35.4 μm) and RTVue (35.5 μm); the corresponding number of steps to reach the RNFL floor were 9.0 on Cirrus, 10.6 on Spectralis, and 8.3 on RTVue. Conclusions The relative VF sensitivity at which average RNFL thickness reaches the measurement floor, the residual layer thickness, and RNFL dynamic measurement range differ among the three devices. However, the number of steps from normal to the RNFL thickness floor is comparable.

Purpose We investigated the progressive nature of neurodegenerative structural changes following injury to retinal ganglion cell (RGC) axons using quantifiable and noninvasive in vivo imaging techniques. Methods To track degenerative RGC progression in retinas following optic nerve crush (ONC) injury, spectral-domain optical coherence tomography (SD-OCT) was used to quantitate the RGC nerve fiber layer (NFL) density. The RGC soma cell density (RCD) was measured by confocal scanning laser ophthalmoscopy (CSLO). The RCD counts were performed using blood vessels as landmarks to anatomically track defined progressive changes in enhanced yellow fluorescent fusion protein (EYFP)-labeled RGCs. Results Following ONC injury, 68% of the observed decrease in RCD measured by CSLO and 54% of the NFL thickness obtained by SD-OCT imaging (N=4 retinas) occurred within the first week. Between days 7 and 14, an additional 22% decrease in RCD was concurrent with a 31% decrease in overall NFL thickness. Finally, between days 14 and 21, an additional 10% decrease in RCD measured in vivo by CSLO and 15% decrease in NFL thickness by SD-OCT was observed. Conclusions Our data suggest that in vivo CSLO imaging of EYFP-RGC expression and SD-OCT measured NFL thickness are fast and reliable methods that longitudinally track neurodegenerative progression following ONC injury. Neurodegenerative changes in NFL thickness measured by SD-OCT imaging have the same overall trajectory as those observed by CSLO for RCD; however, changes in NFL thickness initially lag behind in vivo RGC soma counts with a slower decline in overall measurable change.

Purpose We determined the profiles of sphingomyelin, sphingoid base, sphingoid base-1-phosphate, and ceramide, and their quantitative differences between control and glaucomatous trabecular meshwork (TM) derived from human donors. Methods Control and primary open angle glaucoma (POAG) TM samples were collected from cadaver donors. In addition, POAG TM surgical specimens also were procured. Lipid extraction was performed using suitable modifications of the Bligh and Dyer method. Protein concentrations were determined using Bradford's method. Lipids, identified using standardized class-specific lipid mass spectrometry, were quantified using a two-step ratiometric process. Gomori methenamine silver (GMS) staining was performed for detection of presence of Fusarium in the anterior eye tissue sections. PCR analyses were performed for detection of Fusarium species in the donor TM samples. Results Several species of sphingomyelin, sphingoid base, sphingoid base-1-phosphate, and ceramide were common between control and POAG TM. Only a subset of species in some of these classes were identified uniquely either in controls or in POAG TM. Several lipid species of fungal origin (many from Fungi imperfecti, Fusarium species) were found to be common between control and POAG TM. The GMS staining and PCR analyses showed presence of DNA belonging to Fusarium species suggesting latent commensalism. Conclusions Most sphingolipids and ceramides (except a few unique to a specific donor TM group) were found to be common in the control and POAG TM. Latent commensalism by Fusarium was suggested by identification of Fusarium-specific lipids, which was supported further by PCR amplification and sequencing of DNA.

Purpose We compared phospholipid (phosphatidylcholine, phosphatidylserine, phosphatidylethanolamine, and phosphatidylinositol) profiles of control and glaucomatous trabecular meshwork (TM) derived from human donors. Methods Control TM and most primary open angle glaucoma (POAG) TM were collected from cadaver donors. A select subset of POAG surgical TM samples also were collected for analyses. Lipid extraction was performed using a modification of the Bligh and Dyer method, protein concentrations were determined using the Bradford method, and for select samples confirmed with densitometry of PHAST gels. Lipids were identified and subjected to ratiometric quantification using a TSQ quantum Access Max triple quadrupole mass spectrometer with precursor ion scan (PIS) or neutral ion loss scan (NLS), using appropriate class specific lipid standards. Results The comparative profiles of phosphatidylcholine, phosphatidylserine, phosphoethanolamine, and phosphatidylinositol between control and glaucomatous TM showed several species common between them. A number of unique lipids in all four phospholipid classes also were identified in control TM that were absent in glaucoma TM and vice versa. Conclusions A number of phospholipids were found to be uniquely present in control but absent in glaucomatous TM and vice versa. Compared to a previous study of control and POAG blood, a number of these phospholipids are absent locally (TM), as well as systemically (in blood).

PURPOSE. To investigate the effects of central corneal thickness (CCT)-associated variants on primary open-angle glaucoma (POAG) risk using single nucleotide polymorphisms (SNP) data from the Glaucoma Genes and Environment (GLAUGEN) and National Eye Institute (NEI) Glaucoma Human Genetics Collaboration (NEIGHBOR) consortia. METHODS. A replication analysis of previously reported CCT SNPs was performed in a CCT dataset (n ¼ 1117) and these SNPs were then tested for association with POAG using a larger POAG dataset (n ¼6470). Then a CCT genome-wide association study (GWAS) was performed. Top SNPs from this analysis were selected and tested for association with POAG. cDNA libraries from fetal and adult brain and ocular tissue samples were generated and used for candidate gene expression analysis. RESULTS. Association with one of 20 previously published CCT SNPs was replicated: rs12447690, near the ZNF469 gene (P ¼ 0.001; b ¼� 5.08 lm/allele). None of these SNPs were significantly associated with POAG. In the CCT GWAS, no SNPs reached genome-wide significance. After testing 50 candidate SNPs for association with POAG, one SNP was identified, rs7481514 within the neurotrimin (NTM) gene, that was significantly associated with POAG in a low-tension subset (P ¼ 0.00099; Odds Ratio [OR] ¼ 1.28). Additionally, SNPs in the CNTNAP4 gene showed suggestive association with POAG (top SNP ¼ rs1428758; P ¼ 0.018; OR ¼ 0.84). NTM and CNTNAP4 were shown to be expressed in ocular tissues. CONCLUSIONS. The results suggest previously reported CCT loci are not significantly associated with POAG susceptibility. By performing a quantitative analysis of CCT and a subsequent analysis of POAG, SNPs in two cell adhesion molecules, NTM and CNTNAP4, were identified and may increase POAG susceptibility in a subset of cases. (Invest Ophthalmol Vis

Mice with a targeted disruption of Brn3b (knockout Brn3b(-/-)) undergo the loss of a majority of retinal ganglion cells (RGCs) before birth. Spectral domain optical coherence tomography (SD-OCT) allows for the noninvasive examination of Brn3b(-/-) cellular loss in vivo.The central retinas of Brn3b(-/-) and phenotypically wild-type (Brn3b(+/+) and Brn3b(±)) mice were imaged by SD-OCT. The combined nerve fiber layer (NFL) and inner plexiform layer (IPL) were manually segmented and thickness maps were generated. The results were confirmed by histologic and immunofluorescence cell counts of the RGC layer (RGCL) of the same retinas.The combined NFL and IPL of the Brn3b(-/-) retinas were significantly thinner, and the histologic cell counts significantly lower, than those of the phenotypically wild-type retinas (paired t-test; P0.01 and P0.01, respectively). The combined NFL and IPL thickness and the histologic cell count correlated highly (R(2) = 0.9612). Immunofluorescence staining revealed significant RGC-specific loss in Brn3b(-/-) retinas (paired t-test; P0.01). The distribution of combined central NFL and IPL loss was not localized or sectorial.The strong correlation between the combined layer thickness and histologic cell counts validates manual OCT segmentation as a method of monitoring cell loss in the RGCL. A retinal thickness map assessed if combined NFL and IPL thickness loss in Brn3b(-/-) eyes was topographically specific. Generalized RGC and combined NFL and IPL loss was observed in the Brn3b(-/-) retinas, in contrast to topographically specific RGC loss observed in glaucomatous DBA2/J eyes.