Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Aug 1.
Published in final edited form as: J Glaucoma. 2020 Aug;29(8):613–621. doi: 10.1097/IJG.0000000000001553

Improving the detection of glaucoma and its progression: a topographical approach

Donald C Hood 1,2, Zane Z Zemborain 1, Emmanouil Tsamis 1, C Gustavo De Moraes 2
PMCID: PMC7423747  NIHMSID: NIHMS1592768  PMID: 32459689

Abstract

Glaucoma is typically defined as a progressive optic neuropathy characterized by a specific (arcuate) pattern of visual field (VF) and anatomical changes. Therefore, we should be comparing arcuate patterns of damage seen on VFs with those seen on optical coherence tomography (OCT) maps. Instead, clinicians often use summary metrics such as VF pattern standard deviation (PSD), OCT retinal nerve fiber (RNF) global thickness, etc. There are two major impediments to topographically comparing patterns of damage on VF and OCT maps. First, until recently, it was not easy to make these comparisons with commercial reports. While recent reports do make it easier to compare VF and OCT maps, they have shortcomings. In particular, the 24–2 VF covers a larger retinal region than the commercial OCT scans, and, further, it is not easy to understand the topographical relationship among the different maps/plots within the current OCT reports. Here we show how a model of RNF bundles can overcome these problems. The second major impediment is the lack of a quantitative, and automated, method for comparing patterns of damage seen on VF and OCT maps. However, it is now possible to objectively and automatically quantify this agreement. Together, the RNF bundle model and the automated structure-function method should improve the power of topographical methods for detecting glaucoma and its progression. This should prove useful in clinical studies and trials, as well as for training and validating artificial intelligence/deep learning approaches for these purposes.

Keywords: glaucoma, optical coherence tomography, OCT, perimetry, visual fields, structure-function agreement

Précis:

To improve the power to detect glaucoma and its progression, a model that predicts arcuate patterns and an automated method for assessing structure-function agreement are proposed.

Glaucoma is typically defined as a progressive optic neuropathy characterized by a specific pattern of visual field and anatomical changes.1 Because glaucoma primarily acts in and/or at the disc, the “specific pattern” is arcuate in shape. Before the development of automated perimetry, clinicians typically looked for arcuate patterns of abnormal sensitivity on Goldmann visual fields (VF), and for a corresponding arcuate-shaped fiber bundle defects on fundus exams and photos. Today, the Goldmann VF has been replaced with static automated perimetry (SAP), usually with the 24–2/30–2 pattern, and the fundus exam/photograph augmented with optical coherence tomography (OCT). In this transition, the focus on arcuate patterns of abnormality was largely replaced with an emphasis on summary statistics, because VFs were inherently variable and early OCT instruments had relatively poor resolution.

In fact, many studies employing SAP and OCT do not even examine topographical patterns, but instead look at summary statistics (called metrics here). These metrics include the 24–2 VF pattern standard deviation (PSD), mean deviation (MD), and glaucoma hemifield test (GHT), and the OCT global, and quadrant, circumpapillary retinal nerve fiber layer (cRNFL) thickness. We have called the combination of 24–2 VF and OCT cRNFL tests, combined with a focus on VF and OCT metrics, a common clinical paradigm (CCP).2–4 The CCP misses and/or markedly underestimates glaucomatous damage, including early damage to the central/macular region of the retina,5–16 the critical region for everyday tasks17–19 and for quality of life.20–22

We have argued for a return to a focus on topographical patterns, and that these topographical comparisons should include 10–2 VF, as well as, 24–2 VF, and macular OCT scans, as well as the OCT disc scans.2,4 Failure to include 10–2 VF and/or retinal ganglion cell (RGC) measures will miss and/or underestimate macular damage. In any case, there are currently two major impediments to topographically comparing patterns of damage on VF and OCT maps. In particular, there is need for appropriate commercial reports and for objective and automated methods for assessing VF and OCT agreement.

THE NEED FOR COMMERCIAL REPORTS TO AID IN TOPOGRAPHICAL COMPARISONS

Until recently, the commercial OCT reports were suboptimal for making topographical comparisons between patterns of damage seen with VF and to those seen with OCT. To overcome their relatively poor resolution, early (time-domain) versions of OCT relied heavily on spatial averaging of neighboring regions around the disc to obtain move robust measurements of the RNFL. These summary metrics, and their probabilities relative to normative databases, were provided by the manufacturer in the form of pie charts, which have been used by to aid in clinical decisions. In 2014, we proposed the report in Fig. 1, which we call our “lab report”. This report was designed to more away from metrics and to allow for both a topographical comparison between VF and OCT, and a comparison between OCT scans of the macula and disc.

Figure 1.

Figure 1.

Open in a new tab

A one-page report based spectral domain OCT scans obtained with a commercial instrument (3D OCT-2000 Topcon, Inc.) and three scan protocols (a 6 × 6 mm cube scan centered on the disc; a 6 × 6 mm cube scan centered on the fovea; a 3.4 mm dia. circle scan centered on the disc. See ref 23 for details.) A. The image from the circle scan. B. The cRNFL thickness plot obtained from the thickness map in C (right panel) and shown with macular region in center (NSTIN view). C. RNFL thickness map from the fovea (left) and disc (right) cube scans. D. RGC plus inner plexiform layer thickness map from fovea cube scan. E. RNFL probability map in visual field (VF) view with 24–2 VF locations shown. F. RGC probability map in VF view with 10–2 VF locations shown. For E and F the color bar to the right shows the probability scale. For further details see refs 3, 23 and 24.

For our purposes here, there are four key features of our lab report in Fig. 1. First, the circumpapillary scan image and associated cRNFL thickness curve (A and B, black rectangle) are centered on the temporal quadrant of the disc (at 9 o’clock for the right eye). Thus, the region of the disc associated with the central portion of the VF is in the center of the cRNFL thickness curve. For example, on average the RGCs from the macular region of the retina, ±8° from fixation, send their axons to the region shown by the magenta and blue arrows in panel B. Second, there is an RGC probability map of the macula (F, green rectangle). The probability map, sometimes called a deviation map, is based upon a comparison to a healthy control group. The color codes the significance level, which ranges from p>10% (green) to p≤0.1% (dark red). Third, there is a RNFL probability map of both the macula and disc regions (E, red rectangle). Finally, to allow for ease of comparison of abnormal VF locations to abnormal OCT regions, these OCT probability maps are presented in field view, (i.e., the top of the maps in E and F is superior VF/inferior retina), and the 24–2 and 10–2 VF locations are superimposed, as circles in panels E and F, respectively. Our lab report is more fully explained in previous publications.3,23,24

Our lab report in Fig. 1 was designed to make it easier to compare different OCT maps/plots, as well as to compare the OCT and VF probability/deviation maps. For example, the arrows in Fig. 1 point to the abnormal regions associated with local arcuate defects in the superior retina/inferior VF (black arrows) and inferior retina/superior VF (red arrows). To topographically relate this defect to VF changes, the abnormal regions on the 24–2 and 10–2 VF enclosed by the red and black boundaries in Fig. 2A,B can be shown on the OCT maps, as seen in Fig. 2C,D.

Figure 2.

Figure 2.

Open in a new tab

A comparison of abnormal locations on the VF and abnormal regions on the OCT RNFL and RGC probability plots. The Total Deviation and Pattern Deviation plots of the 24–2 (A) and 10–2 (B) VFs with the boundaries of the abnormal locations shown as the red (superior VF) and black (inferior VF) lines. These same boundaries are superimposed on the RNFL (C) and RGC (D) probability maps from Fig. 1.

While our lab report was based on two cube scans, each 6×6 mm, most OCT manufactures now have a single, but wider, scan, instead, or in addition to, the two smaller cube scans. Further, two manufactures have recently incorporated the key aspects of our lab report into commercial reports, and a third has a similar report, but without the VF locations. Figures 3A and 3B are our in-lab versions of the two commercial reports based upon our lab report, and Fig. 4 the commercial report for the third manufacturer. See captions of Figs. 3 and 4 for details.

Figure 3.

Figure 3.

Open in a new tab

Our in-lab versions of two reports. A. A report based upon a 9 mm x 12 mm OCT cube scan (Topcon, Inc.). Because scans from different eyes are aligned to a common location of the center of the disc and adjusted for disc to fovea angle, the RNFL maps will not be the full size of the scan shown as the 9 ×12 mm black rectangle in upper right panel of panel A. B. A report based upon a circle scan and a 30° x 25° cube scan (Heidelberg, Inc.). C. The 24–2 VF deviation maps. The black rectangle encloses three locations associated with a nasal step. The same locations are enclosed within the black rectangles in panels A and B.

Figure 4.

Figure 4.

Open in a new tab

A. The commercial PanoMap report for the same eye as in Fig. 3 is based upon two 6×6 mm cube scans (Zeiss, Inc.), one centered on the fovea and one centered on the disc. The display within the black rectangle contains the two pseudo-color RNFL thickness maps, based upon the cube scans, centered upon the disc and fovea of the fundus image. The display within the red rectangle contains the deviation/probability map for the RNFL centered on the disc, and the deviation/probability map for the RGC centered on the fovea of the fundus image. B. The display within the red rectangle in A was rotated around the horizontal to present it in VF view (i.e., superior VF/inferior retina on top). In the right panel of B some of the VF locations are shown as the large (24–2) and small (10–2) black squares.

SHORTCOMINGS OF CURRENT TOPOGRPAPHICAL REPORTS

While these topographical reports have already proven useful,24,25 they have at least two shortcomings. First, the 24–2 VF covers a larger retinal region than the commercial OCT scans.

Even the reports with the wider scans in Figs. 3A and 3B do not include more than about 50% of the locations in the 24–2. Thus, when comparing 24–2 VF to OCT RNFL much of the 24–2 information is lost. Figure 3 supplies an example where the 24–2 VF (panel C) shows evidence consistent with arcuate damage. The abnormal points on the 24–2 pattern deviation plot are enclosed within the black rectangle in the OCT RNFL probability maps (upper right corner of panels A and B). Notice that none of the 3 abnormal 24–2 locations coincide with the abnormal (red or yellow) abnormal regions on the OCT RNFL probability maps in Fig. 3B, while only one location coincides in Fig. 3A. [Note: the RNFL maps are almost never as large as the 9×12mm scan image (thin black rectangle in Fig. 3A, upper right panel), because of the alignment of disc centers (see fig. caption). However, the 3 most nasal 24–2 points along the horizontal meridian always fall outside the RNFL map.]

A second shortcoming concerns understanding the topographical relationship among the different maps/plots within the current OCT reports. For example, consider Fig. 1. There are clear local defects seen on the circular b-scan (white and gray arrows) and cRNFL thickness curve (black and red arrows). Without a fair amount of experience with these reports, it is not easy to relate these locate defects on the cRNFL image (A) and thickness plot (B) to the abnormal regions on the RNFL and RGC maps (C-E). We have recently developed an approach to address these two shortcomings (Zane ZZ, et al., IOVS, 2020;XX;ARVO E-Abstract XXX, in press).

THE MODEL OF ARCUATE REGIONS

Our approach is based upon Jansonius et al.’s model26 that describes the path taken by RNFL bundles as they travel from the RGCs to the disc. We modified this model so as to incorporate the predicted pathways into the reports as shown in Figs. 5, which are the same as in Fig. 3 with predictions based upon the Jansonius model added. [For these examples, we adjusted the region along the x-axis of the cRNFL plot (shaded rectangles) so that the defects on the probability maps were included within the model’s predicted arcuates.]

Figure 5.

Figure 5.

Open in a new tab

Same as Fig 3 with the addition of arcuate regions. The arcuate regions are set by the model based upon the width of the blue rectangle on the cRNFL thickness plot. See text for details.

ADVANTAGES OF THIS TOPOGRAPHICAL APPROACH

There are a number of advantages to adding this information to the OCT reports, including addressing the two limitations mentioned above. For example, it is now easier to see the relationship between the regions of the cRNFL thickness plots (transparent blue rectangles in Fig. 5A,B, upper left) and the associated arcuate regions on the RNFL probability maps (arcuate curves in other panels). Further, the model’s predicted region allows us to incorporate more of the information from the 24–2 VF [black rectangle in Fig. 5B]. In particular, more of the 24–2 VF’s abnormal locations now fall within the predicted abnormal arcuate region. In fact, all three abnormal 24–2 VF points now fall within the predicted region on the RNFL probability maps in Fig. 5 (upper right panel in panels A and B).

While the example in Fig. 5 illustrates an eye with clear, widespread, damage on the OCT report, the defect in Fig. 6 is relatively local and relatively subtle. Notice that on the Jan. 2016 report (Fig. 6A), there is a suggestion of very subtle damage near the red arrows in the superior retina when the model’s predicted arcuate regions are superimposed. In particular, there is topographical agreement between the subtle dip in the cpRNFL curve (red arrow within blue rectangle) and the subtle changes in the RGC and RNFL probability maps, near the red arrows. The report (Fig. 6B) based upon a repeat scan about 23 months later indicated that the defect suggested by the report in Fig. 6A is real, and has progressed. Note that this is an example of so-called “green disease”, i.e., OCT pie chart metrics are normal (black arrow) in presence of a clear defect. It is also “green progression”, as the pie charts were green on the both test days, yet the defect had clearly progressed. Further, the metrics (GHT, PSD, MD) of the 24–2 VF test obtained 4.5 months after the second OCT test day were within normal limits, as were the metrics of a 10–2 test obtained between the first and second OCT tests (Fig. 6C). However, there were 4 abnormal locations on the pattern deviation plot of this 10–2 VF, and they fell within the abnormal region on the RGC probability maps, as shown by the black circles in Fig. 6B (lower right panel).

Figure 6.

Figure 6.

Open in a new tab

A,B. Same report as in Fig 5B for an eye with local damage tested on 1/14/2016 (A) and 12/4/2017 (B). C. 10–2 VF deviation maps for this eye.

AUTOMATED STRUCTURE-FUNCTION COMPARISON

Until recently, there was an additional impediment to the acceptance of a topographical approach for comparing VFs and OCT, especially for clinical use and clinical research. Namely, we did not have an objective and automated method for comparing abnormal regions on the OCT with abnormal locations on VF probability/deviation maps. For example, in Fig. 6B, the fact that 4 abnormal 10–2 VF locations coincide with the abnormal region of the RGC+ probability map suggests topographical agreement. However, is this a statistically valid criterion to distinguish glaucoma patients from healthy controls? To address the need for a quantitative and objective approach, we recently developed, and tested, an automated method for comparing abnormal locations on VFs to abnormal regions seen on OCT maps.27,28 The method involves superimposing the abnormal VF points (≤5%) on the OCT probability maps as shown in Fig. 7 and counting the locations that are abnormal on both. (Criteria for abnormal VF and OCT locations, as well as number of points needed can be varied to adjust specificity.28) We have demonstrated that the overwhelming majority (>90%) of eyes, even those with early glaucoma, show excellent structural (OCT) and functional (VF) agreement as long as both 24–2 and 10–2 VFs and both RNFL and RGC probability maps are employed.27 Further, our topographical approach was superior to criteria based upon metrics.27

Figure 7.

Figure 7.

Open in a new tab

An illustration of a method28 for automatically and objectively comparing abnormal locations on VFs with abnormal regions on OCT probability maps. The abnormal locations of the 24–2 and 10–2 VFs in panel A are superimposed upon the RGC and RNFL probability maps by the program and displayed as in B. The region of overlap between the two tests is shown enlarged in C so that the VF locations abnormal on both VF and OCT can be better visualized. These locations are indicated by the open diamonds (24–2) and open squares (10–2). Reproduced from Fig. 2 in ref. 27.

CONCLUSIONS

In conclusion, we should return to an emphasis on arcuate patterns of damage seen on VFs and anatomical (e.g., OCT) tests, and away from using summary metrics such as PSD, etc. In fact, contrary to popular opinion,27,29 there is good agreement between VF and anatomical changes, even in the earliest signs of glaucoma, if topographical agreement is examined between 24–2 and 10–2 deviation maps, on one hand, and RGC and RNFL probability maps, on the other.27 Further, it is possible to objectively and automatically quantify this agreement. Finally, anatomical models of arcuate regions of damage should improve the power of topographical methods for detecting glaucoma and its progression, as well as for training and validating artificial intelligence/deep learning approaches for these purposes.

Acknowledgments

Supported by: NIH/ NEI grant: EY002115 and EY025253.

Footnotes

Financial disclosures: DCH receives lecture fees, research support and equipment from Topcon, Inc. and Heidelberg Engineering. ET received lecture fees from Topcon, Inc.; CGD receives research support from Topcon, Inc. and Heidelberg Engineering and is consultant for Carl Zeiss Meditec.

REFERENCES

  • 1.Weinreb RN, Garway-Heath DF, Leung CKS, Liebmann JM, Medeiros FA. Consensus 10: Diagnosis of primary open angle glaucoma In: World Glaucoma Association. Seattle, Washington, USA: Kugler Publications; 2016. http://www.worldglaucoma.org/consensus-10-preliminary-report/. Accessed January 9, 2016. [Google Scholar]
  • 2.Hood DC, De Moraes CG. Challenges to the Common Clinical Paradigm for Diagnosis of Glaucomatous Damage With OCT and Visual Fields. Investigative Ophthalmology & Visual Science. 2018;59(2):788–791. doi: 10.1167/iovs.17-23713 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Hood DC. Improving our understanding, and detection, of glaucomatous damage: An approach based upon optical coherence tomography (OCT). Prog Retin Eye Res. 2017; 57:46–75. doi: 10.1016/j.preteyeres.2016.12.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Hood DC, De Moraes CG. Four Questions for Every Clinician Diagnosing and Monitoring Glaucoma. J Glaucoma. 2018;27(8):657–664. doi: 10.1097/IJG.0000000000001010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Langerhorst C, Carenini L, Bakker D, De Bie-Raakman M. Measurements for description of very early glaucomatous field defects. In: Wall M, Heiji A, eds. XIIth International Perimetric Society Meeting Wurzburg, Germany: Kugler Publications; 1996:67–73. [Google Scholar]
  • 6.Schiefer U, Papageorgiou E, Sample PA, et al. Spatial Pattern of Glaucomatous Visual Field Loss Obtained with Regionally Condensed Stimulus Arrangements. Invest Ophthalmol Vis Sci. 2010;51(11):5685–5689. doi: 10.1167/iovs.09-5067 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Hood DC, Raza AS, de Moraes CGV, et al. Initial arcuate defects within the central 10 degrees in glaucoma. Invest Ophthalmol Vis Sci. 2011;52(2):940–946. doi: 10.1167/iovs.10-5803 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Hood DC, Raza AS, de Moraes CGV, Liebmann JM, Ritch R. Glaucomatous damage of the macula. Prog Retin Eye Res. 2013; 32:1–21. doi: 10.1016/j.preteyeres.2012.08.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Traynis I, De Moraes CG, Raza AS, Liebmann JM, Ritch R, Hood DC. The Prevalence and Nature of Early Glaucomatous Defects in the Central 10° of the Visual Field. JAMA Ophthalmol. 2014;132(3):291–297. doi: 10.1001/jamaophthalmol.2013.7656 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Wang DL, Raza AS, de Moraes CG, et al. Central Glaucomatous Damage of the Macula Can Be Overlooked by Conventional OCT Retinal Nerve Fiber Layer Thickness Analyses. Transl Vis Sci Technol. 2015;4(6). doi: 10.1167/tvst.4.6.4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Grillo LM, Wang DL, Ramachandran R, et al. The 24–2 Visual Field Test Misses Central Macular Damage Confirmed by the 10–2 Visual Field Test and Optical Coherence Tomography. Transl Vis Sci Technol. 2016;5(2):15. doi: 10.1167/tvst.5.2.15 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Park H-YL, Hwang B-E, Shin H-Y, Park CK. Clinical Clues to Predict the Presence of Parafoveal Scotoma on Humphrey 10–2 Visual Field Using a Humphrey 24–2 Visual Field. Am J Ophthalmol. 2016;161:150–159. doi: 10.1016/j.ajo.2015.10.007 [DOI] [PubMed] [Google Scholar]
  • 13.Sullivan-Mee M, Karin Tran MT, Pensyl D, Tsan G, Katiyar S. Prevalence, Features, and Severity of Glaucomatous Visual Field Loss Measured With the 10–2 Achromatic Threshold Visual Field Test. Am J Ophthalmol. 2016; 168:40–51. doi: 10.1016/j.ajo.2016.05.003 [DOI] [PubMed] [Google Scholar]
  • 14.Jung KI, Park CK. Detection of Functional Change in Preperimetric and Perimetric Glaucoma Using 10–2 Matrix Perimetry. Am J Ophthalmol. 2017;182:35–44. doi: 10.1016/j.ajo.2017.07.007 [DOI] [PubMed] [Google Scholar]
  • 15.De Moraes CG, Hood DC, Thenappan A, et al. 24–2 Visual Fields Miss Central Defects Shown on 10–2 Tests in Glaucoma Suspects, Ocular Hypertensives, and Early Glaucoma. Ophthalmology. 2017;124(10):1449–1456. doi: 10.1016/j.ophtha.2017.04.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Roberti G, Manni G, Riva I, et al. Detection of central visual field defects in early glaucomatous eyes: Comparison of Humphrey and Octopus perimetry. PLoS ONE. 2017;12(10):e0186793. doi: 10.1371/journal.pone.0186793 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Nelson P, Aspinall P, O’Brien C. Patients’ perception of visual impairment in glaucoma: a pilot study. Br J Ophthalmol. 1999;83(5):546–552. doi: 10.1136/bjo.83.5.546 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Ramulu PY, West SK, Munoz B, Jampel HD, Friedman DS. Driving cessation and driving limitation in glaucoma: the Salisbury Eye Evaluation Project. Ophthalmology. 2009;116(10):1846–1853. doi: 10.1016/j.ophtha.2009.03.033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ramulu PY, West SK, Munoz B, Jampel HD, Friedman DS. Glaucoma and Reading Speed: The Salisbury Eye Evaluation Project. Arch Ophthalmol. 2009;127(1):82–87. doi: 10.1001/archophthalmol.2008.523 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Prager AJ, Hood DC, Liebmann JM, et al. Association of Glaucoma-Related, Optical Coherence Tomography-Measured Macular Damage With Vision-Related Quality of Life. JAMA Ophthalmol. 2017;135(7):783–788. doi: 10.1001/jamaophthalmol.2017.1659 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Blumberg DM, De Moraes CG, Prager AJ, et al. Association Between Undetected 10–2 Visual Field Damage and Vision-Related Quality of Life in Patients With Glaucoma. JAMA Ophthalmol. 2017;135(7):742–747. doi: 10.1001/jamaophthalmol.2017.1396 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Blumberg DM, Liebmann JM, Hirji SH, Hood DC. Diffuse Macular Damage in Mild to Moderate Glaucoma Is Associated With Decreased Visual Function Scores Under Low Luminance Conditions. Am J Ophthalmol. 2019;208:415–420. doi: 10.1016/j.ajo.2019.08.024 [DOI] [PubMed] [Google Scholar]
  • 23.Hood DC, Raza AS. On improving the use of OCT imaging for detecting glaucomatous damage. Br J Ophthalmol. 2014;98 Suppl 2:i0i1–9. doi: 10.1136/bjophthalmol-2014-305156 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Hood DC, Cuir ND, Blumberg DM, et al. A Single Wide-Field OCT Protocol Can Provide Compelling Information for the Diagnosis of Early Glaucoma. Trans Vis Sci Tech. 2016;5(6):4–4. doi: 10.1167/tvst.5.6.4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hood DC, Raza AS, Moraes CGD, et al. Evaluation of a One-Page Report to Aid in Detecting Glaucomatous Damage. Trans Vis Sci Tech. 2014;3(6):8–8. doi: 10.1167/tvst.3.6.8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Jansonius NM, Nevalainen J, Selig B, et al. A mathematical description of nerve fiber bundle trajectories and their variability in the human retina. Vision Res. 2009;49(17):2157–2163. doi: 10.1016/j.visres.2009.04.029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hood DC, Tsamis E, Bommakanti NK, et al. Structure-Function Agreement Is Better Than Commonly Thought in Eyes With Early Glaucoma. Invest Ophthalmol Vis Sci. 2019;60(13):4241–4248. doi: 10.1167/iovs.19-27920 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Tsamis E, Bommakanti NK, Sun A, Thakoor KA, Moraes CGD, Hood DC. An Automated Method for Assessing Topographical Structure–Function Agreement in Abnormal Glaucomatous Regions. Trans Vis Sci Tech. 2020;9(4):14–14. doi: 10.1167/tvst.9.4.14 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Hood DC. Does Retinal Ganglion Cell Loss Precede Visual Field Loss in Glaucoma? J Glaucoma. 2019;28(11):945–951. doi: 10.1097/IJG.0000000000001380 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES