Ophthalmic Surgery, Lasers and Imaging Retina

Clinical Science 

Posterior Segment Distortion in Ultra-Widefield Imaging Compared to Conventional Modalities

Luke Nicholson, FRCOphth; Li Yen Goh, BMBS; Elena Marshall, MD; Clara Vazquez-Alfageme, MD; Irini Chatziralli, MD, PhD; Monica Clemo; Philip G. Hykin, FRCOphth; Sobha Sivaprasad, DM, FRCOphth

Abstract

BACKGROUND AND OBJECTIVES:

To describe posterior segment distortions in Optos ultra-widefield images (Optos 200TX; Optos, Dunfermline, United Kingdom) compared to Topcon retinal camera images (Topcon, Tokyo, Japan) using optic disc dimensions and exploring a proposed method for correcting these distortions.

PATIENTS AND METHODS:

Comparative image analysis study on 20 eyes with color fundus images from Optos and Topcon. A model eye with vertical and horizontal grids imaged with Optos in the conventional position and rotated 90° was analyzed.

RESULTS:

Mean vertical-to-horizontal disc diameter ratios were 0.956 [95% CI, 0.919–0.993] for Optos and 1.083 [95% CI, 1.045–1.121] for Topcon (P < .001). This was 0.910 in the conventional position and 1.072 with the object rotated 90° for the model eye with Optos and 1.008 and 0.999, respectively, using Topcon. The average of the measurements taken using both images from Optos yielded a ratio of 0.987.

CONCLUSION:

Optos incorporates a consistent horizontal stretch to images. Combining images taken at right angles reduces the distortion.

[Ophthalmic Surg Lasers Imaging Retina. 2016;47:644–651.]

Abstract

BACKGROUND AND OBJECTIVES:

To describe posterior segment distortions in Optos ultra-widefield images (Optos 200TX; Optos, Dunfermline, United Kingdom) compared to Topcon retinal camera images (Topcon, Tokyo, Japan) using optic disc dimensions and exploring a proposed method for correcting these distortions.

PATIENTS AND METHODS:

Comparative image analysis study on 20 eyes with color fundus images from Optos and Topcon. A model eye with vertical and horizontal grids imaged with Optos in the conventional position and rotated 90° was analyzed.

RESULTS:

Mean vertical-to-horizontal disc diameter ratios were 0.956 [95% CI, 0.919–0.993] for Optos and 1.083 [95% CI, 1.045–1.121] for Topcon (P < .001). This was 0.910 in the conventional position and 1.072 with the object rotated 90° for the model eye with Optos and 1.008 and 0.999, respectively, using Topcon. The average of the measurements taken using both images from Optos yielded a ratio of 0.987.

CONCLUSION:

Optos incorporates a consistent horizontal stretch to images. Combining images taken at right angles reduces the distortion.

[Ophthalmic Surg Lasers Imaging Retina. 2016;47:644–651.]

Introduction

In the past decade, ultra-widefield imaging has contributed significantly to our management of retinal diseases. The need to view the peripheral retina has led to the progression of imaging techniques from 20° posterior pole imaging to 50° imaging and, subsequently, the use of a montage of seven 30° to 50° field images in the Early Treatment Diabetic Retinopathy Study.1

The use of ultra-widefield imaging is increasingly being used in the management of numerous conditions including uveitis, diabetic retinopathy, retinal vein occlusion, choroidal masses, and retinal detachment.2–7 Standard modalities of retinal imaging prior to the introduction of ultra-widefield imaging includes the Topcon Retinal Camera (Topcon, Tokyo, Japan). The revolution of ultra-widefield imaging is not without challenges, as projection from a three-dimensional object — the retina — to a two-dimensional flat image causes peripheral distortions in the image produced. Spaide described these peripheral distortions that occurs with ultra-widefield imaging and proposed an azimuthal projection.8 Sagong et al. have also highlighted the distortions that can occur with varying axial lengths.9

The distortion in the peripheral retina in ultra-widefield imaging is well-established, but the posterior segment has not been thought to suffer this effect significantly. Furthermore, for the optic disc and macula, the projection deformity should only affect the size, if any, and not the vertical-to-horizontal ratio of images.

Oishi et al. imaged an artificial eye with using the Optos ultra-widefield imaging system (Optos 200TX; Optos, Dunfermline, United Kingdom). Using the entire length of the image, they identified a 1.12-factor horizontal stretch and peripheral magnification but did not report posterior pole distortion specifically. Therefore, the “stretch” applied to posterior segment anatomical landmarks such as the optic disc, which are important structures in the function of the visual system, has not yet been explored.

Acknowledgement of any significant posterior pole distortion is important as image analysis and grading tools developed for ultra-widefield images require an understanding of this potential difference. Secondly, we need to ascertain whether clinical decisions based on pre-existing data on the optic disc and fovea from conventional imaging are transferable to ultra-widefield imaging.

The objective of this study was to describe the posterior segment distortion in Optos ultra-wide-field imaging compared to conventional Topcon retinal camera by assessing the vertical-to-horizontal disc diameter ratio of the optic disc. We also hypothesized that variations in the posterior pole, if present, are secondary to the horizontal distortions that may be incorporated to the images from the ultra-widefield imaging system. The ultra-widefield imaging system utilizes an ellipsoidal mirror with a fast vertical scanning laser ophthalmoscope and a motor driven slower horizontal scan. Should the distortion be secondary to projection errors, this will only explain peripheral changes with no horizontal-to-vertical disparity. Therefore, an object that is rotated 90° will impose a distortion that is different compared to the image taken in a conventional position. We also explored the hypothesis that averaging the two images taken 90° apart will reduce the posterior segment distortion and tested this on a simplified model eye.

Patients and Methods

This retrospective image analysis study was performed in the National Institute for Health Research Moorfields Biomedical Research Centre and the Institute of Ophthalmology, University College London, United Kingdom. Local institutional review board approval was obtained for analysis of anonymized data, and the study was conducted in accordance to the tenets of the Declaration of Helsinki.

Image acquisition

Part I: Vertical-to-Horizontal Disc Diameter Ratio Comparison. Color fundus images were taken using the Optos 200TX and Topcon retinal camera. Macula-centered ultra-widefield color fundus image from each eye and a 50° macula-centered color fundus image taken from the Topcon system were obtained from of 20 consecutive eyes with no optic disc pathology.

Part II: Comparison Between Imaging in a Conventional Position and With 90° Rotation. Images from a single patient who has had macula-centered ultra-widefield color fundus image from the Optos 200TX taken in the conventional position and a second image with the head rotated 90° were used. The image taken 90° from the conventional position was then realigned for analysis. A seven-field 35° Topcon color fundus image from the same subject was also used for comparison.

Part III: Proposed Solution. A model eye was made from a 40-mm plastic sphere with a scale along the vertical and horizontal with each mark 0.5 mm apart. The vertical scale is grey and the horizontal scale is white. An aperture with an 8-mm diameter was created to simulate the pupil and a +20.00 diopter (D) lens was used in front of the aperture. Two central Optos images were taken with the model eye rotated 90° apart. The images were taken when the “green in-focus” signal was obtained. This was repeated with the Topcon system. The images taken in the 90° rotated position were realigned for analysis.

Data Collection and Image Analysis

All images were analyzed using ImageJ, an open source software by the National Institutes of Health that can be downloaded from http://imagej.nih.gov/ij/index.html.

Part I. Two investigators (EM, LYG) measured the vertical disc diameter and horizontal disc diameter from 50° using Topcon retinal camera macula-centered color fundus images and ultra-widefield color fundus images from 20 eyes. The vertical-to-horizontal disc diameter ratio was calculated, with a ratio of more than one, indicating the disc is longer vertically and vice versa. An average was obtained from the two investigators.

Refractive errors and axial lengths were obtained, if available, from 20 eyes. Eyes with a spherical equivalent refractive error of less than or equal to −1 D or an axial length of greater than 24 mm were grouped as myopic and refractive errors of greater than or equal to + 1 D or an axial length of less than 23.5 mm were grouped as hyperopic. For eyes that had refractive or cataract surgery, the refractive error prior to surgery was used. The average vertical-to-horizontal disc diameter ratio for these two groups for both the Optos ultra-widefield and Topcon Retinal Camera images were determined.

Part II. The montage of the Topcon seven-field images was produced using a trial version of i2k Retina (DualAlign, Clifton Park, NY). The optic disc and the emanating vessels from the optic disc of each image was used as the center, and all three images were superimposed using Adobe Photoshop CS4 extended edition (Photoshop; Adobe Systems, San Jose, CA) with the optic disc and emanating vessels serving as a reference point. The field of view between the two Optos ultra-widefield images were also studied qualitatively to evaluate the field of view and distortion by tracing the retinal vasculature and superimposing all three.

To quantify angular distortion, bifurcations or arteriovenous crossings that are visible in all three images were marked. The location of these points were grouped to either 0 to 2.5 disc diameters (posterior segment) or greater than 2.5 disc diameters (peripheral) from the disc center. The angles from the center of the optic disc to these points were measured using ImageJ. Absolute values were used; therefore, an angular distortion of 5° clockwise and a 5° distortion anticlockwise will give an average value of 5 rather than 0.

Part III. Two graders (LN, CVA) measured the vertical and horizontal lengths of the grid as in Figure 1. Measurement one was obtained by dividing V1 with H1 and measurement two was determined by dividing (V2 + V3) with (H2 + H3). The average values from the two graders were used for analysis. This was performed for the ultra-widefield images of the model eye in the conventional and 90° rotated position. The same was performed for Topcon 50° images. The average measurements from both the Optos ultra-widefield images in the conventional and 90° rotated images were also calculated as the hypothesis that averaging the values will reduce the distortion obtained. The vertical-to-horizontal ratio should equate to 1.0 if there is no distortion.


A representation of the measurements used to determine the vertical-to-horizontal ratio of the images taken with the Optos ultra-widefield system and the Topcon retinal camera of a model eye in two positions 90° apart. Measurement one was obtained by dividing V1 with H1 and measurement two was determined by dividing (V2 + V3) with (H2 + H3).

Figure 1.

A representation of the measurements used to determine the vertical-to-horizontal ratio of the images taken with the Optos ultra-widefield system and the Topcon retinal camera of a model eye in two positions 90° apart. Measurement one was obtained by dividing V1 with H1 and measurement two was determined by dividing (V2 + V3) with (H2 + H3).

Vertical and horizontal disc diameter measurements were also measured from the images obtained from the human subject in Part II. The Topcon image, Optos ultra-widefield image in the conventional position, and the Optos ultra-widefield image with the subject rotated 90° were analyzed.

Statistical Analysis

Statistical analyses were performed using statistical software (SPSS version 21; IBM, Tokyo, Japan) and Stata version 12.1 (StataCorp LP, College Station, TX) was used for sample size calculation. Descriptive statistics were reported as mean and standard deviation. The Mann-Whitney U test was used to compare means. To calculate the sample size, we used results from an unpublished pilot study identifying the mean values in ultra-widefield and Topcon 3D Optical Coherence Tomography (OCT) (Topcon, Tokyo, Japan) color fundus images. As the vertical-to-horizontal disc diameter ratio was lower in the ultra-widefield images, we have taken the upper 95% confidence interval (CI) value of the ultra-wide-field image measurements, 1.01, and the lower 95% CI value of the Topcon 3D OCT color fundus image measurements, 1.08, with a standard deviation across both groups of 0.074. Sample size required for a 0.05 Type 1 error rate and 90% power was 18 subjects. Significance level was set at 0.05. Inter-rater agreement was determined using the absolute agreement intraclass correlation coefficient. Average measure agreement levels were used to report the agreement levels. Agreement levels were interpreted according to the guidelines proposed by Landis and Koch,11 as detailed in Table 1.


Interpretation of ICC Agreement Scores According to the Guidelines Suggested by Landis and Koch.15

Table 1:

Interpretation of ICC Agreement Scores According to the Guidelines Suggested by Landis and Koch.15

Results

Part I

Twenty eyes from 13 consecutive patients with a mean age of 64.7 years and a male-to-female ratio of 9:4 were analyzed. The vertical-to-horizontal disc diameter ratio agreement between the two investigators were “almost perfect” with an intraclass correlation coefficient of 0.817 for Optos ultra-widefield images and 0.845 for Topcon retinal camera images.

The mean vertical-to-horizontal disc diameter ratio for Optos ultra-widefield images was 0.956 ± 0.078 (95% CI, 0.919–0.993) and 1.083 ± 0.081 (95% CI, 1.045–1.121) for the Topcon retinal camera system, as described in Table 2. The difference between the two groups were statistically significant (P < .001). Figure 2 shows the image of the optic disc for both imaging modalities.


The Vertical-to-Horizontal Disc Diameter Ratio From Optos Ultra-Widefield Images and Topcon Retinal Camera Images From 20 Eyes With Subgroup Analysis Based on Refractive Error/Axial Length

Table 2:

The Vertical-to-Horizontal Disc Diameter Ratio From Optos Ultra-Widefield Images and Topcon Retinal Camera Images From 20 Eyes With Subgroup Analysis Based on Refractive Error/Axial Length


Optic disc images of the same eye. The left image was taken with a Topcon retinal camera, middle image with the Optos ultra-widefield system with the head in a conventional position, and the right image taken with the Optos ultra-widefield system with head in the 90° rotated position with the resulting image rotated back.

Figure 2.

Optic disc images of the same eye. The left image was taken with a Topcon retinal camera, middle image with the Optos ultra-widefield system with the head in a conventional position, and the right image taken with the Optos ultra-widefield system with head in the 90° rotated position with the resulting image rotated back.

The vertical-to-horizontal disc diameter ratio for the myopic group (n = 4) was 0.929 and 1.074 in the ultra-widefield and Topcon images, respectively. As for the hyperopic group (n = 4), the ratios were 0.957 and 1.103 for ultra-widefield and Topcon images, respectively. The difference in ratio between the ultra-widefield and Topcon images in the myopic group was 0.145 and 0.146 for the hyperopic group.

Part II

The clarity of the superior and inferior retina is improved with the head placed in the 90° rotated position and the view of the temporal and nasal retina is better with the eye in the conventional position, as presented in Figure 3 and Figure 4. In Figure 4, the retinal veins and retinal arteries for all three imaging methods were superimposed in a single image to illustrate the field of view imaged and the distortion experienced.


Widefield color fundus images of the same eye taken with the Optos ultra-widefield system and a seven-field Topcon retinal camera montage. Left image: central Optos ultra-widefield color fundus image taken with the head in a conventional position. Middle image: central ultra-widefield image taken with the head rotated 90° and the image is then rotated back. Right image: seven-field montage of 35° Topcon retinal camera color fundus images. The peripheral retina imaged using the Optos ultra-widefield images is wider compared to the seven-field image, as one would expect. Comparing the two Optos ultra-widefield images, the right image has a clear and wider view of the nasal and temporal retina but limited vertically. This is the opposite in the middle image, with a clear and wider view of the superior and inferior retina but limited horizontally.

Figure 3.

Widefield color fundus images of the same eye taken with the Optos ultra-widefield system and a seven-field Topcon retinal camera montage. Left image: central Optos ultra-widefield color fundus image taken with the head in a conventional position. Middle image: central ultra-widefield image taken with the head rotated 90° and the image is then rotated back. Right image: seven-field montage of 35° Topcon retinal camera color fundus images. The peripheral retina imaged using the Optos ultra-widefield images is wider compared to the seven-field image, as one would expect. Comparing the two Optos ultra-widefield images, the right image has a clear and wider view of the nasal and temporal retina but limited vertically. This is the opposite in the middle image, with a clear and wider view of the superior and inferior retina but limited horizontally.


Traces of the main veins (left) and arteries (right) from color fundus images taken with the Optos ultra-widefield system and a seven-field Topcon retinal camera color fundus montage from the same eye. Blue lines represent the tracings of the central Optos ultra-widefield color fundus image taken with the head in a conventional position, and red represents the vessels of the central ultra-widefield image taken with the head rotated 90°. The image is then rotated back, and green represents the tracings of the seven-field Topcon retinal camera color fundus montage. The horizontal field of view of the Optos is markedly wider with the head in the conventional position (blue), and the vertical field of view is increased with the unconventional position (red).

Figure 4.

Traces of the main veins (left) and arteries (right) from color fundus images taken with the Optos ultra-widefield system and a seven-field Topcon retinal camera color fundus montage from the same eye. Blue lines represent the tracings of the central Optos ultra-widefield color fundus image taken with the head in a conventional position, and red represents the vessels of the central ultra-widefield image taken with the head rotated 90°. The image is then rotated back, and green represents the tracings of the seven-field Topcon retinal camera color fundus montage. The horizontal field of view of the Optos is markedly wider with the head in the conventional position (blue), and the vertical field of view is increased with the unconventional position (red).

Forty-seven clear, separate bifurcations were identified in all three images. Twenty-two of the 47 points used to quantify the angular distortion were in the posterior segment and 25 points were in the periphery. The mean absolute angular distortion of all 47 points in the Optos ultra-widefield image in the conventional position when compared with Topcon retinal camera seven-field montage was 3.08° ± 1.65° (95% CI, 2.60–3.56; P= .001). In the 90° rotated position, the absolute angular distortion was 5.04° ± 2.56° (95% CI, 4.29–5.79; P< .0001). Table 3 illustrates the absolute angular differences when compared with Topcon retinal camera seven-field montage.


Mean Absolute Angular Distortion When Compared With Topcon Images (Degrees)

Table 3:

Mean Absolute Angular Distortion When Compared With Topcon Images (Degrees)

Part III

The average vertical-to-horizontal ratio of the Optos ultra-widefield image in the conventional position was 0.910 and in the 90° rotated position was 1.072. When imaged using the Topcon system, the vertical-to-horizontal ratio in the conventional position was 1.008 and 0.999 in the 90° rotated position. Using the average values from the conventional and 90° rotated image of the ultra-widefield image, the vertical-to-horizontal ratio was 0.987.

As for the human subject, the vertical-to-horizontal disc diameter ratio in the Optos ultra-widefield from a conventional position was 1.114 and 1.316 in the rotated position. The average value of the vertical-to-horizontal disc ratio taken in the conventional and 90° rotated position was 1.215. The vertical-to-horizontal disc diameter ratio was 1.219 using the Topcon system.

Discussion

In this study, we described a significant distortion of the posterior segment on Optos ultra-widefield images when compared to Topcon retinal camera. The mean vertical-to-horizontal disc diameter ratio of similar subjects was 0.956 in Optos compared to 1.083 on Topcon. The angular distortion of the blood vessels was also significantly distorted in the ultra-widefield images. With a 90° rotation, the image obtained showed different magnitudes of angular distortions with the image in the conventional position despite using the same imaging system further substantiating the suggestion of a horizontal distortion. However, the field of view appears improved in the horizontal meridian when imaged and, therefore, images taken 90° apart will improve the field imaged in the vertical and horizontal without the need of steering. We have also identified using a model eye that by averaging the values in the two images taken at 90° apart using the Optos system, the vertical-to-horizontal ratio — which should read 1.0 — improved to 0.987 compared to 0.910 in the conventional position.

The horizontal stretch applied throughout the image obtained including the posterior segment is likely to be secondary to the system used to image the retina in the Optos system. It is known that the superior and inferior retina is poorly visualized in Optos images necessitating the need for steering superiorly and inferiorly. The presence of the lids was believed to contribute to the poor visualization of the superior and inferior retina. Oishi et al. have shown in their model eyeball that the vertical peripheries were poorly visualized.10 The Optos system also uses an ellipsoidal mirror to obtain a second focus point at the iris plane and is constructed to obtain images in a non-mydriatic eye so it is unlikely that the lids cause this limitation. The scanning laser ophthalmoscope has fast vertical sections and sweeps horizontally using a slower motor-driven system to obtain the images. The finding in our study that the horizontal stretch is present not only in the peripheries but also the posterior segment augments the theory that the system utilized causes the distortion rather than purely a projection error from a three-dimensional object to a flat, two-dimensional image.

Averaging the topographical location from the two images obtained using Optos in the conventional and 90° rotated position is an interesting solution in reducing the distortion. We explored this proposition using images from a single subject and using a model eye with a grid with images taken 90° apart and realigning the images for analysis. The field image and clarity of the superior and inferior retina is improved in the 90° rotated position without the need for steering. Obviously, rotating one's head is neither convenient nor practical, but a rotating ellipsoidal mirror imaging system taking both vertical and horizontal sweeps and subsequently combining these images can, theoretically at least, increase the clarity of the peripheral retina both in the vertical and horizontal without the need for steering. By taking the average of the topographical location of the images, it will also reduce the distortion introduced in the Optos imaging system as shown in the exploratory section using the model eye.

In conclusion, despite the advantages of the Optos system, there is a horizontal stretch in raw images at the posterior pole. The clinical relevance of this aberration needs to be acknowledged when measurements of anatomical landmarks in the posterior segment are performed such as the optic disc or possibly the foveal avascular zone although the latter is not formally assessed here. The proposed averaging of images taken 90° apart can improve the quality of the images obtained using the Optos system. An acknowledgment and correction of this posterior segment distortion will increase the accuracy that the revolutionary Optos system has to offer.

References

  1. No authors listed. Grading diabetic retinopathy from stereoscopic color fundus photographs – an extension of the modified Airlie House classification. ETDRS report number 10. Early Treatment Diabetic Retinopathy Study Research Group. Ophthalmology. 1991;98(5 Suppl):786–806. doi:10.1016/S0161-6420(13)38012-9 [CrossRef]
  2. Karampelas M, Sim DA, Chu C, et al. Quantitative analysis of peripheral vasculitis, ischemia, and vascular leakage in uveitis using ultra-widefield fluorescein angiography. Am J Ophthalmol. 2015;159(6):1161–1168.e1. doi:10.1016/j.ajo.2015.02.009 [CrossRef]
  3. Lengyel I, Csutak A, Florea D, et al. A population-based ultra-widefield digital image grading study for age-related macular degeneration-like lesions at the peripheral retina. Ophthalmology. 2015;122(7):1340–1347. doi:10.1016/j.ophtha.2015.03.005 [CrossRef]
  4. Sim DA, Keane PA, Rajendram R, et al. Patterns of peripheral retinal and central macula ischemia in diabetic retinopathy as evaluated by ultra-widefield fluorescein angiography. Am J Ophthalmol. 2014;158(1):144–153.e1. doi:10.1016/j.ajo.2014.03.009 [CrossRef]
  5. Tsui I, Kaines A, Havunjian MA, et al. Ischemic index and neovascularization in central retinal vein occlusion. Retina. 2011;31(1):105–110. doi:10.1097/IAE.0b013e3181e36c6d [CrossRef]
  6. Pe'er J, Sancho C, Cantu J, et al. Measurement of choroidal melanoma basal diameter by wide-angle digital fundus camera: a comparison with ultrasound measurement. Ophthalmologica. 2006;220(3):194–197. doi:10.1159/000091765 [CrossRef]
  7. Kornberg DL, Klufas MA, Yannuzzi NA, Orlin A, D'Amico DJ, Kiss S. Clinical utility of ultra-widefield imaging with the Optos Optomap compared with indirect ophthalmoscopy in the setting of non-traumatic rhegmatogenous retinal detachment. Semin Ophthalmol. 2015Aug21:1–8. [Epub ahead of print]
  8. Spaide RF. Peripheral areas of nonperfusion in treated central retinal vein occlusion as imaged by wide-field fluorescein angiography. Retina. 2011;31(5):829–837. doi:10.1097/IAE.0b013e31820c841e [CrossRef]
  9. Sagong M, van Hemert J, Olmos de Koo LC, Barnett C, Sadda SR. Assessment of accuracy and precision of quantification of ultra-widefield images. Ophthalmology. 2015;122(4):864–866. doi:10.1016/j.ophtha.2014.11.016 [CrossRef]
  10. Oishi A, Hidaka J, Yoshimura N. Quantification of the image obtained with a wide-field scanning ophthalmoscope. Invest Ophthalmol Vis Sci. 2014;55(4):2424–2431. doi:10.1167/iovs.13-13738 [CrossRef]
  11. Landis JR, Koch GG. The measurement of observer agreement for categorical data. Biometrics. 1977;33(1):159–174. doi:10.2307/2529310 [CrossRef]

Interpretation of ICC Agreement Scores According to the Guidelines Suggested by Landis and Koch.15

ICC ScoreInterpretation
0.81 – 1.00Almost Perfect
0.61 – 0.80Substantial
0.41 – 0.60Moderate
0.21 – 0.40Fair
< 0.21Slight

The Vertical-to-Horizontal Disc Diameter Ratio From Optos Ultra-Widefield Images and Topcon Retinal Camera Images From 20 Eyes With Subgroup Analysis Based on Refractive Error/Axial Length

Vertical-to-Horizontal Disc Diameter RationOptos Ultra-Widefield ImageTopcon Retinal Camera ImageMean DifferenceP Value
All eyes200.956 [95% CI, 0.919–0.993]1.083 [95% CI, 1.045–1.121]0.127< .001
Myopic eyes (Spherical equivalent = −1 Diopter or axial length > 24 mm)40.9291.0740.145
Hyperopic eyes (Spherical equivalent = −1 Diopter or axial length > 24 mm)40.9571.1030.146

Mean Absolute Angular Distortion When Compared With Topcon Images (Degrees)

Optos Ultra-Widefield With Conventional Head PositionOptos Ultra-Widefield With 90° Rotation
Posterior segment (22 points)2.60 ± 1.38 (1.99, 3.21)5.15 ± 2.46 (4.06, 6.24)
Periphery (25 points)3.49 ± 1.78 (2.70, 4.28)4.95 ± 2.69 (3.76, 6.14)
All 47 points3.08± 1.65 (2.60, 3.56)5.04 ± 2.56 (4.29, 5.79)
Authors

From the National Institute for Health Research Moorfields Biomedical Research Centre, Moorfields Eye Hospital and University College London Institute of Ophthalmology, London, United Kingdom (LN, EM, CV, IC, MC, PGH, SS); and the Royal Derby Hospital, Derby, Uttoxeter New Rd, DE22 3NE, Derby, United Kingdom (LYG).

Dr. Vazquez-Alfageme has received funding from the Spanish Retina and Vitreous Society, Spain. Dr. Hykin has received grants from Novartis, Allergan, and Bayer and is on the advisory board and receives speaker fees from Novartis and Bayer. Dr. Sivaprasad has received grants from Novartis, Allergan, and Bayer and is on the advisory board and receives speaker fees from Novartis, Allergan, and Bayer. The remaining authors report no relevant financial disclosures.

Address correspondence to Sobha Sivaprasad, DM, FRCOphth, National Institute for Health Research Moorfields Biomedical Research Centre and University College London Institute of Ophthalmology, Moorfields Eye Hospital, 162 City Road, EC1V 2PD, London, United Kingdom; email: senswathi@aol.com.

Received: November 21, 2015
Accepted: April 22, 2016

10.3928/23258160-20160707-06

Sign up to receive

Journal E-contents