A machine vision method for automated alignment of fundus imaging systems.

BACKGROUND AND OBJECTIVE: To report a method to track the pupil in three axes simultaneously prior to imaging the fundus. PATIENTS AND METHODS: The system is based on parallax optical alignment to detect the center of the pupil. The system consists of two cameras acquiring pupil images from two distinct directions and an operator-supervised algorithm to derive the coordinates of the pupil center and output of commands to drive a three-axes computer-controlled stage. The system was tested in a cohort of 45 individuals 61 ± 15 years of age, 26 with and 19 without glaucoma. The tracking was performed without pharmacologic pupil dilation. RESULTS: The variability of the pupil center determination (assessed by the standard deviation) was ± 0.19 and ± 0.33 mm for the vertical and horizontal directions, respectively. The processing time of the algorithm was 0.75 msec. The tracking converged to within a preset tolerance of ± 0.5 mm in 45 of the 45 eyes. CONCLUSION: Image acquisition and processing of pupil images can be used to align fundus imaging systems rapidly, accurately, and with minimal operator intervention.

A computer algorithm to quantitatively assess quality of digital optic disc images.

Optic disc photography is used in the management and study of glaucoma. Quality assessment is needed at the time of acquisition and during review. A computerized algorithm for objective quality assessment was developed to mimic the procedure used by human observers. It was tested on film-based images obtained with mydriasis (40 normal and 46 glaucomatous eyes) and non-mydriatic digital images (30 normal and 38 glaucomatous eyes). The image sharpness was graded by six masked readers into four categories. The area under the receiver operating characteristic curve for identifying unreadable images was 1.0 for the digital and film-based images and 0.91 and 1.0 for differentiating between unreadable and mediocre images for digital and film-based images, respectively. This pilot study demonstrates that the algorithm can identify all unreadable images. Further studies are necessary to test whether it can be applied to images obtained in other locations on the fundus and with additional cameras.

An image based auto-focusing algorithm for digital fundus photography.

In fundus photography, the task of fine focusing the image is demanding and lack of focus is quite often the cause of suboptimal photographs. The introduction of digital cameras has provided an opportunity to automate the task of focusing. We have developed a software algorithm capable of identifying best focus. The auto-focus (AF) method is based on an algorithm we developed to assess the sharpness of an image. The AF algorithm was tested in the prototype of a semi-automated nonmydriatic fundus camera designed to screen in the primary care environment for major eye diseases. A series of images was acquired in volunteers while focusing the camera on the fundus. The image with the best focus was determined by the AF algorithm and compared to the assessment of two masked readers. A set of fundus images was obtained in 26 eyes of 20 normal subjects and 42 eyes of 28 glaucoma patients. The 95% limits of agreement between the readers and the AF algorithm were -2.56 to 2.93 and -3.7 to 3.84 diopter and the bias was 0.09 and 0.71 diopter, for the two readers respectively. On average, the readers agreed with the AF algorithm on the best correction within less than 3/4 diopter. The intraobserver repeatability was 0.94 and 1.87 diopter, for the two readers respectively, indicating that the limit of agreement with the AF algorithm was determined predominantly by the repeatability of each reader. An auto-focus algorithm for digital fundus photography can identify the best focus reliably and objectively. It may improve the quality of fundus images by easing the task of the photographer.

Retinal thickness in eyes of older normal individuals and its implication for the diagnosis of glaucoma.

PURPOSE: Retinal thickness (RT) is a useful measurement for describing diseases that affect the thickness of the retina, such as glaucoma. Existing normative data are derived from relatively young individuals; however, glaucoma is most prevalent in older individuals. We therefore studied the RT in older normal individuals. PATIENTS AND METHODS: Participants of the Baltimore Eye Study, persons accompanying patients, and staff were recruited and underwent visual field testing and a comprehensive eye examination by a glaucoma specialist. RT was measured with the retinal thickness analyzer (RTA, Talia Technology) and RT values in specific regions were derived using a custom-designed MatLab program. RESULTS: One hundred and three eyes of sixty-two individuals were studied. Mean age was 61 years. Sixty-six percent were female and 82% were of European descent. The average mean deviation on visual field testing was 0.03 dB and the average pattern SD was 1.51 dB. The mean RT of the entire macula was 159+/-16 microm, and was lowest in the foveal pit and highest in the parafoveal annulus. The average distance from the foveal pit to the thickest point in the parafoveal annulus was 1240+/-138 microm. The mean RT of the entire macula was slightly less in older individuals (slope=-5.7 microm/10 y, P=0.02) but the height of the parafoveal annulus relative to the foveal pit, which is determined by the combined thickness of the parafoveal nerve fibers, ganglion cells, inner plexiform layer, and inner nuclear layer, did not vary with age (P=0.62). CONCLUSIONS: Although the average RT of the entire macula was slightly thinner with increasing age, the height of the parafoveal annulus relative to the foveal pit did not change with age and would therefore seem to be a better marker of neuronal tissue health than the average RT of the entire macula.

Test-retest variability in structural and functional parameters of glaucoma damage in the glaucoma imaging longitudinal study.

PURPOSE: To determine the test-retest variability in perimetric, optic disc, and macular thickness parameters in a cohort of treated patients with established glaucoma. PATIENTS AND METHODS: In this cohort study, the authors analyzed the imaging studies and visual field tests at the baseline and 6-month visits of 162 eyes of 162 participant in the Glaucoma Imaging Longitudinal Study (GILS). They assessed the difference, expressed as the standard error of measurement, of Humphrey field analyzer II (HFA) Swedish Interactive Threshold Algorithm fast, Heidelberg retinal tomograph (HRT) II, and retinal thickness analyzer (RTA) parameters between the two visits and assumed that this difference was due to measurement variability, not pathologic change. A statistically significant change was defined as twice the standard error of measurement. RESULTS: In this cohort of treated glaucoma patients, it was found that statistically significant changes were 3.2 dB for mean deviation (MD), 2.2 for pattern standard deviation (PSD), 0.12 for cup shape measure, 0.26 mm for rim area, and 32.8 microm and 31.8 microm for superior and inferior macular thickness, respectively. On the basis of these values, it was estimated that the number of potential progression events detectable in this cohort by the parameters of MD, PSD, cup shape measure, rim area, superior macular thickness, and inferior macular thickness was 7.5, 6.0, 2.3, 5.7, 3.1, and 3.4, respectively. CONCLUSIONS: The variability of the measurements of MD, PSD, and rim area, relative to the range of possible values, is less than the variability of cup shape measure or macular thickness measurements. Therefore, the former measurements may be more useful global measurements for assessing progressive glaucoma damage.

Results of implementation of the DigiScope for diabetic retinopathy assessment in the primary care environment.

Approximately 50% of patients with diabetes in the United States do not undergo recommended ocular evaluations for diabetic retinopathy. The DigiScope (EyeTel Imaging, Inc., Columbia, MD) was developed as a cost-effective and practical telemedicine digital imaging system to screen for diabetic retinopathy in the primary care physician's (PCP) office. The DigiScope has been validated against seven-field stereo color fundus photography for the detection of diabetic retinopathy. This study reports on the implementation of the DigiScope for diabetic retinopathy assessment in the primary care environment. In PCP's offices, patients with diabetes who had not undergone an eye examination in the past year were imaged with the DigiScope. The images were transmitted to a reading center where the need for referral to an ophthalmologist was determined. Nonurgent referral was recommended for patients with diabetic retinopathy greater than a few micro-aneurysms, other ocular pathology, or unreadable images. Referral was deemed "urgent" for patients with sight-threatening disease and evaluation by an ophthalmologist within 72 hours was recommended. Between October 1, 2002 and March 31, 2771 patients with diabetes underwent DigiScope imaging at multiple sites. Nonurgent referral was recommended for 468 patients (17%). Urgent referral was recommended for 71 patients (3%). The images were unreadable in 295 cases (11%). This study indicates that implementation of the DigiScope in the primary care setting is practical and allows screening of patients with diabetes who are otherwise not receiving recommended eye examinations.

Feasibility of screening for high-risk age-related macular degeneration with an Internet-based automated fundus camera.

BACKGROUND AND OBJECTIVE: The objective of this pilot study was to determine whether a new screening system, the DigiScope (EyeTel Imaging, Inc., Columbia, MD), can detect the presence of age-related macular degeneration (AMD) at a level requiring referral to an ophthalmologist for further evaluation and possible treatment. PATIENTS AND METHODS: The DigiScope is an Internet-based semi-automated digital imaging system designed to be in primary care physicians' offices. Forty-two eyes of 21 patients with different categories of AMD were imaged with both the DigiScope and a standard color fundus camera. The imaging capability of the two modalities was compared for identification of lesions associated with AMD and classification into stages. RESULTS: There was good agreement for low-risk lesions and excellent agreement for high-risk lesions. Thirty-five of 36 eyes with intermediate or advanced disease were correctly identified with DigiScope images. Choroidal neovascularization was identified in all cases with the DigiScope due to the presence of subretinal hemorrhage or subretinal fibrosis. The DigiScope was found less capable of detecting subretinal fluid than standard stereo fundus photographs. CONCLUSIONS: This pilot study suggests that the DigiScope may be a useful screening tool for AMD.

Comparison of a digital retinal imaging system and seven-field stereo color fundus photography to detect diabetic retinopathy in the primary care environment.

BACKGROUND AND OBJECTIVE: Because patients with diabetes mellitus may visit their primary care physician regularly but not their ophthalmologist, a retinal risk assessment in the primary care setting could improve the screening rate for diabetic retinopathy. An imaging system for use in the primary care setting to identify diabetic retinopathy requiring referral to an ophthalmologist was evaluated. PATIENTS AND METHODS: In a masked prospective study, images were obtained from 11 patients with diabetes mellitus using both the digital retinal imaging system and seven-field stereo color fundus photography. The ability to obtain gradable images and to identify diabetic retinal lesions was compared. RESULTS: Of all images, 85% of digital retinal imaging system images and 88% of seven-field images were gradable. Agreement based on "no retinopathy" versus "any retinopathy" was excellent (Kappa = 0.96). Agreement based on "microaneurysms or less retinopathy" versus "retinal hemorrhages or worse retinopathy" was very good (Kappa = 0.83). CONCLUSIONS: The agreement between the digital retinal imaging system and seven-field photography indicates that the digital retinal imaging system may be useful to screen for diabetic retinopathy.

A comparison of quantitative mapping and stereoscopic fundus photography grading of retinal thickness in diabetic eyes with macular edema.

OBJECTIVE: To evaluate the ability of a retina specialist's grading of 30 degrees color stereoscopic fundus photographs to identify areas of significant retinal thickening as assessed by the Retinal Thickness Analyzer (RTA) and to determine whether this ability was affected by the presence of retinal pathology. MATERIALS AND METHODS: Thirty-two eyes in 29 patients clinically diagnosed as having diabetic macular edema underwent RTA scanning and nonsimultaneous 30 degrees color stereoscopic fundus photography. Retinal thickness maps of the macular area were generated, and regions with significant retinal thickening (> or = 2 SD above normal values) were identified. A retina specialist reader, masked to the RTA measurements, identified areas with macular edema on the stereoscopic fundus photographs, which subsequently were overlaid on the retinal thickness maps. The sensitivity (percent of significant retinal thickening areas identified by the retina specialist grading the stereoscopic fundus photographs) was calculated separately for areas with and without retinal pathology. Specificity of the stereoscopic fundus photograph grading was assessed similarly. RESULTS: The retina specialist's stereoscopic fundus photography grading identified 78.8% of areas with significant retinal thickening (range over eyes: 20.4%-100%) and was slightly more likely to identify significant retinal thickening when pathology was present (89.6%) than when pathology was not present (78.4%; pooled risk ratio, 1.14 [95% CI = 0.54, 2.42]). Specificity of stereoscopic fundus photography grading was 58%, ie, 42% of areas without significant retinal thickening were (incorrectly) identified as edematous by the stereoscopic fundus photograph grading. This misidentification was more likely if pathology was present (76.9%) than if pathology was not present (41.1%; pooled risk ratio, 1.87 [95% CI = 1.28, 2.73]). CONCLUSION: This study shows the determination of macular edema by a retina specialist reading color stereoscopic fundus photographs is sensitive but not specific with reference to edema identified by the RTA. Furthermore, the presence of retinopathy tends to cause false-positive readings with reference to edema identified by the RTA. These findings indicate the need to use objective, quantitative methods in clinical studies to detect and monitor macular edema.

Comparison between retinal thickness analyzer and optical coherence tomography for assessment of foveal thickness in eyes with macular disease.

PURPOSE: To use the retinal thickness analyzer (RTA) and optical coherence tomography (OCT) scanners for quantitative measurement of retinal thickness in eyes with macular disease. DESIGN: In a cross-sectional study, 44 patients (55 eyes) with macular disease and sufficient media clarity to visualize the fundus using clinical biomicroscopy underwent an ophthalmologic examination, fluorescein angiography, RTA, and OCT during the same visit. METHODS: Foveal and foveal center (foveolar) retinal thickness measurements were obtained by RTA and by OCT. RESULTS: Retinal thickness measurements were obtained by OCT in all 55 eyes and by RTA in 34 eyes (62%, primarily due to interference from media opacities). In the 34 eyes in which measurements were obtained by both instruments, mean foveal thickness was 291 and 269 microm for OCT and RTA, respectively; foveolar thickness was 277 and 265 microm, respectively. OCT and RTA measurements of foveal thickness were strongly correlated (intraclass correlation coefficient = 0.89), as were measurements of the foveolar thickness (intraclass correlation coefficient = 0.94). Topographic maps generated by the two techniques yielded qualitatively similar information. CONCLUSIONS: Overall, there was excellent agreement between RTA and OCT measurements. Each technique has advantages that may make its use preferable in a particular subgroup of eyes or to describe a particular disease process. An important consideration is that media opacities create less interference for OCT than for RTA, so that in study populations with a moderate-to-high prevalence of media opacity, images can be obtained in a greater percentage of eyes by OCT than by RTA.

A fundus camera dedicated to the screening of diabetic retinopathy in the primary-care physician's office.

PURPOSE: To increase the number of diabetic patients being screened for retinopathy, an instrument, the DigiScope, was specifically designed to operate in primary-care physicians' offices. The DigiScope is described and its automated functions are evaluated. METHODS: The DigiScope consists of a semiautomated optical head to acquire fundus images, evaluate visual acuity, and transmit the data to a remote reading center through telephone lines. Normal volunteers and 17 consecutive diabetic patients visiting their primary-care physician were recruited, and nonophthalmic staff performed the acquisition session. RESULTS: The pupil center and working distance were set automatically. Centering was achieved within 750 microm in less than 500 ms. The fundus was successfully focused by an automated algorithm, and an imaging session covering 71 degrees of the posterior pole of both eyes lasted 5.6 +/- 2.4 minutes. It was found that a file-compression ratio of 12 did not degrade the clinical information and allowed data transfer in less than 6 minutes. A pilot study in normal eyes showed that the DigiScope images yielded the same amount of details as conventional color fundus photographs obtained by an expert photographer. CONCLUSIONS: The DigiScope fulfills the instrumental requirements for a practical and cost-effective tool to acquire data needed to identify diabetic patients who must be referred to an eye-care specialist. Widespread screening with the DigiScope may help reduce the risk of vision loss in an estimated 4 million individuals in the United States alone, who currently do not undergo an annual eye examination.

Relation between retinal thickening and clinically visible fundus pathologies in mild nonproliferative diabetic retinopathy.

BACKGROUND AND OBJECTIVE: To determine the association of retinal thickening (RT) with clinically observable retinal pathologies in eyes with mild nonproliferative diabetic retinopathy. PATIENTS AND METHODS: Using an objective quantitative imaging method (Retinal Thickness Analyzer), the ratio relative to normal RT (RTI) was measured in 23 eyes with and 35 eyes without clinically observable diabetic fundus pathology. RTI was analyzed in relation to presence of mild diabetic retinal lesions in the +/-0.5 mm vicinity. RESULTS: The percent of eyes with RTI significantly above normal values did not differ significantly between eyes with and without retinopathy (30% vs 34%). Mean RTI was not associated with local presence of microaneurysms (P=0.92), soft exudates (P=0.55), or retinal hemorrhages (P=0.31). Areas without hard exudates had significantly greater mean RTI (1.10) than areas with exudates (0.97, P=0.009). CONCLUSION: In diabetic patients with mild retinopathy, areas with and without clinically observable retinal pathologies had similar retinal thickness. We conclude that clinical strategies for detection of retinal thickening should not be limited to areas with visible fundus pathologies.

Laser targeted photo-occlusion of rat choroidal neovascularization without collateral damage.

Laser targeted photo-occlusion (LTO) is a novel method being developed to treat choroidal neovascular membranes (CNV) in age-related and other macular degenerations. A photosensitive agent, encapsulated in heat-sensitive liposomes, is administered intravenously. A low power laser warms the targeted tissue and releases a bolus of photosensitizer. The photosensitizer is activated after it clears from the normal choriocapillaris but not from the CNV. Forty-five experimental CNV were induced in seven rats. Five weeks after LTO, complete occlusion was observed by laser targeted angiography (LTA) in 76% of treated CNV, and partial occlusion was found in the remaining 24%. The tissues outside the CNV but within the area treated by LTO showed no flow alteration and no dye leakage. All untreated CNV were patent on LTA at 5 weeks. Light microscopy and electron microscopy confirmed the results in treated and control lesions. Moreover, treated areas next to lesions showed normal photoreceptors, retinal pigment epithelium (RPE), Bruch's membrane and choriocapillaris. These results indicate that LTO may improve current photodynamic therapy by alleviating the need for repeated treatments and by avoiding the long-term risks associated with damage to the RPE and occlusion of normal choriocapillaries.