Validated Telemedicine for Diabetic Retinopathy

The Joslin Vision Network Diabetes Eye Care Program

JERRY CAVALLERANO, OD, PhD · SVEN-ERIK BURSELL, PhD · LLOYD M. AIELLO, MD

Approximately 40% of Americans with diabetes mellitus (DM) do not receive eye examinations for diabetic retinopathy (DR), as suggested by the American Diabetes Association Clinical Practice Recommendations;¹ consequently, only 60% of Americans with DM receive the sight-saving care that has been proved effective over the last 40 years.^2-6 Obstacles that limit or prevent access to appropriate diabetes eye care include geographic and economic barriers, patient or provider unawareness of the value of regular diabetes eye care, patient fear or reluctance to undergo dilated fundus exam, and the fact that DR, even in advanced states, is frequently asymptomatic. Ocular telemedicine for DR, appropriately deployed and properly validated, has the potential to deliver high-quality, chronic, urgent, and emergent care, limited only by the ability to deploy the latest technologies.

WHAT IS TELEMEDICINE?

The American Telemedicine Association (ATA) defines telemedicine as "the use of medical information exchanged from 1 site to another via electronic communications to improve patients' health status."⁷ Telemedicine is the practice of medicine that is interventional and includes relationships between patients and physicians or their surrogates. Telehealth is closely associated with telemedicine and encompasses a broader definition of remote health care that relies on videoconferencing, electronic patient portals to record and monitor vital signs, transmission of digital images and records, electronic consultation, patient education, and continuing medical education.

For many reasons, DR presents an ideal model for disease management by telemedicine and telehealth. There are an estimated 21 million Americans with DM and more than 150 million people worldwide with the condition, with the number of cases increasing at epidemic rates.^8,9 DM is a complex, chronic disease that requires lifelong care. Everyone with DM is at risk of developing DR, which is a well-characterized, sight-threatening microvascular complication of DM;^10,11 there are solid, evidence-based clinical management strategies established over the last 35 years that are effective in preserving vision; DR is frequently asymptomatic in its most treatable stages; the prevalence and incidence of DM, with its accompanying complications, are growing at alarming rates worldwide; and many persons with DM do not receive regular eye care and sight-preserving treatments, even when such care and treatments may be readily available. DM poses significant personal and societal problems, remaining a leading cause of vision loss in industrialized countries. Multiple nationwide clinical trials, including the Diabetic Retinopathy Study,¹² the Early Treatment Diabetic Retinopathy Study (ETDRS),^13,14 the Diabetes Control and Complications Trial,¹⁵ and the United Kingdom Prospective Diabetes Study,¹⁶ have demonstrated effective care and treatment programs for DR, but because many individuals who would benefit from sight-preserving care do not receive this care, DM is a significant cause of preventable vision loss and blindness worldwide.¹⁷

While telemedicine has the potential to extend sight-preserving diabetes eye care, it is crucial that telemedicine programs match the quality of care expected in traditional clinical settings. Because there are evidenced-based methods to preserve vision for those with DM faced with an increased risk of vision loss, ocular telemedicine should include and apply these proven methods. The ATA, recognizing the importance of evidence-based care in telemedicine, established consensus recommendations for ocular telemedicine for DR with the intent to improve clinical outcomes and promote informed and reasonable patient expectations. In 2003, the ATA, in cooperation with the United States National Institute of Standards and Technology and the ATA Ocular Telehealth Special Interest Group, initiated a program to prepare and publish Telehealth Practice Recommendations for Diabetic Retinopathy.^18,19 Because ETDRS 30° stereo 7-standard fields color 35-mm slides are an accepted standard for evaluating DR, the recommendations accepted ETDRS 7-standard fields photography as the criterion to assess the accuracy of a telemedicine system for DR.^19,20 The guidelines suggest that telemedicine programs for DR compare favorably with ETDRS film photography as reflected in kappa values for agreement of diagnosis, false positive and false negative readings, positive predictive value, negative predictive value, sensitivity, and specificity of diagnosing levels of DR. While there are standards other than ETDRS 30° stereo 7-standard fields for grading DR, protocols should define alternate standards used for validation if ETDRS photos are not used.

Telehealth programs for DR should clearly define program goals and program performance in relation to accepted clinical standards. In general, the selection of an ocular telehealth system for evaluating DR should be based on the unique needs of the healthcare setting. The Telehealth Practice Recommendations for Diabetic Retinopathy prepared by the ATA clearly delineate performance standards for clinical, technical, and administrative elements of ocular telemedicine for DR (Tables 1, 2, and 3).^18,19

The ATA Practice Recommendations defined 4 clinical categories of assessment for validation (Table 1). Technology considerations addressed the following areas: image acquisition, compression, analysis, data communication and transfer, display capabilities, archiving and retrieval, security, reliability and redundancy, and documentation. The recommendations are summarized in Table 2. Operational and business recommendations are summarized in Table 3. Importantly, by clearly defining and applying standards, the Telehealth Practice Recommendations for Diabetic Retinopathy apply evidence-based clinical care to ocular telemedicine for DR.

THE JOSLIN VISION NETWORK DIABETES EYE CARE PROGRAM

The Joslin Vision Network (JVN) Diabetes Eye Care Program is an ATA-defined category 3 telemedicine program for DR. The JVN has been rigorously validated to demonstrate reliable diagnosis compared to current clinical standards and identifies clinical level of DR accurately compared to ETDRS 7-standard fields 35-mm stereoscopic slides and clinical examination by a retinal specialist. It has demonstrated the ability to identify accurately non–diabetes-related eye disease. Reports of clinical applications of the JVN demonstrate the value and potential of this telemedicine program in providing eye care for DR.

The JVN Diabetes Eye Care Program has been described elsewhere.^21,22 The commercially available non-mydriatic retinal fundus camera (Topcon TRC-NW5S or TRC-NW6S, Paramus, NJ) was optimized for low–light-level imaging of the retina. For each retinal field, a stereoscopic pair of images was acquired by a manual horizontal translation of the fundus camera, as is standard for obtaining nonsimultaneous stereoscopic retinal photographs. An external view image was also acquired for evaluation of ocular adnexa and media clarity. These images are labeled, digitized, and stored on the JVN system using proprietary software. Of note, in the earlier applications involving validation studies, images were compressed (10:1 JPEG/95% picture fidelity) because of restriction in bandwidth, while current JVN operation does not require image compression, with all retinal images transmitted and stored in their native state. The images are captured in true color (24 bits) at a resolution of 640 × 480 pixels. Relevant patient demographics, medical histories, related laboratory results, and clinical study data are also recorded.

Imagers and readers who successfully complete a JVN certification program and who demonstrate ongoing ability to meet quality standards according to JVN protocol can acquire and grade JVN images. Images are acquired with no pharmacological pupil dilation and archived on JVN servers. Three 45° retinal fields [(1) the optic disc and macula, (2) superior temporal to the optic disc, and (3) nasal to the optic disc] and two 30° retinal fields (corresponding to ETDRS standard fields 1 and 2) are captured using nonsimultaneous stereo imaging procedure according to JVN protocol.^21-23 The JVN field definitions using a nonmydriatic camera can be accessed online at: www.retinalphysician.com/table2.htm. Images are graded by certified JVN readers at the Beetham Eye Institute of the Joslin Diabetes Center (Figures 1 and 2).^22-24 Stereo image viewing is achieved using LCD-shuttered goggles (Stereographics, San Rafael, CA) and results are recorded on JVN templates designed for clinical diagnosis. The JVN template can be accessed online at: www.retinalphysician.com/table2.htm. Upon full population of the templates, the JVN software calculates the clinical level of DR based on ETDRS algorithms.

JVN VALIDATION

JVN Compared to 35-mm Slides

A prospective study validated the ability of the JVN to determine clinical levels of DR and DME, timing of next retinal evaluation, and need to refer to ophthalmology specialists using JVN stereoscopic nonmydriatic digital-video color retinal images as compared to ETDRS 7-standard fields 35-mm stereoscopic color fundus photographs (ETDRS photos).²² The study enrolled 54 patients (108 eyes) with type 1 or type 2 DM with various levels of DR and DME. Nonsimultaneous 45°-field stereoscopic digital-video color images (JVN images) were obtained according to JVN protocol prior to pupil dilation. Following pupil dilation, ETDRS photos were obtained. JVN images and ETDRS photos were graded on a lesion-by-lesion basis by 2 independent, masked readers to assess ETDRS clinical level of DR. An independent retinal specialist adjudicated inter-reader disagreements in a masked fashion. ETDRS clinical level of DR, timing of next ophthalmic evaluation of DR, and need for prompt referral to retinal specialist were compared.

Figure 1. Joslin Vision Network Fields compared to Early Treatment Diabetic Retinopathy (ETDRS) 7-standard fields. NM-1 = JVN 45° nonmydriatic field 1, NM-2 = JVN 45° nonmydriatic field 2, NM-3 = JVN nonmydriatic field 45° 3. F1-F7 = ETDRS 30° mydriatic fields 1-7. JVN also captures 30° nonmydriatic images of ETDRS fields 1 and 2. All fields are imaged using a nonsimultaneous stereo image capture protocol.

There was substantial agreement (κ=.65) between the clinical level of DR assessed from the undilated JVN images and the dilated ETDRS photos. Agreement was excellent (κ=.87) for suggested referral to specialists for eye examinations (Table 4). This study showed that undilated digital-video images using the JVN system were comparable to dilated ETDRS photographs in determining the clinical level of DR and validated the agreement between nonmydriatic JVN images and dilated ETDRS photos.

Figure 2. Representative JVN fields for the right eye. Top row: NM-2 – superior temporal field; external view. Second row: NM-1 – optic disc and macula field; NM-3 – nasal field. Bottom row: macula 30° field and optic disc 30° field.

Retrospective Study: JVN vs Retinal Specialist

A retrospective study measured the accuracy of the JVN to diagnose clinical level of DR, determine follow-up, and appropriately refer for comprehensive eye examination in patients who had both JVN imaging and clinical retinal examination through dilated pupils by a retinal specialist.²³ Patients who received diabetes care in Joslin Diabetes Center's adult diabetes clinic as part of Joslin's 3.5–day-long Diabetes Outpatient Intensive Treatment (DO IT) program, an intensive program of diabetes care and education, had retinal imaging according to JVN protocol. Images were graded by certified JVN readers and patients with significant retinal disease underwent evaluation by retinal specialists.

Overall, 268 (51.1%) of 525 imaged patients had comprehensive eye examination by a retinal specialist due to referable JVN-assessed clinical level of DR, nondiabetic ocular disease, ungradable images, last eye examination > 12 months, or patient request for examination. JVN diagnosis of clinical level of DR agreed exactly with clinical findings in 388 eyes (72.5%) and was at or within 1 level in 478 eyes (89.3%). JVN referral based on most severe diagnosis in either eye matched retinal specialist recommended follow-up in 248/268 of patients (92.5%) (Table 5). Additionally, 136/525 of JVN patients (25.9%) had nondiabetic ocular abnormalities requiring referral. Recommended follow-up from JVN imaging compared favorably to clinical examination by a retinal specialist and significant nondiabetic ocular pathology was identified.

Prospective Study: JVN vs Retinal Specialist

In an attempt to assess whether the JVN diabetes eyecare program had the potential to be a surrogate exam for patients with previously documented little or no DR, a prospective study evaluated the JVN for follow-up annual retinal examination for level of DR.²⁴ Fifty-two patients with documented no or mild nonproliferative DR (ETDRS level ≤35) and no DME at dilated retinal exam ≥11 months earlier were imaged according to JVN protocol. Patients then had dilated retinal examination and ETDRS photos. Levels of DR determined from JVN images, clinical exam, and ETDRS photos were compared.

Two (1.9%) eyes had JVN images ungradable for level of DR. In the 102 (98.1%) gradable eyes, JVN diagnosis exactly matched clinical examination for level of DR in 82 (77.9%) eyes and was at or within 1 level of DR in all eyes (100%). Three eyes (2.9%) had JVN images ungradable for DME and 1 of these eyes had DME by clinical examination. JVN diagnosis matched clinical examination for DME in all eyes (101) gradable by JVN. In a survey following all imaging sessions, 96.1% of patients reported JVN imaging improved their understanding of eye disease, 100% were satisfied with JVN, and 92.3% would consider replacing dilated examination by their eye doctor with JVN imaging. JVN digital imaging closely matched clinical examination for level of DR and DME, would have resulted in no patients receiving less stringent follow-up, and was well accepted by patients.

JVN vs Retinal Specialist: Nondiabetic-Related Eye Disorders

Because inability to identify significant nondiabetic-related eye disease would pose a serious shortcoming to any DR telemedicine program, a study was conducted to evaluate the ability of the JVN to detect ocular pathology other than DR in patients with DM compared with dilated retinal examination by retinal specialist ophthalmologists.²⁵ A retrospective chart review of 280 Joslin Diabetes Center outpatients (560 eyes) with type 1 ortype 2 DM who had both JVN imaging and clinical comprehensive eye examination by retinal specialists at the Beetham Eye Institute of the Joslin Diabetes Center was conducted. Nonsimultaneous stereoscopic nonmydriatic digital retinal images (640 × 480 pixels) of three 45° retinal fields were acquired and graded for clinical level of DR and other ocular pathology by certified readers according to JVN protocol. Findings from JVN digital images were compared with findings from dilated retinal examination by retinal specialists performed within an average of 39.6 days of digital imaging. An independent senior retinal specialist adjudicated disagreements by review of JVN images and ETDRS photos when available.

Nonmydriatic digital evaluation identified at least1 non-DR ocular finding in 40.7% of patients (114/280), including cataract (n=100), age-related maculopathy (n=52), suspicion of glaucoma (n=18), choroidal lesions (n=18), evidence of systemic disorder (eg, hypertension or renal disease, n=15), epiretinal membrane (n=11), chorioretinal atrophy and/or scar (n=6), retinal emboli (n=3), retinitis pigmentosa (n=1), and asteroid hyalosis (n=1). Agreement of nonmydriatic imaging with clinical examination for presence and absence of these findings was 95.4%, 91.3%, 98.2%, 98.6%, 98.2%, 99.6%, 100%, 100%, 100%, and 100%, respectively. Kappa values for all non-DR lesions demonstrated near perfect agreement (κ≥.80) except for age-related maculopathy (κ=.71) and choroidal lesions (κ=.73), where agreement was substantial. Overall, JVN nonmydriatic digital imaging demonstrated excellent agreement with dilated ophthalmic examination by retinal specialists in the detection of ocular disease other than DR (Figure 4).

JVN IN CLINICAL ENVIRONMENTS

JVN in the Indian Health Service

JVN has been successfully deployed in more than 50 clinical sites within the Indian Health Service, and in each case, the JVN has resulted in increased access to diabetes eye care. In the Indian Health Service, the impact of JVN retinal imaging technology was measured on the rate of surveillance and treatment of DR in this large, well-defined patient population over a 5-year period.²⁶ A computerized patient information system described, on an annual basis, the patient population, the number of patients with DM, and the proportion of the diabetic patients who received appropriate medical services as measured against standards of care before and after implementation of the JVN digital retinal imaging system in a primary care setting. A procedure log ascertained the proportion of patients who received laser treatments for DR per year.

The rate of annual retinal exams increased from 50% (95% CI 44%-56%) to 75% (95% CI 70%-80%)(chi²=58.5, trend P=.000000) representing a 50% increase in the retinal examination rate. The laser therapy rate rose from 19.6/1000 people with DM in 1999 to 29.5/1000 people in 2003 for a 51% increase in the laser treatment rate. Implementation of retinal imaging technology in a primary care setting led to a significant increase in the rate of DR surveillance and a proportional increase in the rate of laser treatment for DR for a large patient population.

JVN in the Veterans Administration

In another study, the effect of the JVN was evaluated in a retrospective observational cohort study. Veterans at the Togus Veterans Health Administration outpatient clinic with DM, impaired fasting glucose, or impaired glucose tolerance were imaged according to JVN protocol prior to scheduled nonophthalmic appointments or based on ophthalmic history.²⁷ Images were transmitted to the Beetham Eye Institute of the Joslin Diabetes Center for grading and recommended treatment plan. Patients with significant diabetic eye disease, with nondiabetic ocular findings, or who were overdue for ocular examination were referred for ophthalmic care.

A total of 1219 patients (2437 eyes) were imaged; of these, 1536 eyes (63.0%) had no DR, 389 (16.0%) mild nonproliferative DR (NPDR), 105 (4.3%) moderate NPDR, 35 (1.4%) severe NPDR, 20 (0.8%) very severe NPDR, and 21 (0.9%) proliferative DR (PDR). Additionally, 1907 eyes (78.3%) had no DME, 34 (1.4%) early DME, and 16 (0.7%) clinically significant macular edema (CSME). 354 patients (29.0%) had either no DR or mild NPDR in both eyes, no evidence of DME, and no significant nondiabetic findings. 679 patients (55.7%) had no DR in either eye and 229 (18.8%) had mild NPDR in the more severe eye. Of the 908 patients (74.5%) with either no DR or mild NPDR in the more severe eye, 533 (58.7%) had at least 1 nondiabetic ocular finding necessitating referral. 320 eyes (13.1%) were ungradable for level of both DR and DME and 160 (6.6%) were ungradable for DME alone. In a nonophthalmic setting, JVN identified severity of DR permitting appropriate triage for eye care, and detected nondiabetic ocular conditions suggesting ophthalmic evaluation in a large percentage of patients, even in the absence of significant DR. Subsequently, the US Veterans Administration initiated their National Teleretinal Program, which is based on the JVN model.

Additional studies have reinforced the sensitivity and specificity of the JVN nonmydriatic diabetes eyecare program in identifying diabetic eye disease²⁸ and have demonstrated cost-effectiveness of the JVN telemedicine system in the US Department of Defense, US Department of Veterans Affairs, and the US Indian Health Service.²⁹

FUTURE OF TELEMEDICINE

Telemedicine in general, and telemedicine for DM and DR in particular, are likely to play an increasing role in the delivery of health care. Telemedicine has the potential to expand the delivery of care, establish a widespread standard of care that is based on validated clinical standards, and allow patients to participate more fully in their health care. For ocular telemedicine, advancing technologies are likely to allow automated grading of DR, with computer-assisted identification of lesions of DR and other retinal disorders. Integrating diabetic eye care into a comprehensive diabetes care program will allow eye care that considers and addresses risk factors for onset and progression of DR and allows a continuum of care that reduces the risks of complications from DM. The result will be a continuum-of-care model that fosters productive interactions between patients who take an active part in their care and a variety of providers backed up by resources and expertise. Thus, the patient is an informed, proactive self-manager using telemedicine technology as a tool to keep DM in control and provide ongoing education and care to minimize complications, symptoms, and disability. The core of such programs is the ability to empower the patient.

Figure 4. Overall agreement for specific non-diabetic retinopathy findings as determined by clinical dilated fundus exam, Joslin Vision Network evaluation.

A comprehensive disease management program (CDMP) development effort is under way involving leaders in diabetes clinical management, education, lifestyle modification, and medical informatics from the Joslin Diabetes Center, the Department of Defense, the Veterans Affairs, and the Indian Health Service. The rationale for this effort was the recognized need to provide a continuum of care for diabetic patients. This need is underscored by results from the Diabetes Prevention Program (DPP). In the DPP, patients were randomized to either intensive lifestyle modification, metformin, or placebo treatment. After follow-up of 4.6 years, lifestyle modification reduced the progression to DM by 58%, reinforcing the value of case-management programs. The CDMP is a case-management program that provides continuous and immediate contact between patients, care managers and physicians over secure Web sites. It is anticipated that the development of the interactive Web-based education and behavior modules will provide the largest potential benefit with respect to motivating patients to set reasonable goals for their management of DM and thus maximize the clinical benefit.

Philosophically, this management program has been developed to facilitate an interactive and continuous connection between patients and their care teams. The program is able to aggregate clinical data from diverse sources, such as electronic medical record systems, lab systems, and data from the home, through the use of physiologic home-monitoring devices. In this way, the system is able to present data to a physician in a medically relevant manner that allows patient-doctor communications to be optimized over what is generally a very short patient visit. The robust clinical decision support system also rapidly identifies patients at risk or who have other medical issues that need to be addressed. It is expected that the management and healthcare delivery services provided through this application will allow a primary-care practitioner to appropriately manage patients with chronic diseases such as diabetes for longer periods of time before having to refer patients to more expensive subspecialty services.

SUMMARY

Diabetes and its complications, particularly DR, provide an ideal opportunity for care with telemedicine. The JVN is a telemedicine diabetes eyecare program that allows accurate assessment of level of DR, is sensitive in identifying nondiabetic related eye disease, and has the potential to be an integral part of a comprehensive diabetes management program. RP

Acknowledgement: Special thanks to Cheri C. Dunn, OD, DO, MPH for reviewing this manuscript.

REFERENCES

1. American Diabetes Association. Standards of medical care in diabetes—2007. Diabetes Care 2007;30(Suppl. 1):S4-S41.
2. National Diabetes Data Group. Diabetes in America, 2nd ed. Washington, DC: US Government Printing Office; 1995. Report No.: NIH no. 95-1468.
3. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL. The Wisconsin Epidemiologic Study of Diabetic Retinopathy. VI. Retinal Photocoagulation. Ophthalmology. 1987;94:747-53.
4. Brechner RJ, Cowie CC, Howie LJ, Herman WH, Will JC, Harris MI. Ophthalmic examination among adults with diagnosed diabetes mellitus. JAMA. 1993; 279:1714-1719.
5. Will JC, German RR, Schurman E, Michael S, Kurth DM, Deeb L. Patient adherence to guidelines for diabetes eye care: results from the Diabetic Eye Disease Follow-up Study. Am J Pub Health. 1994;84:1669-1671.
6. Wang F, Javitt JC. Eye care for elderly Americans with diabetes mellitus. Ophthalmology. 1996;103:1744-1750.
7. American Telemedicine Association Web site. http://www.atmeda.org/news/definition.html. Accessed September 17, 2007.
8. King H, Aubert R, Herman W. Global burden of diabetes, 1995-2025: prevalence, numerical estimates and projections. Diabetes Care. 1998;21:1414-1431.
9. World Health Organization – Regional Office for Americas: Diabetes in selected study populations. January 2003. World Health Organzation Web site. http://www.who.int/ncd/dia/databases4.htm. Accessed September 17, 2007.
10. Aiello LP, Gardner TW, King GL, et al. Diabetic retinopathy: technical review. Diabetes Care. 1998;21:143-156.
11. Klein R, Klein BEK. Vision disorders in diabetes. In: Harris MI, Cowie CC, Reiber G, Boyko E, Stern M, Bennett P, eds. Diabetes in America. 2nd ed. Washington, DC: U S Government Printing Office; 1995:293-338.
12. Diabetic Retinopathy Study Report Number 2: Photocoagulation of proliferative diabetic retinopathy. Ophthalmology. 1978;85:82-106.
13. Early Treatment Diabetic Retinopathy Study Research Group. Photocoagulation for diabetic macular edema; Early Treatment Diabetic Retinopathy Study report number 1. Arch Ophthalmol. 1985;103:1796-1806.
14. Early Treatment Diabetic Retinopathy Study Research Group. Early Treatment Diabetic Retinopathy Study Report Number 9: Early photocoagulation for diabetic retinopathy. Ophthalmology. 1991;98:766-785.
15. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977-986.
16. United Kingdom Prospective Diabetes Study Group. Effect of intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes: UKPDS 33. Lancet. 1998;352:837-853.
17. Ferris FL III. How effective are treatments for diabetic retinopathy? JAMA. 2003;269:1290-1291.
18. Telehealth practice recommendations for diabetic retinopathy. ATA Web site. http://www.atmeda.org/ICOT/diabeticretionpathy.FINAL.pdf. Accessed September 17, 2007.
19. Cavallerano J, Lawrence MG, Zimmer-Galler I, et al; American Telemedicine Association; Ocular Telehealth Special Interest Group; National Institute of Standards and Technology Working Group. Telehealth Practice Recommendations for Diabetic Retinopathy. Telemed J and e-Health. 2004;10:469-482.
20. 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.
21. Aiello LM, Cavallerano J, Cavallerano A, Bursell S-E. Preserving human vision: The Joslin Vision Network (JVN) innovative telemedicine care for diabetes. Ophthalmol Clin N Am. 2000;13:213-224.
22. Bursell SE, Cavallerano JD, Cavallerano AA, et al. Stereo nonmydriatic digital-video color retinal imaging compared to ETDRS 7-standard field 35-mm stereo color photos to diagnose level of diabetic retinopathy. Ophthalmology. 2001;108:572-585.
23. Cavallerano AA, Cavallerano JD, Katalinic P, Tolson AM, Aiello LP, Aiello LM. Use of Joslin Vision Network Digital-Video Nonmydriatic Retinal Imaging to assess diabetic retinopathy in a clinical program. Retina. 2003;23:215-223.
24. Cavallerano JD, Aiello LP, Cavallerano AA, et al; Joslin Vision Network Research Team. Nonmydriatic digital retinal imaging alternative for annual retinal examination in persons with previously documented no or mild diabetic retinopathy. Am J Ophthalmol. 2005;140:667-673.
25. Chow S-P, Aiello LM, Cavallerano JD, et al. Comparison of nonmydriatic digital retinal imaging vs dilated ophthalmic examination for nondiabetic eye disease in persons with diabetes. Ophthalmology. 2006;113:833-840.
26. Wilson C, Horton M, Cavallerano J, Aiello LM. Addition of primary-care based retinal imaging technology to an existing eye care professional referral program increased the rate of surveillance and treatment of diabetic retinopathy. Diabetes Care. 2005;28:318-322.
27. Cavallerano AA, Cavallerano JD, Katalinic P, et al. A telemedicine program for diabetic retinopathy in a Veterans Affairs Medical Center—the Joslin Vision Network Eye Health Care Model. Am J Ophthalmol. 2005:139;597-604.
28. Ahmed J, Ward TP, Bursell SE, Aiello LM, Cavallerano JD. The sensitivity and specificity of nonmydriatic digital stereoscopic retinal imaging in detecting diabetic retinopathy. Diabetes Care. 2006;29:2205-2209.
29. Whited JD, Datta SK, Aiello LM, et al. A modeled economic analysis of a digital teleophthalmology system as used by three federal healthcare agencies for detecting proliferative diabetic retinopathy. Telemed and e-Health. 2005;11:641-651.