Applying “Informatics” Technology to Retinal Practice

Applying “Informatics” Technology to Retinal Practice


If you are unfamiliar with the word “informatics,” don't feel that you are somehow behind the curve when it comes to keeping up with advances in information technology. Informatics is essentially an umbrella term that encompasses a range of tools that can be used to create more efficient operation of your retina practice.

Informatics includes the science of information, the art of info processing, and the management of info systems. It studies the structure and interactions of systems that store, process, and communicate information, and has its own conceptual and theoretical basis.1

Over time, several informatics developments have had significant impacts on retinal practice, including electronic health records (EHR), digital imaging, telemedicine, computer-aided screening, and decision-support systems. But the implementation of informatics technology in retinal practice has not been broadly covered in medical literature. This article will trace important developments affecting retinal practice and explain the benefits and barriers in applying informatics technology in ophthalmology.

Many concepts discussed are probably already familiar to you but, taken in total, the technologies presented here represent the spectrum of informational tools that are now available — or that soon will be available — to retina practices seeking to be on the leading edge of progress.


EHR systems organize the storage and retrieval of digitized records. Over time, EHR systems should enable paperless retinal clinics and make the clinic workflow fully automatic and efficient. For example, a case study of an EHR system start-up for Retina Consultants (Phoenix, AZ) in 2002 revealed a return on investment within eight months of implementation. After three years, the practice claimed a 95% decrease in staff overtime, 20% decrease in staff salary, 40% decrease in office supplies, 100% decrease in transcription costs, 30% decrease in total data processing, and 20% decrease in staff concurrent with adding another provider.2

It's no secret that ophthalmology is an image-based specialty, and retinal diagnoses are mainly based on imaging. A major issue in retinal practice is image archiving and digital data capture from diagnostic devices to the EHR. The EHR should be connected to existing imaging (fundus camera, slit-lamp, OCT, and Heidelberg retinal tomography) and diagnostic devices (visual field machine and various tonometers). Other measurements such as visual acuity are important for diabetic retinopathy diagnosis and should also be fed into the EHR, either automatically from computer-aided visual acuity measurement programs or manually. The specialist should be able to access all info at a touch of a button from the consultation room (Figure 1).

Kambiz Bahaadini, MD, is a physician and PhD fellow at the Lions Eye Institute in Perth, Australia. Kanagasingam Yogesan, PhD, is a Medical Informatics specialist and professor in e-Medicine at Lions Eye Institute and University of Western Australia. Dr. Bahaadini reports no financial interest in any products mentioned in this article. Professor Yogesan reports significant financial interest in the eye imaging device and information kiosks mentioned here. Dr. Bahaadini can be reached via e-mail at


The retina is the only part of the body where blood vessels can be directly visualized noninvasively. Advanced technological developments and fast computer processing, which led to the development of digital imaging systems over the past 20 years, have revolutionized fundus imaging.3 As digital imaging and computing power rapidly expand, the potential use in ophthalmology increases.

Digital imaging systems use a digital fundus camera to obtain standard field color and/or monochromatic images of the retina. While this type of retinopathy screening is suggested as an alternative to the conventional dilated fundus examination, it is not an alternative to a comprehensive ophthalmologic examination.4 With the advent of high-resolution digital photography, retinal imaging has become the new gold standard for retinal diagnosis. These digital images, including videos and stereoimages, can be transferred automatically to the EHR for storage and archiving.


Image processing, analysis, and computer vision techniques are increasing in prominence in all fields of medicine. They are especially relevant to ophthalmology, which is profoundly reliant on visually oriented signs. Automatic image analysis and grading for the diagnosis of diabetic retinopathy and glaucoma benefit mass screening by health professionals.3 Given the diversity and complexity of eye functions, a large number of equipment, automatic methods, and algorithms for diagnosis have been developed.

Although a doctor can often identify certain diseases after a visual analysis of an image representing the affected area, in some cases the diagnosis is difficult due to lack of experience, fatigue, poor image quality, etc. In such situations, a second opinion from another expert or, preferably, a computer-aided diagnosis (CADx) system would be very useful.3 In particular, CADx systems reduce the level of uncertainty regarding similar diseases, improve the primary and evolutional detection of disease, and allow monitoring of the health status of a patient during treatment.5

Computer-aided Detection vs Diagnosis. While computer-aided detection methods can only detect the location of a possible disease, CADx systems provide area analysis, diagnostics, and disease recognition with high precision.6Computer-aided detection systems are characterized by sensitivity: the percent of images correctly detected as potentially being pathogenic. In contrast, CADx systems are characterized by specificity: the percent of images correctly recognized as diseased. Image classification recognition methods and techniques often use linear or higher order classifiers,7 decision trees, machine learning,8 or neural networks5,9 for their operation.

Support Systems for Diabetic Retinopathy. There has recently been an increase in the use of digital image processing to screen and grade diabetic retinopathy, including the detection of micro-aneurysms, hemorrhages, retinal exudates, cotton wool spots, and clinically significant macular edema.3 In terms of automatic screening for early-stage (nonproliferative) diabetic retinopathy, Reza et al. developed a system that could detect bright lesions with an average accuracy of ~97%.10

Figure 1. An ophthalmology-specific EHR system.


Support Systems for Glaucoma. Research by Wrobleski et al. demonstrated that the automatic classification of perimetry data may be useful for glaucoma screening, staging, and follow-up.11 The optic cup-to-disc ratio, a key indicator of glaucoma, is determined manually, limiting its use in mass screening.

However, recently computer advancement has come to help physicians in their automatic estimation of this ratio.12 The cup-to-disc ratio, intraocular pressure, visual field measurement, disc asymmetry, family history, and age could be combined to create a computer-aided decision support system for the diagnosis of glaucoma.

A single clinical parameter is not adequate to diagnose glaucoma with accuracy during screenings.


Screening examinations for diabetic retinopathy and other retinal diseases require that a patient make regular visits to an eye specialist. Many individuals in rural or remote areas do not have easy access to an ophthalmologist. If patients could obtain fundus exams through their primary care physician, then diseases such as diabetic retinopathy, AMD, and even retinal tears could be diagnosed and treated.

For example, using a digital retinal imaging device, primary care doctors can take dilated fundus photographs and send the images via the Internet to a centralized reading center. The images are examined and interpreted by a retinal specialist, then sent back to the primary care physician for management. Patients who need treatment and surgery can be referred to an ophthalmologist.13

An example of a successful teleophthalmology program links optometrists to retinal specialists in Canada. This program reduced office visits to retinal specialists by 48% while improving the efficiency of examination, testing, and treatment. Patients benefited from reduced travel time and distance.14 The LRMC remote teleretinology and teleophthalmology project is another example of the successful implementation of telemedicine in retinal practice.6 The Internet can also facilitate communication between ophthalmologists and patients and among eyecare professionals for better delivery of health and advice.15 Indeed, Internet-based telemedicine is in greater need for retinal disease than other eye problems; in Western Australia, only 3% of the patients use teleophthalmology for emergency consultation, while 94% use it for glaucoma and diabetic retinopathy testing.16

Recently developed low-cost retinal imagers can capture high-quality images of the retina through a nondilated pupil. These imagers are well suited for ophthalmic telemedicine. Additional developments such as three-dimensional viewing and image-enhancement software facilitate interpretation and provide objective measurements to aid in diagnosis. While digital retinal imaging is unlikely to replace face-to-face examination with an ophthalmologist, digital-guided ophthalmic telemedicine promises to play a prominent role in patient care for years to come.13


In addition to the above benefits of informatics technology in retinal practice, such technologies can also be applied in clinics to provide retinal education. Eye information kiosks can be installed in consultation rooms to educate patients about their disease and improve clinical outcomes. Providing general information on eye anatomy, diseases, and treatment options, the tactile screens and audio support of these kiosks make them ideal for access by children, people with low vision, and the elderly. These kiosks can also be used as an education tool for nurses, allied health care professionals, and others who screen for eye disease.17


There are several important reasons why the implementation of recent information technological advances has been slow in retinal clinics.15

Cost. Changing from the current paper-based system to digital EHR is costly, as all devices must be upgraded to provide digital output. Although it may be cheaper to buy the latest devices rather than to upgrade existing ones (eg, photographs), the cost of diagnostic devices is prohibitively expensive. Other costs include capital costs of digital equipment, storage media, servers, recurrent maintenance costs, upgrades, and overhead costs for training support.

• Mass Storage. One of the major issues with digital images is storage and archiving. Around 5 to 20 MB of storage space is required per image, and at least four images (20-80 MB) need to be stored for each patient, amounting to massive amounts of data to be stored and archived (on the order of terabytes per year). Backup storage is also required. Although the cost involved in the maintenance and storage hardware is considerable, the benefits of image-based EHR are enormous.

• Interoperability. Each diagnostic device vendor has its own database and image capture software, making it very difficult to communicate all systems to enable one single EHR for the whole clinic. Message exchange standards such as HL7 can be used to overcome this problem.

Workflow Disruption. The development and implementation of new systems requires significant effort and commitment from the entire institution, and the deployment of new online technologies in individual departments can lead to unpredictable disruptions in workplace practices.

Data Security. An important limiting factor has been concerns regarding patient confidentiality, and the ownership of and accountability for patient information. Data integrity and security are critical issues that need to be addressed by both local and national regulatory bodies.15


Informatics technology has the potential to significantly improve retinal practice, making it more efficient with the paperless option. Its benefits overweigh any barriers. The era of a fully integrated, interoperable and intelligent health informatics technology system is no longer science fiction, but an emerging reality. We should expect to see fully automated retinal clinics by the year 2020. RP


  1. What is the difference between Informatics and Information Technology? Web site. between_lnformatics_and_lnformation_Technology. Accessed August 26, 2009.
  2. Associated Retina Consultants Implements EMR. Chart Logic Web site. Accessed August 26, 2009.
  3. Patton N, Aslam TM, MacGillivray T, et al. Retinal image analysis: concepts, applications and potential. Prog Retin Eye Res. 2006;25:99-127.
  4. Digital retinal Imaging systems for the screening and detection of diabetic retinopathy. Wellmark Blue Cross Blue Shield Web site, e_business/provlder/medical_pollcies/policles/_Digltal_Retlnal_lmaging.htm Acccessed August 26, 2009.
  5. Luculescu MC, Lache S. Computer-aided diagnosis system for retinal diseases In medical imaging. WSEAS Trans Systems. 2008;7:264-276.
  6. Erickson BJ, Bartholmai B. Computer-aided detection and diagnosis at the start of the third millennium. J Digit Imaging. 2002;15:59-68.
  7. Burroni M, Corona R, Dell'Eva G, et al. Melanoma computer-aided diagnosis: reliability and feasibility study. Clin Cancer Res. 2004;10:1881-1886.
  8. Niemeijer M, van Ginneken B, Russell SR, Suttorp-Schulten MS, Abràmoff MD. Automated detection and differentiation of drusen, exudates, and cotton-wool spots In digital color fundus photographs for diabetic retinopathy diagnosis. Invest Ophthalmol Vis Sci. 2007;48:2260-2267.
  9. Osman MK, Mashor MY, Arshad MR. 3D object recognition using mutiple views, affine moments invariants and multilayered perceptron network. WSEAS Trans Systems. 2004;3:3065-3069.
  10. Reza AW, Eswaran C. A decision support system for automatic screening of nonproliferative diabetic retinopathy. J Med System. 2009 Jul 4. [Epub ahead of print]
  11. Wroblewski D, Francis BA, Chopra V, et al. Glaucoma detection and evaluation through pattern recognition In standard automated perimetry data. GraefesArch Clin Exp Ophthalmol. 2009 Jul 5. [Epub ahead of print]
  12. Wong DK, Liu J, Lim JH, et al. Level-set based automatic cup-to-disc ratio determination using retinal fundus images in ARGALI. Conf Proc IEEE Eng Med Biol Soc. 2008;2008:2266-2269.
  13. Kalayoglu MV. Digital retinal Imaging and the new era of ophthalmic telemedicine. Ophthalmology Web Web site, Spotlight.aspx?spid=23&aid=262. Accessed August 26, 2009.
  14. Hanson C, Tennant MT, Rudnlsky CJ. Optometrie referrals to retina specialists: evaluation and triage via teleophthalmology. Telemed J E Health. 2008: 14:441-445.
  15. Kumar S, Yogesan K. Internet-based eye care: VISION 2020. Lancet. 2005: 366:1244-1245.
  16. Kumar S, Tay-Kearney ML, Constable IJ, Yogesan K. Internet based ophthalmology service: Impact assessment. Br J Ophthalmol. 2005;89:1382-1383.
  17. Lions Eye Institute annual report 2007. Lions Eye Institute Web site. Accessed August 26, 2009.
  18. Scanlon P. Digital retinal photography in diabetic eye screening. Diabetes Technol Ther. 2001;3:187-191.