The sensitivity and specificity of single-field nonmydriatic monochromatic digital fundus photography with remote image interpretation for diabetic retinopathy screening: a comparison with ophthalmoscopy and standardized mydriatic color photography11InternetAdvance publication at ajo.com. April 12, 2002.

To determine, from the health insurer's perspective, the cost of preventing vision loss in patients with diabetes mellitus through ophthalmologic screening and treatment and to calculate the cost-effectiveness of these interventions as compared with that of other medical interventions. Computer modeling, incorporating data from population-based epidemiologic studies and multicenter clinical trials. Monte Carlo simulation was used, combined with sensitivity analysis and present value analysis of cost savings. Screening and treatment of eye disease in patients with diabetes mellitus costs $3190 per quality-adjusted life-year (QALY) saved. This average cost is a weighted average (based on prevalence disease) of the cost-effectiveness of detecting and treating diabetic eye disease in those with insulin-dependent diabetes mellitus ($1996 per QALY), those with non-insulin-dependent diabetes mellitus (NIDDM) who use insulin for glycemic control ($2933 per QALY), and those with NIDDM who do not use insulin for glycemic control ($3530 per QALY). Our analysis indicates that prevention programs aimed at improving eye care for diabetic persons not only result in substantial federal budgetary savings but are highly cost-effective health investments for society. Ophthalmologic screening for diabetic persons is more cost-effective than many routinely provided health interventions. Because diabetic eye disease is the leading cause of new cases of blindness among working-age Americans, these results support the widespread use of screening and treatment for diabetic eye disease.

To determine if neural networks can detect diabetic features in fundus images and compare the network against an ophthalmologist screening a set of fundus images. 147 diabetic and 32 normal images were captured from a fundus camera, stored on computer, and analysed using a back propagation neural network. The network was trained to recognise features in the retinal image. The effects of digital filtering techniques and different network variables were assessed. 200 diabetic and 101 normal images were then randomised and used to evaluate the network's performance for the detection of diabetic retinopathy against an ophthalmologist. Detection rates for the recognition of vessels, exudates, and haemorrhages were 91.7%, 93.1%, and 73.8% respectively. When compared with the results of the ophthalmologist, the network achieved a sensitivity of 88.4% and a specificity of 83.5% for the detection of diabetic retinopathy. Detection of vessels, exudates, and haemorrhages was possible, with success rates dependent upon preprocessing and the number of images used in training. When compared with the ophthalmologist, the network achieved good accuracy for the detection of diabetic retinopathy. The system could be used as an aid to the screening of diabetic patients for retinopathy.

Diabetic retinopathy, which leads to macular edema and retinal neovascularization, is the leading cause of blindness among working-age Americans. Previous research has demonstrated significant cost savings associated with detection of eye disease in Americans with type I diabetes. However, detection and treatment of eye disease among those with type II diabetes was previously thought not to be cost-saving. Our purpose was to estimate the current and potential federal savings resulting from the screening and treatment of retinopathy in patients with type II diabetes, based on recently available data concerning efficacy of treating both macular edema and neovascularization along with new data on federal budgetary costs of blindness. We used computer modeling, incorporating data from population-based epidemiological studies and multicenter clinical trials. Monte Carlo simulation was used, combined with sensitivity analysis and present value analysis of cost savings. Screening and treatment for eye disease in patients with type II diabetes generates annual savings of $247.9 million to the federal budget and 53,986 person-years of sight, even at current suboptimal (60%) levels of care. If all patients with type II diabetes receive recommended care, the predicted net savings (discounted at 5%) exceeds $472.1 million and 94,304 person-years of sight. Nearly all savings are associated with detection and treatment of diabetic macular edema. Enrolling each additional person with type II diabetes into currently recommended ophthalmological care results in an average net savings of $975/person, even if all costs of care are borne by the federal government. Our analysis indicates that prevention programs aimed at improving eye care for patients with diabetes not only reduce needless vision loss but also will provide a financial return on the investment of public funds.