Purchase this article with an account.
Loic Bazin, Sabine Derrien, MARTIANO David, Lamard Mathieu, Beatrice Cochener; Diabetic retinopathy : A descriptive study of a method of investigation: BOA’s project. Invest. Ophthalmol. Vis. Sci. 2014;55(13):2281. doi: https://doi.org/.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The purpose of this study is to report the results of 208 patients observation comparing two different interfaces of delayed reading retinophotographies and thus two different logistics managements of diabetic patients at the University Hospital of Brest.
Prospective study of 208 diabetic patients followed in the endocrinology department of Brest from January to April 2013. The retinophotographies were automatically and randomly assigned to one of two consoles for delayed reading by an ophthalmologist of the CHU Brest. The « examination time » and « writing report time » were identified. The settings « diagnosis », « image quality » and «follow-up » were graduated.
208 patients were included. A high risk diabetic retinopathy was diagnosed in 22.6% of patients. The examination time did not differ between the two consoles (Student t test: t = 0.623, p = 0.534). « Writing report time » differ a statistically significant manner (p << 10-4). There is a positive correlation between « examination time » and « writing report time » (r = 0.382, p <10-4) for both consoles. There are no statistically significant differences in the « image quality » setting between the two systems (p = 0.623). There was no statistically significant difference between the two consoles on the variables « follow-up » (p = 0.067). 18.3% of the retinophotographies are not interpretable and 10.6% of the patients were referred to an ophthalmologist for a retinal diagnosis other than that of diabetic retinopathy.
The BOA project that focused on the comparison of two consoles has shown an improvement in the compendium and interpretation of data with an optimized console. The new studied interface has yet to be apprehended by the ophthalmologist and the study of a larger cohort should be able to reveal significant differences. Reading retinophotographies in a thought and ergonomic interface allows better reproducibility and diagnostic reliability. The automated detection and diagnosis assistance provided to the ophthalmologist in the monitoring and the management of diabetic retinopathy are major assets in the development of our specialty.
This PDF is available to Subscribers Only