gms | German Medical Science

102. Jahrestagung der DOG

Deutsche Ophthalmologische Gesellschaft e. V.

23. bis 26.09.2004, Berlin

Telemedical evaluation of retinal images (TalkingEyes)

Meeting Abstract

  • corresponding author G. Michelson - University Erlangen-Nurnberg, Dept. Ophthalmology
  • P. Kolominsky-Rabas - University Erlangen-Nurnberg, Dept. Neurology
  • M. Groh - Private Center of Preventive Medicine, Erlangen
  • Y. Dhein - SBK Siemens Betriebskrankenkasse, München
  • W. Striebel - Siemens AG, Erlangen
  • R. Horstmann - Private Center of Preventive Medicine, Erlangen

Evidenzbasierte Medizin - Anspruch und Wirklichkeit. 102. Jahrestagung der Deutschen Ophthalmologischen Gesellschaft. Berlin, 23.-26.09.2004. Düsseldorf, Köln: German Medical Science; 2004. Doc04dogFR.12.06

The electronic version of this article is the complete one and can be found online at:

Published: September 22, 2004

© 2004 Michelson et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License ( You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.




To screen retinal vessels in respect to microvascular abnormalities in a multicentric telemedical setting.


By trained technicians the retinal vessels from both eyes were registered by a non-mydriatic fundus camera (45°, Kowa). A standardised medical history was performed and completed. The images and the data were electronically sent by MedStage (Siemens AG) to a web-server. By trained technicians the arterio-venous ratio was calculated using the Parr-Hubbard-formula. Ophthalmologists evaluated the retinal fundus image using a standardised web-based document. The result of the evaluation became available for the examined subject by a web-based report. From 1.9.2001 to 1.10.2003 in the ongoing prospective study 27.234 subjects were examined and evaluated. The mean age was 48±8 years. The subjects reported in 27% arterial hypertension, in 26% hyperlipidemia, in 4% coronary heart disease, and in 2.5% diabetes. The arterial and diastolic blood pressure was in average 129±14mmHg,and 82±9mmHg, respectively.


The intraobserver- and interobserver- reproducibility (alpha-Kronbach-coefficient) of the measurement of the arterio-venous ratio was 0,97, and 0,798, respectively. The quality of images was good. In respect to the measurement of the arterio-venous ratio 4.3% of the images were unable to process, 15,3%, and 80.5% of the images showed moderate, and good quality, respectively measuring the arterio-venous ratio. In respect to the ophthalmological evaluation in 1,1% of the images an evaluation was not possible due to bad quality, 10,2%, and 88,7% of the images showed moderate, and good quality, respectively.


A multicenter approach of screening of the retina with a central evaluation by ophthalmologists in a reading center allows screenings of more than 100 persons per day with good quality and good reproducibility.