Fundus autofluorescence and colour fundus imaging compared during telemedicine screening in patients with diabetes

Anton M. Kolomeyer, Benjamin R. Baumrind, Bernard C. Szirth, Khadija Shahid, Albert S. Khouri

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


We investigated the use of fundus autofluorescence (FAF) imaging in screening the eyes of patients with diabetes. Images were obtained from 50 patients with type 2 diabetes undergoing telemedicine screening with colour fundus imaging. The colour and FAF images were obtained with a 15.1 megapixel non-mydriatic retinal camera. Colour and FAF images were compared for pathology seen in nonproliferative and proliferative diabetic retinopathy (NPDR and PDR, respectively). A qualitative assessment was made of the ease of detecting early retinopathy changes and the extent of existing retinopathy. The mean age of the patients was 47 years, most were male (82%) and most were African American (68%). Their mean visual acuity was 20/45 and their mean intraocular pressure was 14.3 mm Hg. Thirty-eight eyes (76%) did not show any diabetic retinopathy changes on colour or FAF imaging. Seven patients (14%) met the criteria for NPDR and five (10%) for severe NPDR or PDR. The most common findings were microaneurysms, hard exudates and intra-retinal haemorrhages (IRH) (n = 6 for each). IRH, microaneurysms and chorioretinal scars were more easily visible on FAF images. Hard exudates, pre-retinal haemorrhage and fibrosis, macular oedema and Hollenhorst plaque were easier to identify on colour photographs. The value of FAF imaging as a complementary technique to colour fundus imaging in detecting diabetic retinopathy during ocular screening warrants further investigation.

Original languageEnglish (US)
Pages (from-to)209-212
Number of pages4
JournalJournal of Telemedicine and Telecare
Issue number4
StatePublished - 2013

All Science Journal Classification (ASJC) codes

  • Health Informatics


Dive into the research topics of 'Fundus autofluorescence and colour fundus imaging compared during telemedicine screening in patients with diabetes'. Together they form a unique fingerprint.

Cite this