Abstract
Abstract: :
Purpose: Quality fundus photography plays an essential role in phenotyping for ophthalmic genetic studies. In cases where patients are unwilling or unable to travel to hospital, examination in the community becomes a necessity. The aim of this study is to compare the accuracy of diagnoses made on the same image taken with the Nidek NM200D against the Topcon TRC 50–EX Methods: Images were taken from each individual patient at the same sitting on both systems. The images were stored in TIFF format on CDs. These were graded independently in a masked fashion by an ophthalmologist. The images were graded in terms of the most likely diagnosis and for the identification of specific lesions such as drusen, haemorrhages, exudates, pigment, fibrosis and neovascularization. A subset of the images which had a diagnoses of age–related macular degeneration (AMD), were also graded independently at the Belfast Grading Centre for Fundus Photography to investigate if subtle grading differences were detectable using the NM200D system. Results: Images were acquired from 40 consecutive patients who were willing to be photographed on both systems The photographs from each system were reviewed in random order from each system by the grader and a diagnoses was made. Twenty–two patients were female and 18 were male. On chart review 12 patients had diabetic retinopathy, 15 had AMD, 5 were normal and 8 had miscellaneous diagnoses. The grader was not in possession of any clinical data at the time of grading. The results show 87.5% agreement in terms of diagnosis and 82.2% agreement on the identification of specific lesions. Detailed AMD grading is currently in progress. Conclusions:We present data which compares the quality of images from the standard high quality TOPCON digital system with the portable NM200D camera. The results indicate that although the quality of images from the portable camera was less – the images from the NM200D are of sufficient quality to allow a high level of diagnostic accuracy and also a good level of agreement on the presence or absence of disease. Further analysis of the AMD images will indicate whether the quality is sufficient to allow levels of grading of AMD.
Keywords: imaging/image analysis: clinical • retina • imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound)