TOSCA-Imaging--developing Internet based image processing software for screening and diagnosis of diabetic retinopathy.

The primary aim of TOSCA-Imaging, which was a part of the TOSCA Project financed by EU's Fifth Framework IST Programme, was to develop Internet based software and image data bases for screening and diagnosis of diabetic retinopathy and implementing it into a real life situation. The work consisted of: 1) Construction of an Internet based communication platform for transmitting and analyzing retinal images. 2) Implementation of routines for detecting the first microaneurysm (transition from normal to pathologic), detecting patients needing referral for treatment (presence of venous beading or hard exudates near the fovea), and for serial analysis (image alignment). 3) Construction of a reference image data base. A preliminary validation showed that decisions that depended on a precise detection of individual lesions, e.g. the detection of normality, had a sensitivity and specificity of around 80%, whereas decisions that depended on the detection of lesion patterns, e.g. clinically significant macular oedema, had a sensitivity and specificity of more than 95%. Validation of the reference image data base by double grading by two expert graders suggested a sensitivity and a specificity of just below 90% for any lesion and of more than 95% for predicting the overall retinopathy grade. TOSCA-Imaging has succeeded in its primary aim of developing Internet based software and implementing it into a real life situation, integrating work within image processing done in four different European countries (England, Germany, Ireland, and Denmark) to be accessed from one Internet web site.

The relationship between age and colour content in fundus images.

INTRODUCTION: The morphological appearance of the ocular fundus is one of the key parameters used in the diagnosis and management of retinal disease. However, optical imperfections in the refractive media result in blurring, low luminance and contrast, and changes in the colour composition of the image which can be seen as an increasing yellowish appearance with age. The introduction of a method for quantifying this age-related change in colour content may help in diagnosing and grading pathological changes in the eye lens which are secondary to ocular and systemic diseases. METHODS: A total of 102 digitized fundus images from 102 healthy subjects (mean age = 50.4 years, range 7.0-94.3 years) were used to build a model for estimating the age of the subject from the colour content of the images. RESULTS: Estimation of age from the fundus images could be done within approximately 16 years. This variation could be reduced considerably by analysis of repeated photographs from the same examination. CONCLUSION: The colour content of fundus images can be used to estimate the ages of healthy subjects. Furthermore, when the colour content of fundus images deviates from that expected according to subject age, this may indicate causes other than age of increased light absorption in the lens, such as cumulative exposure to hyperglycaemia in diabetic patients. This could potentially be used to identify patients with undiagnosed type 2 diabetes in the general population and help to establish their risk of developing late diabetic complications as the cumulative exposure to hyperglycaemia is unknown at the time of diagnosis of the disease.

Using a model of the colour content in retinal fundus images to screen for sight threatening diabetic retinopathy.

The aim of the study was to assess the potential use of a model of the colour content in retinal fundus images to screen for sight threatening retinopathy in diabetes. Diabetic retinopathy is the most frequent cause of blindness in the population of working age in industrialised countries, but efficient therapies do exist, and accurate and early diagnosis, and correct treatment can prevent blindness in more than 50% of all cases. However, up to 50% of cases of type 2 diabetes, which comprises 85-90% of all patients, are undiagnosed, with an average delay of 10 years between the onset of the condition and diagnosis. In an other study we have described how there is a linear relation between age and the colour composition of retinal images from non-diabetic subjects. In the present study this relation was compared to the colour composition of retinal images from diabetes patients. We found that for the patients in the present study there is a significant difference in the colour composition between normal subjects and diabetic subjects with retinopathy. Although the number of patients in our study is too small to allow any conclusion, we suggest that this difference potentially may be used as the basis for a simple screening method for sight threatening retinopathy in unrecognised diabetes, or potentially may help estimating the risk of developing diabetic late complications in newly diagnosed type 2 diabetes.