Abstract
Aim: To assess the efficacy of automated “disease/no disease” grading for diabetic retinopathy within a systematic screening programme.
Methods: Anonymised images were obtained from consecutive patients attending a regional primary care based diabetic retinopathy screening programme. A training set of 1067 images was used to develop automated grading algorithms. The final software was tested using a separate set of 14406 images from 6722 patients. The sensitivity and specificity of manual and automated systems operating as “disease/no disease” graders (detecting poor quality images and any diabetic retinopathy) were determined relative to a clinical reference standard.
Results: The reference standard classified 8.2% of the patients as having ungradable images (technical failures) and 62.5% as having no retinopathy. Detection of technical failures or any retinopathy was achieved by manual grading with 86.5% sensitivity (95% confidence interval 85.1 to 87.8) and 95.3% specificity (94.6 to 95.9) and by automated grading with 90.5% sensitivity (89.3 to 91.6) and 67.4% specificity (66.0 to 68.8). Manual and automated grading detected 99.1% and 97.9%, respectively, of patients with referable or observable retinopathy/maculopathy. Manual and automated grading detected 95.7% and 99.8%, respectively, of technical failures.
Conclusion: Automated “disease/no disease” grading of diabetic retinopathy could safely reduce the burden of grading in diabetic retinopathy screening programmes.
Methods: Anonymised images were obtained from consecutive patients attending a regional primary care based diabetic retinopathy screening programme. A training set of 1067 images was used to develop automated grading algorithms. The final software was tested using a separate set of 14406 images from 6722 patients. The sensitivity and specificity of manual and automated systems operating as “disease/no disease” graders (detecting poor quality images and any diabetic retinopathy) were determined relative to a clinical reference standard.
Results: The reference standard classified 8.2% of the patients as having ungradable images (technical failures) and 62.5% as having no retinopathy. Detection of technical failures or any retinopathy was achieved by manual grading with 86.5% sensitivity (95% confidence interval 85.1 to 87.8) and 95.3% specificity (94.6 to 95.9) and by automated grading with 90.5% sensitivity (89.3 to 91.6) and 67.4% specificity (66.0 to 68.8). Manual and automated grading detected 99.1% and 97.9%, respectively, of patients with referable or observable retinopathy/maculopathy. Manual and automated grading detected 95.7% and 99.8%, respectively, of technical failures.
Conclusion: Automated “disease/no disease” grading of diabetic retinopathy could safely reduce the burden of grading in diabetic retinopathy screening programmes.
Original language | English |
---|---|
Pages (from-to) | 1512-1517 |
Number of pages | 6 |
Journal | British Journal of Ophthalmology |
Volume | 91 |
Issue number | 11 |
Early online date | 15 May 2007 |
DOIs | |
Publication status | Published - Nov 2007 |
Keywords
- Aged
- Aged, 80 and over
- Algorithms
- Diabetic Retinopathy
- Female
- Humans
- Male
- Mass Screening
- Middle Aged
- Program Evaluation
- Sensitivity and Specificity
- Severity of Illness Index
Fingerprint
Dive into the research topics of 'The efficacy of “disease/no disease” grading for diabetic retinopathy in a systematic screening programme'. Together they form a unique fingerprint.Impacts
-
Automated screening systems for the early detection of diabetic retinopathy: development and incorporation into a national screening programme
John A. Olson (Coordinator), Peter Sharp (Coordinator) & John Forrester (Coordinator)
Impact