Automated detection of diabetic retinopathy on digital fundus images

Diabetic Medicine - Tập 19 Số 2 - Trang 105-112 - 2002
Chanjira Sinthanayothin1, J.F. Boyce2, Tom H. Williamson3, Helen Cook3, Evelyn Mensah3, Sadhna Lal3, David Usher2
1Image Processing Group, Department of Physics, King's College London , UK.
2Image Processing Group, Department of Physics, King’s College London and
3Department of Ophthalmology, St. Thomas' Hospital, London, UK

Tóm tắt

AbstractAims The aim was to develop an automated screening system to analyse digital colour retinal images for important features of non‐proliferative diabetic retinopathy (NPDR).Methods High performance pre‐processing of the colour images was performed. Previously described automated image analysis systems were used to detect major landmarks of the retinal image (optic disc, blood vessels and fovea). Recursive region growing segmentation algorithms combined with the use of a new technique, termed a ‘Moat Operator’, were used to automatically detect features of NPDR. These features included haemorrhages and microaneurysms (HMA), which were treated as one group, and hard exudates as another group. Sensitivity and specificity data were calculated by comparison with an experienced fundoscopist.Results The algorithm for exudate recognition was applied to 30 retinal images of which 21 contained exudates and nine were without pathology. The sensitivity and specificity for exudate detection were 88.5% and 99.7%, respectively, when compared with the ophthalmologist. HMA were present in 14 retinal images. The algorithm achieved a sensitivity of 77.5% and specificity of 88.7% for detection of HMA.Conclusions Fully automated computer algorithms were able to detect hard exudates and HMA. This paper presents encouraging results in automatic identification of important features of NPDR.Diabet. Med. 19, 105–112 (2002)

Từ khóa


Tài liệu tham khảo

10.2337/diacare.17.8.909

1990, Diabetes care and research in Europe: the St Vincent Declaration, Diabet Med, 7, 60

British Diabetic Association., 1995, St Vincent Joint Task Force for Diabetes. Final Report.

Bagga P, 1998, Survey of diabetic retinopathy screening services in England and Wales, Diabet Med, 15, 80, 10.1002/(SICI)1096-9136(199809)15:9<780::AID-DIA632>3.0.CO;2-K

10.1002/(SICI)1096-9136(199712)14:5 <S7::AID-DIA522>3.0.CO;2-R

10.1038/eye.1999.37

10.1002/(SICI)1096-9136(199803)15:3<254::AID-DIA543>3.0.CO;2-C

10.1002/(SICI)1096-9136(199803)15:3<250::AID-DIA565>3.0.CO;2-G

10.1136/bjo.83.8.902

10.1136/bjo.82.1.5

10.1038/eye.2000.80

ToulsonDL BoyceJF.Image segmentation using a neural network. IEE Colloquium on 3‐D Imaging Techniques For Medicine April 1991.

ToulsonDL BoyceJF.Segmentation of MR images using neural nets.Proceedings of the British Machine Vision Conference1991: 284–292.

10.1016/0262-8856(92)90050-D

Gonzales RC, 1993, Digital Image Processing., 229

10.1136/bjo.80.11.940

British Diabetic Association., 1997, Retinal Photography Screening for Diabetic Eye Disease.

10.1016/S0161-6420(90)32573-3