Skip to main content
NCBI home page
The British Journal of Ophthalmology logoLink to The British Journal of Ophthalmology
. 1996 Nov;80(11):940–944. doi: 10.1136/bjo.80.11.940

Automatic detection of diabetic retinopathy using an artificial neural network: a screening tool.

G G Gardner 1, D Keating 1, T H Williamson 1, A T Elliott 1
PMCID: PMC505667  PMID: 8976718

Abstract

AIMS: To determine if neural networks can detect diabetic features in fundus images and compare the network against an ophthalmologist screening a set of fundus images. METHODS: 147 diabetic and 32 normal images were captured from a fundus camera, stored on computer, and analysed using a back propagation neural network. The network was trained to recognise features in the retinal image. The effects of digital filtering techniques and different network variables were assessed. 200 diabetic and 101 normal images were then randomised and used to evaluate the network's performance for the detection of diabetic retinopathy against an ophthalmologist. RESULTS: Detection rates for the recognition of vessels, exudates, and haemorrhages were 91.7%, 93.1%, and 73.8% respectively. When compared with the results of the ophthalmologist, the network achieved a sensitivity of 88.4% and a specificity of 83.5% for the detection of diabetic retinopathy. CONCLUSIONS: Detection of vessels, exudates, and haemorrhages was possible, with success rates dependent upon preprocessing and the number of images used in training. When compared with the ophthalmologist, the network achieved good accuracy for the detection of diabetic retinopathy. The system could be used as an aid to the screening of diabetic patients for retinopathy.

Full text

PDF
940

Images in this article

941Image
on p.941
941Image
on p.941
944Image
on p.944

Selected References

These references are in PubMed. This may not be the complete list of references from this article.

  1. Astion M. L., Wilding P. The application of backpropagation neural networks to problems in pathology and laboratory medicine. Arch Pathol Lab Med. 1992 Oct;116(10):995–1001. [PubMed] [Google Scholar]
  2. Goldbaum M. H., Katz N. P., Nelson M. R., Haff L. R. The discrimination of similarly colored objects in computer images of the ocular fundus. Invest Ophthalmol Vis Sci. 1990 Apr;31(4):617–623. [PubMed] [Google Scholar]
  3. Goldbaum M. H., Sample P. A., White H., Côlt B., Raphaelian P., Fechtner R. D., Weinreb R. N. Interpretation of automated perimetry for glaucoma by neural network. Invest Ophthalmol Vis Sci. 1994 Aug;35(9):3362–3373. [PubMed] [Google Scholar]
  4. Javitt J. C., Canner J. K., Sommer A. Cost effectiveness of current approaches to the control of retinopathy in type I diabetics. Ophthalmology. 1989 Feb;96(2):255–264. doi: 10.1016/s0161-6420(89)32923-x. [DOI] [PubMed] [Google Scholar]
  5. Kalm H., Egertsen R., Blohmé G. Non-stereo fundus photography as a screening procedure for diabetic retinopathy among patients with type II diabetes. Compared with 60D enhanced slit-lamp examination. Acta Ophthalmol (Copenh) 1989 Oct;67(5):546–553. doi: 10.1111/j.1755-3768.1989.tb04106.x. [DOI] [PubMed] [Google Scholar]
  6. Kalm H. Non-stereo photographic screening in long-term follow-up for detection of proliferative diabetic retinopathy. Acta Ophthalmol (Copenh) 1992 Apr;70(2):228–234. doi: 10.1111/j.1755-3768.1992.tb04129.x. [DOI] [PubMed] [Google Scholar]
  7. Lee V. S., Kingsley R. M., Lee E. T., Lu M., Russell D., Asal N. R., Bradford R. H., Jr, Wilkinson C. P. The diagnosis of diabetic retinopathy. Ophthalmoscopy versus fundus photography. Ophthalmology. 1993 Oct;100(10):1504–1512. doi: 10.1016/s0161-6420(93)31449-1. [DOI] [PubMed] [Google Scholar]
  8. Lo S. C., Freedman M. T., Lin J. S., Mun S. K. Automatic lung nodule detection using profile matching and back-propagation neural network techniques. J Digit Imaging. 1993 Feb;6(1):48–54. doi: 10.1007/BF03168418. [DOI] [PubMed] [Google Scholar]
  9. Mutlukan E., Keating D. Visual field interpretation with a personal computer based neural network. Eye (Lond) 1994;8(Pt 3):321–323. doi: 10.1038/eye.1994.65. [DOI] [PubMed] [Google Scholar]
  10. Phillips R. P., Spencer T., Ross P. G., Sharp P. F., Forrester J. V. Quantification of diabetic maculopathy by digital imaging of the fundus. Eye (Lond) 1991;5(Pt 1):130–137. doi: 10.1038/eye.1991.24. [DOI] [PubMed] [Google Scholar]
  11. Pugh J. A., Jacobson J. M., Van Heuven W. A., Watters J. A., Tuley M. R., Lairson D. R., Lorimor R. J., Kapadia A. S., Velez R. Screening for diabetic retinopathy. The wide-angle retinal camera. Diabetes Care. 1993 Jun;16(6):889–895. doi: 10.2337/diacare.16.6.889. [DOI] [PubMed] [Google Scholar]
  12. Ryder R. E., Vora J. P., Atiea J. A., Owens D. R., Hayes T. M., Young S. Possible new method to improve detection of diabetic retinopathy: Polaroid non-mydriatic retinal photography. Br Med J (Clin Res Ed) 1985 Nov 2;291(6504):1256–1257. doi: 10.1136/bmj.291.6504.1256. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Schachat A. P., Hyman L., Leske M. C., Connell A. M., Hiner C., Javornik N., Alexander J. Comparison of diabetic retinopathy detection by clinical examinations and photograph gradings. Barbados (West Indies) Eye Study Group. Arch Ophthalmol. 1993 Aug;111(8):1064–1070. doi: 10.1001/archopht.1993.01090080060019. [DOI] [PubMed] [Google Scholar]
  14. Singer D. E., Nathan D. M., Fogel H. A., Schachat A. P. Screening for diabetic retinopathy. Ann Intern Med. 1992 Apr 15;116(8):660–671. doi: 10.7326/0003-4819-116-8-660. [DOI] [PubMed] [Google Scholar]
  15. Spenceley S. E., Henson D. B., Bull D. R. Visual field analysis using artificial neural networks. Ophthalmic Physiol Opt. 1994 Jul;14(3):239–248. doi: 10.1111/j.1475-1313.1994.tb00004.x. [DOI] [PubMed] [Google Scholar]
  16. Spencer T., Phillips R. P., Sharp P. F., Forrester J. V. Automated detection and quantification of microaneurysms in fluorescein angiograms. Graefes Arch Clin Exp Ophthalmol. 1992;230(1):36–41. doi: 10.1007/BF00166760. [DOI] [PubMed] [Google Scholar]
  17. Taylor R., Lovelock L., Tunbridge W. M., Alberti K. G., Brackenridge R. G., Stephenson P., Young E. Comparison of non-mydriatic retinal photography with ophthalmoscopy in 2159 patients: mobile retinal camera study. BMJ. 1990 Dec 1;301(6763):1243–1247. doi: 10.1136/bmj.301.6763.1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Ward N. P., Tomlinson S., Taylor C. J. Image analysis of fundus photographs. The detection and measurement of exudates associated with diabetic retinopathy. Ophthalmology. 1989 Jan;96(1):80–86. [PubMed] [Google Scholar]
  19. Wareham N. J. Cost-effectiveness of alternative methods for diabetic retinopathy screening. Diabetes Care. 1993 May;16(5):844–844. doi: 10.2337/diacare.16.5.844a. [DOI] [PubMed] [Google Scholar]
  20. Williams R., Nussey S., Humphry R., Thompson G. Assessment of non-mydriatic fundus photography in detection of diabetic retinopathy. Br Med J (Clin Res Ed) 1986 Nov 1;293(6555):1140–1142. doi: 10.1136/bmj.293.6555.1140. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Wu Y., Giger M. L., Doi K., Vyborny C. J., Schmidt R. A., Metz C. E. Artificial neural networks in mammography: application to decision making in the diagnosis of breast cancer. Radiology. 1993 Apr;187(1):81–87. doi: 10.1148/radiology.187.1.8451441. [DOI] [PubMed] [Google Scholar]

Articles from The British Journal of Ophthalmology are provided here courtesy of BMJ Publishing Group

RESOURCES