DEVELOPMENT OF AN EXPERT SYSTEM BASED ON FUZZY LOGIC FOR EARLY DIAGNOSIS OF GLAUCOMA

Nowadays, decision-making systems that rely on images are becoming increasingly crucial, especially in the medical field. Images have become a fundamental tool for clinical research and diagnosing illnesses. In the case of glaucoma, a disease that can damage the optic nerve head and result in irreversible vision loss, a new Fuzzy Expert System has been developed for early diagnosis. Original ONH images are preprocessed with filters to remove noise, followed by using the Canny detector algorithm to detect contours. Key parameters are then extracted by identifying elliptical forms of the optic disc and excavation using the Randomized Hough Transform. A classification algorithm based on fuzzy logic is used to assess patients' conditions, taking into account both instrumental parameters and risk factors such as age, race, and family history. The system is tested on a dataset of ophthalmologic images, showing a significant improvement in predictions compared to existing methods, with over 96% accuracy in identifying cases suspected to have glaucoma.

1. Dahl A.A. Glaucoma Causes, Symptoms. Treatment & Surgery, 2018. [Online]. https://www.medicinenet.com/glaucoma/article.htm. [Accessed: 10-Available at: Mar-2018].

2. Nayak J., Acharya U.R., Bhat P.S., Shetty A., Lim T.C. (2009) Automated diagnosis ofglaucoma using digital fundus images. J. Med. Syst., vol. 33, no. 5, pp. 337–346.7.

3. Abysheva L.D., Aldasheva N.A., Utel'baeva Z.T. (2019) Sravnitel'naja ocenka metodov lechenija refrakternoj glaukomy. Vestnik KazNMU, no. 3, pp. 185–188.

4. Kourkoutas D., Karanasiou I.S., Tsekouras G.J., Moshos M., Iliakis E., Georgopoulos G. (2012) Glaucoma risk assessment using a non-linear multivariableregression method. Comput. Methods Programs Biomed, vol.108, no. 3, pp. 1149–1159.

5. Greany M.J. et al. (2002) Comparisons of optic nerve imaging methods to distinguishnormal eyes from those with glaucomaInvest. Ophthalmol. Visual Sci., vol. 43, no. 1, pp. 140–14.

6. Lachkar Y., Sellem E. (2024) La papille glaucomateuse en pratique. Laboratoire Chauvin Bausch & Lombo, no. 80.

7. Man Hua, Chenghao Zhub, Xiaoxing Lib, Yongli Xub. (2016) Optic Cup Segmentationfrom Fundus Images for Glaucoma Diagnosis, Taylor & Francis, 2016, pp. 1–8. [Online].

8. Chandrika S., Nirmala K. (2013) Analysis of CDR detection for glaucoma diagnosis. Int. J. Eng. Res. Appl. (IJERA), ISSN: 2248-9622, NCACCT-19.

9. Sobia Nazi, Sheela N. Rao. (2014) Glaucoma detection in color fundus images usingcup to disc ratio. Int. J. Eng. Sci. (IJES), vol. 3, no. 6, pp. 51–58.

10. Kim P.Y., Iftekharuddin K.M. (2013) Novel fractal feature-Based multiclass glaucomadetection and progression prediction. IEEE J. Biomed. Health. Inf., vol. 17, no. 2.

11. Tan M.H., Sun Y., Ong S.H., Liu J., Baskaran M., Aung T., Wong T.Y. (2013) Automaticnotch detection in retinal images. IEEE 10th International Symposium onBiomedical Imaging, April, San Francisco, CA, USA.

12. Yadav D., Sarathi M.P., Dutta M.K. (2014) Classification of glaucoma based on texturefeatures using neural networks. IEEE 7th International Conference onContemporary Computing, August, Noida, India.

13. Noronha K.P., Acharya U.R., Nayaka K.P., Martis R.J., Bhandary S.V. (2014) Automatedclassification of glaucoma stages using higher order cumulant features Biomed. Signal Process. Control, no. 10, pp. 174–183.

14. Raja C., Gangatharan N. (2015) A hybrid swarm algorithm for optimizing glaucomadiagnosis. Comput. Biol. Med., no. 63, pp. 196–207.

15. Ashish Issac, M. Partha Sarathi, Malay Kishore Dutta. (2015) An adaptive thresholdbased image processing technique for improved glaucoma detection andclassification. Comput. Methods Programs Biomed., vol. 122, no. 2, pp. 229–244.

16. Cosmin A. Basca, Mihai Talos, Remus Brad. (2005) Randomized hough transform forellipse detection with result clustering. IEEE International Conference onComputer as a Tool, Serbia & Montenegro, Belgrade.

17. Bennett N., Burridge R., Saito N. (1999) A method to detect and characterize ellipsesusing the Hough transform. IEEE Trans. Pattern Anal. Mach. Intell., vol. 21, no. 7, pp. 652–657.

18. Corine Schnyder, André Mermoud, et al. Glaucome (2005) Atlas En Ophtalmologie, p. 463.

19. Van Broekhoven E., De Baets B. (2004) A comparison of three methods for computingthe center of gravity defuzzification. IEEE International Conference onFuzzy Systems, Budapest, Hungary