Affiliation:
1. Murom Institute (Branch) Vladimir State University named after A. G. and N. G. Stoletov
Abstract
The purpose of the research to develop a technique for detecting glaucoma, which is based on calculating the size ratio of the optical cup and optical disc and the "four quadrants" rule. Their use increases the accuracy of glaucoma detection in human retina images.Methods. A glaucoma detection technique is proposed that uses the ratio of the vertical diameter of the cup to the vertical diameter of the disc and the "four quadrants" rule as the two main parameters for the detection of glaucoma. The optic nerve disc (OD), the ocular cup (OCH) are segmented using the area extension method and the watershed method, and then combined to obtain the final results. Their union is performed using the logical operation OR. The resulting images are approximated using circular approximation, since its implementation is simple by calculating a single center and radius. For diagnostics, it was decided to use two parameters: the ratio of the cup and the disc (OCD) and the rule of "four quadrants". Their combined assessment makes it possible to increase the accuracy of glaucoma detection.Results: the study of the proposed technique was performed on retinal images obtained from 4 databases: HRF, DIARETDB1, DRIONS-DB, Messidor. The study showed that the proposed technique correctly identifies 75 retinal images as glaucoma out of 84 with a total sensitivity of 91.67%. Of the 163 normal images, 154 were correctly classified as normal with a specificity of 94.47%.Conclusion. The proposed method is simple and computationally efficient. It can be effectively used in computer diagnostics of glaucoma in the early stages of the disease.
Publisher
Southwest State University
Reference20 articles.
1. Abdel-Razik Youssif A., Ghalwash A., Abdel-Rahman Ghoneim A. Optic disc detection from normalized digital fundus images by means of a vessels’ direction matched filter. Medical Imaging, 2018, vol. 27, no. 1, pp. 11–18.
2. Abrmoff M. D., Niemeijer M., Suttorp-Schulten M. S. A., Viergever M. A., Russell S. R., Ginneken B. V. Evaluation of a system for automatic detection of diabetic retinopathy from color fundus photographs in a large population of patients with diabetes. Diabetes Care, 2008, vol. 31, no. 2, pp. 193–198.
3. Komkova S. V. Metodika formirovaniya vektora priznakov po izobrazheniyam setchatki glaza [Methodology for forming a feature vector based on retinal images]. Estestvennye i tekhnicheskie nauki = Natural and Technical Sciences, 2021, no. 4, pp. 250‒252.
4. Komkova S. V. Algoritm identifikacii ekssudatov na izobrazheniyah setchatki glaza cheloveka [Algorithm for identification of exudates in human retinal images]. Vestnik komp'yuternyh i informacionnyh tekhnologij = Bulletin of Computer and Information Technologies, 2022, vol. 19, no. 1, pp. 47‒51. https://doi.org/10.14489/vkit.2022.01.pp.047-051
5. Cheng J., Liu J., Xu Y., Yin F., Wong D. W. K. Glaucoma detection using retinal fundus images: A review. Computers in Biology and Medicine, 2019, no. 109, pp. 216‒229. https://doi.org/10.1016/j.compbiomed.2019.04.010