Deep Feature Glaucoma Diagnosis

• Lisle Faray de Paiva UFMA
• José Mateus Boaro Carvalho UFMA
• Arthur Guilherme Santos Fernandes UFMA
• Caio Manfredini da Silva Martins UFMA
• Geraldo Braz Junior UFMA

Abstract

Glaucoma is a silent disease which permanently damages the optic nerve. Asymptomatic in it’s early stages, it’s the second leading cause of blindness in the world. Several automatic diagnostic systems have been proposed, however, these systems are not capable of handling a wide range of images. Therefore, such methods are not feasible for use in screening programs. This paper proposes a methodology for efficiently detecting glaucoma that is capable of handling diverse images through feature extraction using Convolutional Neural Networks (CNNs). In this proposal, a total of 1090 images from four public data-sets were evaluated and it was concluded that the combination of pre trained CNNs and Specific Optimized Networks together with the use of the Logistic Regression classifier is promising in the detection of this pathology, obtaining accuracy’s of 86.8% e 86.3%

Keywords: Neural Networks, Logistic Regression, Glaucoma

References

Abbas, Q. (2017). Glaucoma-deep: detection of glaucoma eye disease on retinal fundus images using deep learning. Int J Adv Comput Sci Appl, 8(6):41–5.

Bergstra, J., Yamins, D., and Cox, D. D. (2013). Hyperopt: A python library for optimizing the hyperparameters of machine learning algorithms. In Proceedings of the 12th Python in science conference, pages 13–20. Citeseer.

Fernandes, A. G. S., da Silva Martins, C. M., de Moura Lima, A. C., Junior, G. B., de Almeida, J. D. S., and de Paiva, A. C. (2019). Meta aprendizagem de extrac¸ ˜ao de caracterýsticas aplicada ao diagn´ostico de glaucoma. In Anais do XIX Simp´osio Brasileiro de Computac¸ ˜ao Aplicada `a Sa´ude, pages 342–347. SBC.

Fumero, F., Alay´on, S., Sanchez, J., Sigut, J., and Gonzalez-Hernandez, M. (2011). Rimone: An open retinal image database for optic nerve evaluation. pages 1 – 6.

Han, S., Mao, H., and Dally, W. J. (2015). Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding. arXiv preprint arXiv:1510.00149.

linchundan (2018). 1000 fundus images with 39 categories.

Matsugu, M., Mori, K., Mitari, Y., and Kaneda, Y. (2003). Subject independent facial expression recognition with robust face detection using a convolutional neural network. Neural Networks, 16(5-6):555–559.

Mittapalli, P. S. and Kande, G. B. (2016). Segmentation of optic disk and optic cup from digital fundus images for the assessment of glaucoma. Biomedical Signal Processing and Control, 24:34–46. of Technology, B. I. (2018). ichallenge-gon.

Quigley, H. A. and Broman, A. T. (2006). The number of people with glaucoma worldwide in 2010 and 2020. British journal of ophthalmology, 90(3):262–267.

Raghavendra, U., Fujita, H., Bhandary, S. V., Gudigar, A., Tan, J. H., and Acharya, U. R. (2018). Deep convolution neural network for accurate diagnosis of glaucoma using digital fundus images. Information Sciences, 441:41–49.

Yu, H.-F., Huang, F.-L., and Lin, C.-J. (2011). Dual coordinate descent methods for logistic regression and maximum entropy models. Machine Learning, 85(1):41–75.