Aplicação de imagens sintéticas para otimização de modelos computacionais de detecção do estrabismo

  • Jonathan Santos UPM
  • Ismar Frango UPM


Strabismus is among the eye diseases that most lead to blindness or low vision, affecting about 4 % of the world population. Fortunately, the disease can be treated. Diagnosis even in the first moments of its manifestation dramatically increases the possibility of successful treatment. There are several proposals in the scientific literature for detecting and supporting the diagnosis of pathology, however, we have not found studies that seek to propose means of optimization for these techniques. This article presents a methodology for optimizing supervised strabismus detection models by increasing data using realistic synthetic samples. In evaluation, the proposed technique resulted in a gain of 7 % accuracy.


Celebrities images | kaggle. https://www.kaggle.com/greg115/ celebrities-100k. (Accessed on 02/15/2020).

Eye disease dataset | kaggle. https://www.kaggle.com/kondwani/ eye-disease-dataset. (Accessed on 11/02/2019).

Home - keras documentation. https://keras.io/. (Accessed on 01/27/2020).

Keras-gan/dcgan.py at master · eriklindernoren/keras-gan. https://github. com/eriklindernoren/Keras-GAN/blob/master/dcgan/dcgan.py. (Accessed on 01/27/2020).

Losses - keras documentation. https://keras.io/losses/. (Accessed on 01/27/2020).

Metrics - keras documentation. https://keras.io/metrics/. (Accessed on 01/27/2020).

Optimizers - keras documentation. https://keras.io/optimizers/. (Accessed on 01/27/2020).

Python.org. https://www.python.org/. (Accessed on 01/27/2020).

Tensorflow. https://www.tensorflow.org/. (Accessed on 01/27/2020).

Abrahamsson, M., Fabian, G., and Sjöstrand, J. (1986). Photorefraction: a useful tool to detect small angle strabismus. Acta ophthalmologica, 64(1):101–104.

Calimeri, F., Marzullo, A., Stamile, C., and Terracina, G. (2017). Biomedical data augmentation using generative adversarial neural networks. In International conference on artificial neural networks. Springer.

Chen, Z., Fu, H., Lo, W.-L., and Chi, Z. (2018). Strabismus recognition using eyetracking data and convolutional neural networks. Journal of healthcare engineering, 2018.

De Almeida, J. D. S., Silva, A. C., De Paiva, A. C., and Teixeira, J. A. M. (2012). Computational methodology for automatic detection of strabismus in digital images through hirschberg test. Computers in biology and medicine, 42(1):135–146.

De Almeida, J. D. S., Silva, A. C., Teixeira, J. A. M., Paiva, A. C., and Gattass, M. (2015). Computer-aided methodology for syndromic strabismus diagnosis. Journal of digital imaging, 28(4):462–473.

Frid-Adar, M., Diamant, I., Klang, E., Amitai, M., Goldberger, J., and Greenspan, H. (2018). Gan-based synthetic medical image augmentation for increased cnn performance in liver lesion classification. Neurocomputing, 321:321–331.

Fukushima, K. (1980). Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position. Biological cybernetics, 36(4):193–202.

Goodfellow, I., Bengio, Y., and Courville, A. (2016). Deep learning. MIT press.

Goodfellow, I., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A., and Bengio, Y. (2014). Generative adversarial nets. In Advances in neural information processing systems, pages 2672–2680.

Hwang, S.-K. and Kim, W.-Y. (2006). A novel approach to the fast computation of zernike moments. Pattern Recognition, 39(11):2065–2076.

Ingram, R. (1977). Refraction as a basis for screening children for squint and amblyopia. British Journal of Ophthalmology, 61(1):8–15.

Kaakinen, K. (1979). A simple method for screening of children with strabismus, anisometropia or ametropia by simultaneous photography of the corneal and the fundus reflexes. Acta ophthalmologica, 57(2):161–171.

Kelleher, J. D. (2019). Deep Learning. MIT press.

Kim, H.-J. and Kim, W.-Y. (2008). Eye detection in facial images using zernike moments with svm. ETRI journal, 30(2):335–337.

LeCun, Y., Boser, B., Denker, J. S., Henderson, D., Howard, R. E., Hubbard, W., and Jackel, L. D. (1989). Backpropagation applied to handwritten zip code recognition. Neural computation, 1(4):541–551.

Lopes, A. C. (2006). Diagnóstico e tratamento, volume 1. Editora Manole Ltda.

Lu, J., Fan, Z., Zheng, C., Feng, J., Huang, L., Li, W., and Goodman, E. D. (2018). Automated strabismus detection for telemedicine applications. arXiv preprint arXiv:1809.02940, page 2.

Radford, A., Metz, L., and Chintala, S. (2015). Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprint arXiv:1511.06434.

Santos, J. S. and Silveira, I. F. (2020). Generating photorealistic images of people’s eyes with strabismus using deep convolutional generative adversarial networks.

Valente, T. L. A., de Almeida, J. D. S., Silva, A. C., Teixeira, J. A. M., and Gattass, M. (2017). Automatic diagnosis of strabismus in digital videos through cover test. Computer methods and programs in biomedicine, 140:295–305.
Como Citar

Selecione um Formato
SANTOS, Jonathan; FRANGO, Ismar. Aplicação de imagens sintéticas para otimização de modelos computacionais de detecção do estrabismo. In: SIMPÓSIO BRASILEIRO DE COMPUTAÇÃO APLICADA À SAÚDE (SBCAS), 20. , 2020, Evento Online. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2020 . p. 13-24. ISSN 2763-8952. DOI: https://doi.org/10.5753/sbcas.2020.11498.