Application of Convolutional Neural Network in Classification of Autofluorescence Image of Diabetic Retina Fundus

  IJETT-book-cover  International Journal of Engineering Trends and Technology (IJETT)          
  
© 2020 by IJETT Journal
Volume-68 Issue-12
Year of Publication : 2020
Authors : Suzhe Ye?Daguan Ke
DOI :  10.14445/22315381/IJETT-V68I12P210

Citation 

MLA Style: Suzhe Ye?Daguan Ke. Application of Convolutional Neural Network in Classification of Autofluorescence Image of Diabetic Retina Fundus International Journal of Engineering Trends and Technology 68.12(2020):54-57. 

APA Style:Suzhe Ye?Daguan Ke. Nallusamy. Application of Convolutional Neural Network in Classification of Autofluorescence Image of Diabetic Retina Fundus.  International Journal of Engineering Trends and Technology, 68(12), 54-57.

Abstract
Diabetic retinopathy is one of the complications of diabetes. The common medical diagnosis method is fluorescein fundus angiography, but the diagnosis process requires fluorescein sodium injection. The fundus autofluorescence technology used in this study can be harmless and has a better application prospect for patients who cannot be angiographic examinations. However, the naked eye cannot recognize the early fundus images and need to introduce computer-aided diagnosis. This paper`s research object is 190 fundus autofluorescence images, and the accuracy of the 10-fold cross-check is used as the evaluation index. Compare the effects of convolutional neural network algorithms on classification performance under different image resolutions and image enhancement operations. The optimal image resolution is 64*64, the image enhancement operation is horizontal flip, and the optimal accuracy rate is 0.92105. After exploring the network structure, it is found that there is a better result without modifying the network. This article summarizes the following training steps: first, use the basic model to select the appropriate image resolution and image enhancement operation, and secondly, modify the network layer and explore the network through trial and error.

Reference
[1] Shannon C E. A mathematical theory of communication[J]. The Bell system technical journal, 27(3) (1948) 379-423.
[2] LeCun Y, Bengio Y. Convolutional networks for images, speech, and time series[J]. The handbook of brain theory and neural networks, 3361(10) 1995.
[3] Hinton G E, Salakhutdinov R R. Reducing the dimensionality of data with neural networks[J]. science, , 313(5786) (2006) 504-507.
[4] Krizhevsky A, Sutskever I, Hinton G E. Imagenet classification with deep convolutional neural networks[C]//Advances in neural information processing systems. 2012: 1097-1105.
[5] Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.
[6] Szegedy C, Vanhoucke V, Ioffe S, et al. Rethinking the inception architecture for computer vision[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. (2016) 2818-2826.
[7] He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. (2016) 770-778.
[8] Szegedy C, Ioffe S, Vanhoucke V, et al. Inception-v4, inceptionresnet, and the impact of residual connections on learning[J]. arXiv preprint arXiv:1602.07261, 2016.
[9] Xie S, Girshick R, Dollár P, et al. Aggregated residual transformations for deep neural networks[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. (2017) 1492-1500.
[10] Kermany D, Zhang K, Goldbaum M. Large dataset of labeled optical coherence tomography (oct) and chest x-ray images[J]. Mendeley Data, v3 http://dx. doi. org/10.17632/rscbjbr9sj, 2018, 3.
[11] Sabour S, Frosst N, Hinton G E. Dynamic routing between capsules[C]//Advances in neural information processing systems. (2017) 3856-3866.
[12] Zhang Z, Ning G, Cen Y, et al. Progressive neural networks for image classification[J]. arXiv preprint arXiv:1804.09803, 2018.
[13] Zhang X, Zhou X, Lin M, et al. Shufflenet: An extremely efficient convolutional neural network for mobile devices[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. (2018) 6848-6856.
[14] Narendra Mohan, Recognition of Skin Diseases using Deep Neural Network Optimized by Group Teaching Algorithm, International Journal of Engineering Trends and Technology 68(9)(2020):109-120.

Keywords
Autofluorescence image of fundus; Convolutional neural network; Computer-aided diagnosis