Optical Coherence Tomography Image Classification Using Hybrid Deep Learning and Ant Colony Optimization

Abstract

Optical coherence tomography (OCT) is widely used to detect and classify retinal diseases. However, OCT-image-based manual detection by ophthalmologists is prone to errors and subjectivity. Thus, various automation methods have been proposed; however, improvements in detection accuracy are required. Particularly, automated techniques using deep learning on OCT images are being developed to detect various retinal disorders at an early stage. Here, we propose a deep learning-based automatic method for detecting and classifying retinal diseases using OCT images. The diseases include age-related macular degeneration, branch retinal vein occlusion, central retinal vein occlusion, central serous chorioretinopathy, and diabetic macular edema. The proposed method comprises four main steps: three pretrained models, DenseNet-201, InceptionV3, and ResNet-50, are first modified according to the nature of the dataset, after which the features are extracted via transfer learning. The extracted features are improved, and the best features are selected using ant colony optimization. Finally, the best features are passed to the k-nearest neighbors and support vector machine algorithms for final classification. The proposed method, evaluated using OCT retinal images collected from Soonchunhyang University Bucheon Hospital, demonstrates an accuracy of 99.1% with the incorporation of ACO. Without ACO, the accuracy achieved is 97.4%. Furthermore, the proposed method exhibits state-of-the-art performance and outperforms existing techniques in terms of accuracy.

Keywords: age-related macular degeneration; ant colony optimization; branch retinal vein occlusion; central retinal vein occlusion; central serous chorioretinopathy; convolutional neural network; deep learning; diabetic macular edema; feature selection; machine learning; optical coherence tomography.

Figure 1

Flow diagram of the proposed method. Features are extracted by transfer learning, a feature vector is constructed, and ant colony optimization (ACO) is applied on the feature vector for optimization and final classification.

Figure 2

Modified proposed architecture of ResNet-50.

Figure 3

Modified architecture of DenseNet-201.

Figure 4

Modified proposed architecture of InceptionV3.

Figure 5

Confusion matrix of cubic support vector machine using deep learning models: (a) modified ResNet-50, (b) modified InceptionV3, and (c) modified DenseNet-201.

Figure 6

Confusion matrix of cubic support vector machine using deep learning models: (a) modified ResNet-50, (b) modified InceptionV3, and (c) modified DenseNet-201 with ACO.

Figure 7

ROC plots for the selected OCT image classes of modified DenseNet-201 using cubic SVM after applying ACO.

Figure 8

Sample of the OCT images from our proposed dataset and corresponding color images for understanding the input dataset images.