Classification of SD-OCT volumes with multi pyramids, LBP and HOG descriptors: application to DME detections

Retinal OCT image classification based on MGR-GAN

Accurately classifying optical coherence tomography (OCT) images is essential for diagnosing and treating ophthalmic diseases. This paper introduces a novel generative adversarial network framework called MGR-GAN. The masked image modeling (MIM) method is integrated into the GAN model's generator, enhancing its ability to synthesize more realistic images by reconstructing them based on unmasked patches. A ResNet-structured discriminator is employed to determine whether the image is generated by the generator. Through the unique game process of the generative adversarial network (GAN) model, the discriminator acquires high-level discriminant features, essential for precise OCT classification. Experimental results demonstrate that MGR-GAN achieves a classification accuracy of 98.4% on the original UCSD dataset. As the trained generator can synthesize OCT images with higher precision, and owing to category imbalances in the UCSD dataset, the generated OCT images are leveraged to address this imbalance. After balancing the UCSD dataset, the classification accuracy further improves to 99%.

Using optical coherence tomography to assess luster of pearls: technique suitability and insights

Luster is one of the vital indexes in pearl grading. To find a fast, nondestructive, and low-cost grading method, optical coherence tomography (OCT) is introduced to predict the luster grade through the texture features. After background removal, flattening, and segmentation, the speckle pattern of the region of interest is described by seven kinds of feature textures, including center-symmetric auto-correlation (CSAC), fractal dimension (FD), Gabor, gray level co-occurrence matrix (GLCM), histogram of oriented gradients (HOG), laws texture energy (LAWS), and local binary patterns (LBP). To find the relations between speckle-derived texture features and luster grades, four Four groups of pearl samples were used in the experiment to detect texture differences based on support vector machines (SVMs) and random forest classifier (RFC)) for investigating the relations between speckle-derived texture features and luster grades. The precision, recall, F1-score, and accuracy are more significant than 0.9 in several simulations, even after dimension reduction. This demonstrates that the texture feature from OCT images can be applied to class the pearl luster based on speckle changes.

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

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.

FN-OCT: Disease Detection Algorithm for Retinal Optical Coherence Tomography Based on a Fusion Network

Optical coherence tomography (OCT) is a new type of tomography that has experienced rapid development and potential in recent years. It is playing an increasingly important role in retinopathy diagnoses. At present, due to the uneven distributions of medical resources in various regions, the uneven proficiency levels of doctors in grassroots and remote areas, and the development needs of rare disease diagnosis and precision medicine, artificial intelligence technology based on deep learning can provide fast, accurate, and effective solutions for the recognition and diagnosis of retinal OCT images. To prevent vision damage and blindness caused by the delayed discovery of retinopathy, a fusion network (FN)-based retinal OCT classification algorithm (FN-OCT) is proposed in this paper to improve upon the adaptability and accuracy of traditional classification algorithms. The InceptionV3, Inception-ResNet, and Xception deep learning algorithms are used as base classifiers, a convolutional block attention mechanism (CBAM) is added after each base classifier, and three different fusion strategies are used to merge the prediction results of the base classifiers to output the final prediction results (choroidal neovascularization (CNV), diabetic macular oedema (DME), drusen, normal). The results show that in a classification problem involving the UCSD common retinal OCT dataset (108,312 OCT images from 4,686 patients), compared with that of the InceptionV3 network model, the prediction accuracy of FN-OCT is improved by 5.3% (accuracy = 98.7%, area under the curve (AUC) = 99.1%). The predictive accuracy and AUC achieved on an external dataset for the classification of retinal OCT diseases are 92 and 94.5%, respectively, and gradient-weighted class activation mapping (Grad-CAM) is used as a visualization tool to verify the effectiveness of the proposed FNs. This finding indicates that the developed fusion algorithm can significantly improve the performance of classifiers while providing a powerful tool and theoretical support for assisting with the diagnosis of retinal OCT.

Fully automated detection of retinal disorders by image-based deep learning

PURPOSE

With the aging population and the global diabetes epidemic, the prevalence of age-related macular degeneration (AMD) and diabetic macular edema (DME) diseases which are the leading causes of blindness is further increasing. Intravitreal injections with anti-vascular endothelial growth factor (anti-VEGF) medications are the standard of care for their indications. Optical coherence tomography (OCT), as a noninvasive imaging modality, plays a major part in guiding the administration of anti-VEGF therapy by providing detailed cross-sectional scans of the retina pathology. Fully automating OCT image detection can significantly decrease the tedious clinician labor and obtain a faithful pre-diagnosis from the analysis of the structural elements of the retina. Thereby, we explore the use of deep transfer learning method based on the visual geometry group 16 (VGG-16) network for classifying AMD and DME in OCT images accurately and automatically.

METHOD

A total of 207,130 retinal OCT images between 2013 and 2017 were selected from retrospective cohorts of 5319 adult patients from the Shiley Eye Institute of the University of California San Diego, the California Retinal Research Foundation, Medical Center Ophthalmology Associates, the Shanghai First People's Hospital, and the Beijing Tongren Eye Center, with 109,312 images (37,456 with choroidal neovascularization, 11,599 with diabetic macular edema, 8867 with drusen, and 51,390 normal) for the experiment. After images preprocessing, 1000 images (250 images from each category) from 633 patients were selected as validation dataset while the rest images from another 4686 patients were used as training dataset. We used deep transfer learning method to fine-tune the VGG-16 network pre-trained on the ImageNet dataset, and evaluated its performance on the validation dataset. Then, prediction accuracy, sensitivity, specificity, and receiver-operating characteristic (ROC) were calculated.

RESULTS

Experimental results proved that the proposed approach had manifested superior performance in retinal OCT images detection, which achieved a prediction accuracy of 98.6%, with a sensitivity of 97.8%, a specificity of 99.4%, and introduced an area under the ROC curve of 100%.

CONCLUSION

Deep transfer learning method based on the VGG-16 network shows significant effectiveness on classification of retinal OCT images with a relatively small dataset, which can provide assistant support for medical decision-making. Moreover, the performance of the proposed approach is comparable to that of human experts with significant clinical experience. Thereby, it will find promising applications in an automatic diagnosis and classification of common retinal diseases.

Automatic detection and recognition of multiple macular lesions in retinal optical coherence tomography images with multi-instance multilabel learning

Detection and recognition of macular lesions in optical coherence tomography (OCT) are very important for retinal diseases diagnosis and treatment. As one kind of retinal disease (e.g., diabetic retinopathy) may contain multiple lesions (e.g., edema, exudates, and microaneurysms) and eye patients may suffer from multiple retinal diseases, multiple lesions often coexist within one retinal image. Therefore, one single-lesion-based detector may not support the diagnosis of clinical eye diseases. To address this issue, we propose a multi-instance multilabel-based lesions recognition (MIML-LR) method for the simultaneous detection and recognition of multiple lesions. The proposed MIML-LR method consists of the following steps: (1) segment the regions of interest (ROIs) for different lesions, (2) compute descriptive instances (features) for each lesion region, (3) construct multilabel detectors, and (4) recognize each ROI with the detectors. The proposed MIML-LR method was tested on 823 clinically labeled OCT images with normal macular and macular with three common lesions: epiretinal membrane, edema, and drusen. For each input OCT image, our MIML-LR method can automatically identify the number of lesions and assign the class labels, achieving the average accuracy of 88.72% for the cases with multiple lesions, which better assists macular disease diagnosis and treatment.