They defined about 2% of the images of observation discrepancy between graders as ‘ambiguous’ category and found that they confuse the network and decrease overall performance when used for training. All the images were labelled by three retinal image analysis experts including one ophthalmologist from the EyePACS. 20 trained a deep learning framework using 3428 images (3179 accept, 249 reject, and 147 ambiguous), and the rest 3425 images (3302 accept and 123 reject) were used for evaluation. They extract CNN-based features as well as the unsupervised saliency map-based features and fuse them. 19 used a dataset including the training set with 3000 images and the test set with 2200 images from the Kaggle database labelled by the professionals. A dataset from the diabetic retinopathy (DR) screening initiative with 9653 ungradable retinal images and 11,347 gradable images was used and graded by human graders. 18 combined unsupervised information from saliency maps and supervised information from CNN with 5 convolution layers to assess image quality. The main studies about CNN method are summarized in the following paragraphs and Table 1. As the most popular deep learning architecture, a deep convolutional neural network (CNN) can automatically identify and extract hidden or latent features inherent in the input images with no need to define hand-crafted features 17, and shows superior performance than traditional machine learning methods. Recently, some research focused on using the deep learning (DL) approach to assess image quality on color fundus retinal images. However, despite low computational complexity, traditional methods requiring human intervention only can identify some characteristics of image quality and have poor generalization to other datasets. It has been investigated for decades to improve the automatic mage quality assessment, such as clarity assessment techniques including spatial techniques 3, 4, 5, 6, 7, 8, 9, 10, 11 and wavelet transform (WT) techniques 12, 13, 14, 15, 16. Poor images are the main reason for decreasing the accuracy of retinopathy detection 2. Image quality, in real-world settings, is a significant aspect for diagnosis because the proportion of poor-quality images has been reported to reach up to 19.7% in non-mydriatic retinal photography 1. It suggested that the ARIA can be used as a screening tool in the preliminary stage of retinopathy grading by telemedicine or artificial intelligence analysis. This study provides a novel angle for image quality screening based on the different poor quality types and corresponding dealing methods. The proposed approach, ARIA, showed good performance in testing, 10-fold cross validation and external validation. In external validation, our method achieved an area under the ROC curve of 0.997 for the overall quality classification and 0.915 for the classification of two types of poor quality. The sensitivity, specificity, and accuracy for testing good quality against poor quality were 98.0%, 99.1%, and 98.6%, and for differentiating between eye-abnormality-associated-poor-quality and artefact-associated-poor-quality were 92.2%, 93.8%, and 93.0%, respectively. We also analyzed the external validation with the clinical diagnosis of eye abnormality as the reference standard to evaluate the performance of the method. A total of 2434 retinal images, including 1439 good quality and 995 poor quality (483 eye-abnormality-associated-poor-quality and 512 artefact-associated-poor-quality), were used for training, testing, and 10-ford cross-validation. This study developed an automatic retinal image analysis (ARIA) method, incorporating transfer net ResNet50 deep network with the automatic features generation approach to automatically assess image quality, and distinguish eye-abnormality-associated-poor-quality from artefact-associated-poor-quality on color fundus retinal images. However, most studies focused on the classification of good and poor quality without considering the different types of poor quality. Image quality assessment is essential for retinopathy detection on color fundus retinal image.
0 kommentar(er)
