DeepUWF: An Automated Ultra-Wide-Field Fundus Screening System via Deep Learning

Zhang, Wei; Zhao, Xiujuan; Chen, Yuanyuan; Zhong, Jie; Zhang, Yi

doi:10.1109/jbhi.2020.3046771

Cited by 17 publications

(9 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are still relatively few UWFI-based multidisease AI models for the fundus [17][18][19]. Recently, in parallel to our present study, Cao et al [19] innovatively constructed a four-hierarchical interpretable eye diseases screening system using UWFIs.…”

Section: Human Doctors Versus Deep-learning Modelsmentioning

confidence: 53%

“…Antaki et al [17] applied automatic machine learning in Google Cloud AutoML Vision to implement a multiclassification task for UWFIs. Zhang et al [18] improved the performance of deep learning models in classifying UWFIs by optimizing image preprocessing. However, in both of these studies, only normal eyes and three fundus diseases were identified.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for the Detection of Multiple Fundus Diseases Using Ultra-widefield Images

Sun

Wang

Xu³

et al. 2022

Ophthalmol Ther

View full text Add to dashboard Cite

Introduction:To design and evaluate a deep learning model based on ultra-widefield images (UWFIs) that can detect several common fundus diseases. Methods: Based on 4574 UWFIs, a deep learning model was trained and validated that can identify normal fundus and eight common fundus diseases, namely referable diabetic retinopathy, retinal vein occlusion, pathologic myopia, retinal detachment, retinitis pigmentosa, age-related macular degeneration, vitreous opacity, and optic neuropathy. The model was tested on three test sets with data volumes of 465, 979, and 525. The performance of the three deep learning networks, EfficientNet-B7, Den-seNet, and ResNet-101, was evaluated on the internal test set. Additionally, we compared the performance of the deep learning model with that of doctors in a tertiary referral hospital. Results: Compared to the other two deep learning models, EfficientNet-B7 achieved the best performance. The area under the receiver operating characteristic curves of the Effi-cientNet-B7 model on the internal test set, external test set A and external test set B were 0.9708 (0.8772, 0.9849) to 1.0000 (1.0000, 1.0000), 0.9683 (0.8829, 0.9770) to 1.0000 (0.9975, 1.0000), and 0.8919 (0.7150, 0.9055) to 0.9977 (0.9165, 1.0000), respectively. On a data set of 100 images, the total accuracy of the deep learning model was 93.00%, the average Gongpeng Sun and Xiaoling Wang contributed equally.

show abstract

Section: Human Doctors Versus Deep-learning Modelsmentioning

confidence: 53%

Section: Introductionmentioning

confidence: 99%

Deep Learning for the Detection of Multiple Fundus Diseases Using Ultra-widefield Images

Sun

Wang

Xu³

et al. 2022

Ophthalmol Ther

View full text Add to dashboard Cite

show abstract

“…In the next phase, multi-object adaptive CNN, called AdaResU-Net [27], was proposed with the ability to automatically adapt to new datasets and residual learning paradigms. U-Net++ [28], a U-Netbased model, was also used on high-resolution CT images for MM detection. Moreover, the MM's detection has been tested with many variants of transfer learning [29] algorithms.…”

Section: Literature Reviewmentioning

confidence: 99%

A Semantic Adversarial Network for Detection and Classification of Myopic燤aculopathy

Abbas¹,

Baig²,

Hussain³

2023

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

The diagnosis of eye disease through deep learning (DL) technology is the latest trend in the field of artificial intelligence (AI). Especially in diagnosing pathologic myopia (PM) lesions, the implementation of DL is a difficult task because of the classification complexity and definition system of PM. However, it is possible to design an AI-based technique that can identify PM automatically and help doctors make relevant decisions. To achieve this objective, it is important to have adequate resources such as a high-quality PM image dataset and an expert team. The primary aim of this research is to design and train the DLs to automatically identify and classify PM into different classes. In this article, we have developed a new class of DL models (SAN-FSL) for the segmentation and detection of PM through semantic adversarial networks (SAN) and few-short learning (FSL) methods, respectively. Compared to DL methods, the conventional segmentation methods use supervised learning models, so they (a) require a lot of data for training and (b) fixed weights are used after the completion of the training process. To solve such problems, the FSL technique was employed for model training with few samples. The ability of FSL learning in UNet architectures is being explored, and to fine-tune the weights, a few new samples are being provided to the UNet. The outcomes show improvement in the detection area and classification of PM stages. Betterment in the result is observed by sensitivity (SE) of 95%, specificity (SP) of 96%, and area under the receiver operating curve (AUC) of 98%, and the higher F1-score is achieved using 10-fold crossvalidation. Furthermore, the obtained results confirmed the superiority of the SAN-FSL method.

show abstract

“…The second approach is to classify DR into multiple categories according to the severity of Kanth et al (2013) proposed applying "gray scale conversion, " "histogram equalization, " "application of digital filters, " "gradient magnetics segmentation, " and "finally fuzzy c clustering" to extract three features including the sum, average, and sum of exudates of the white pixels with a value of "1" in the binary image, and then used the multi-layer perceptron to classify images. Zhang et al (2021) developed an early fundus abnormality screening system (DeepUWF). Firstly, six image pre-processing techniques were used to enhance the image, and then the image was input into the CNN for image classification.…”

Section: Dichotomymentioning

confidence: 99%

DR-IIXRN : Detection Algorithm of Diabetic Retinopathy Based on Deep Ensemble Learning and Attention Mechanism

Huang

Fan

et al. 2021

Front. Neuroinform.

View full text Add to dashboard Cite

Diabetic retinopathy (DR) is one of the common chronic complications of diabetes and the most common blinding eye disease. If not treated in time, it might lead to visual impairment and even blindness in severe cases. Therefore, this article proposes an algorithm for detecting diabetic retinopathy based on deep ensemble learning and attention mechanism. First, image samples were preprocessed and enhanced to obtain high quality image data. Second, in order to improve the adaptability and accuracy of the detection algorithm, we constructed a holistic detection model DR-IIXRN, which consists of Inception V3, InceptionResNet V2, Xception, ResNeXt101, and NASNetLarge. For each base classifier, we modified the network model using transfer learning, fine-tuning, and attention mechanisms to improve its ability to detect DR. Finally, a weighted voting algorithm was used to determine which category (normal, mild, moderate, severe, or proliferative DR) the images belonged to. We also tuned the trained network model on the hospital data, and the real test samples in the hospital also confirmed the advantages of the algorithm in the detection of the diabetic retina. Experiments show that compared with the traditional single network model detection algorithm, the auc, accuracy, and recall rate of the proposed method are improved to 95, 92, and 92%, respectively, which proves the adaptability and correctness of the proposed method.

show abstract

DeepUWF: An Automated Ultra-Wide-Field Fundus Screening System via Deep Learning

Cited by 17 publications

References 32 publications

Deep Learning for the Detection of Multiple Fundus Diseases Using Ultra-widefield Images

Deep Learning for the Detection of Multiple Fundus Diseases Using Ultra-widefield Images

A Semantic Adversarial Network for Detection and Classification of Myopic燤aculopathy

DR-IIXRN : Detection Algorithm of Diabetic Retinopathy Based on Deep Ensemble Learning and Attention Mechanism

Contact Info

Product

Resources

About