A cascade eye diseases screening system with interpretability and expandability in ultra-wide field fundus images: A multicentre diagnostic accuracy study

Cao, Jing; You, Kun; Zhou, Jingxin; Xu, Mingyu; Xu, Pinghong; Wen, Lei; Wang, Shengzhan; Jin, Kai; Lou, Lixia; Wang, Yao; Ye, Juan

doi:10.1016/j.eclinm.2022.101633

Cited by 7 publications

(11 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Section: Human Doctors Versus Deep-learning Modelsmentioning

confidence: 56%

“…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. In the study, extensive image-level and lesion-level labeling was performed, the images were divided into 21 anatomical regions based on the anatomical location of the optic nerve head and macular fovea, and a lesion atlas mapping module was constructed to distinguish abnormal findings and determine their anatomical locations.…”

Section: Human Doctors Versus Deep-learning Modelsmentioning

confidence: 56%

“…However, in both of these studies, only normal eyes and three fundus diseases were identified. Recently, Cao et al [19] proposed a UWFI-based four-hierarchical interpretable eye diseases screening system that can identify 30 common fundus diseases and lesions. We also carried out some work on multidisease identification with UWFIs.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

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: 56%

Section: Human Doctors Versus Deep-learning Modelsmentioning

confidence: 56%

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

show abstract

“…Due to the relatively lower cost and irreplaceable advantages of evaluating ischemic lesions, FFA devices have commonly been used in general hospitals in recent decades. 9 , 15 However, there is only one ophthalmologist per 110,000 people in developing countries, a tenth of the ophthalmologist-to-population ratio in developed countries, 24 and there are even fewer retinal ophthalmologists. Some ophthalmologists might be able to determine whether a patient with retinal disease needs FFA examination but not be able to accurately interpret the FFA images, posing great challenges to service coverage.…”

Section: Discussionmentioning

confidence: 99%

An artificial intelligence system for the whole process from diagnosis to treatment suggestion of ischemic retinal diseases

Zhao,

Lin,

et al. 2023

Cell Reports Medicine

View full text Add to dashboard Cite

“…In related work, subjective assessment methods in ophthalmology are commonly implemented as a part of a broader evaluation of newly developed automated diagnostic methods. Applications include, but are not limited to, glaucoma detection from stereoscopic [ 17 , 18 , 19 ] and monoscopic [ 20 ] optical disk images, diabetic retinopathy [ 21 , 22 ], diabetic macular edema classification [ 23 ], age-related macular degradation [ 19 , 22 , 24 , 25 , 26 ], central/branch retinal vein occlusion [ 22 ], detachment of the retina [ 22 ], as well as cataract [ 21 , 27 ], retinopathy of prematurity classification [ 28 ], and cardiovascular disease detection [ 29 ]. Related studies address disease classification tasks, where both human observers and proposed artificial intelligence (AI) methods take the same piece of data as an input and output the same kind of information: a textual diagnosis, a bounding box marking a region of interest where a certain pattern is observed in an image, etc.…”

Section: Introductionmentioning

confidence: 99%

Comparing the Clinical Viability of Automated Fundus Image Segmentation Methods

Gojić

Petrović

Dragan

et al. 2022

Sensors

View full text Add to dashboard Cite

Recent methods for automatic blood vessel segmentation from fundus images have been commonly implemented as convolutional neural networks. While these networks report high values for objective metrics, the clinical viability of recovered segmentation masks remains unexplored. In this paper, we perform a pilot study to assess the clinical viability of automatically generated segmentation masks in the diagnosis of diseases affecting retinal vascularization. Five ophthalmologists with clinical experience were asked to participate in the study. The results demonstrate low classification accuracy, inferring that generated segmentation masks cannot be used as a standalone resource in general clinical practice. The results also hint at possible clinical infeasibility in experimental design. In the follow-up experiment, we evaluate the clinical quality of masks by having ophthalmologists rank generation methods. The ranking is established with high intra-observer consistency, indicating better subjective performance for a subset of tested networks. The study also demonstrates that objective metrics are not correlated with subjective metrics in retinal segmentation tasks for the methods involved, suggesting that objective metrics commonly used in scientific papers to measure the method’s performance are not plausible criteria for choosing clinically robust solutions.

show abstract

A cascade eye diseases screening system with interpretability and expandability in ultra-wide field fundus images: A multicentre diagnostic accuracy study

Cited by 7 publications

References 37 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

An artificial intelligence system for the whole process from diagnosis to treatment suggestion of ischemic retinal diseases

Comparing the Clinical Viability of Automated Fundus Image Segmentation Methods

Contact Info

Product

Resources

About