Artificial intelligence for the detection of age-related macular degeneration in color fundus photographs: A systematic review and meta-analysis

Dong, Li; Yang, Qiong; Zhang, Rui Heng; Wei, Wen Bin

doi:10.1016/j.eclinm.2021.100875

Cited by 62 publications

(36 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Due to these complexities, the deep learning methods have been preferred over the conventional machine learning approaches of setting detailed criteria to satisfy each required feature. For this reason, this section only describes the deep learning methods [ 37 , 38 , 39 , 40 ].…”

Section: Machine Learning Methodsmentioning

confidence: 99%

“…Referring to the methods developed so far, the possible ways to develop such systems are by detecting referable cases, detecting the presence of disease, classifying the degree of severity, and localizing lesions. Based on [ 37 , 39 ], the development of frameworks for DR has been the most researched, followed by frameworks for AMD. Frameworks for glaucoma have not been studied as much as DR and AMD [ 37 , 39 , 40 ].…”

Section: Machine Learning Methodsmentioning

confidence: 99%

“…Based on [ 37 , 39 ], the development of frameworks for DR has been the most researched, followed by frameworks for AMD. Frameworks for glaucoma have not been studied as much as DR and AMD [ 37 , 39 , 40 ]. The DR framework mentioned in this section encompasses not only a general explanation about the developed deep learning methods, but also detailed descriptions of two methods that achieved relatively better sensitivity for the widely used public datasets shown in Table 3 [ 39 ].…”

Section: Machine Learning Methodsmentioning

confidence: 99%

“…AMD can be graded based on these lesions, but there are different schemes for the grading [ 33 , 34 ]. Many machine learning works used AREDS (Age-Related Eye Disease Study) criteria [ 37 ]. These criteria divide AMD into eight grades based on the size of the drusen, increased pigment, depigmentation, and geographic atrophy (GA).…”

Section: Domain Knowledgementioning

confidence: 99%

“…In this paper, in order to leverage the strengths of both approaches, we restricted our scope to applications of fundus imaging and their most important uses are covered by presenting detailed examples. In particular, we have selected several important, recently published reviews that cover the sub-domain tasks for major eye diseases and the structure of the eye [ 37 , 38 , 39 , 40 ]. The following section introduces domain knowledge in the field of ophthalmology, such as the imaging modality, retinal structures, and retinopathy.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Review of Machine Learning Applications Using Retinal Fundus Images

2022

View full text Add to dashboard Cite

Automating screening and diagnosis in the medical field saves time and reduces the chances of misdiagnosis while saving on labor and cost for physicians. With the feasibility and development of deep learning methods, machines are now able to interpret complex features in medical data, which leads to rapid advancements in automation. Such efforts have been made in ophthalmology to analyze retinal images and build frameworks based on analysis for the identification of retinopathy and the assessment of its severity. This paper reviews recent state-of-the-art works utilizing the color fundus image taken from one of the imaging modalities used in ophthalmology. Specifically, the deep learning methods of automated screening and diagnosis for diabetic retinopathy (DR), age-related macular degeneration (AMD), and glaucoma are investigated. In addition, the machine learning techniques applied to the retinal vasculature extraction from the fundus image are covered. The challenges in developing these systems are also discussed.

show abstract

Section: Machine Learning Methodsmentioning

confidence: 99%

Section: Machine Learning Methodsmentioning

confidence: 99%

Section: Machine Learning Methodsmentioning

confidence: 99%

Section: Domain Knowledgementioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Review of Machine Learning Applications Using Retinal Fundus Images

2022

View full text Add to dashboard Cite

show abstract

Evaluation of Generative Adversarial Networks for High-Resolution Synthetic Image Generation of Circumpapillary Optical Coherence Tomography Images for Glaucoma

et al. 2022

View full text Add to dashboard Cite

IMPORTANCE Deep learning (DL) networks require large data sets for training, which can be challenging to collect clinically. Generative models could be used to generate large numbers of synthetic optical coherence tomography (OCT) images to train such DL networks for glaucoma detection.OBJECTIVE To assess whether generative models can synthesize circumpapillary optic nerve head OCT images of normal and glaucomatous eyes and determine the usability of synthetic images for training DL models for glaucoma detection.DESIGN, SETTING, AND PARTICIPANTS Progressively growing generative adversarial network models were trained to generate circumpapillary OCT scans. Image gradeability and authenticity were evaluated on a clinical set of 100 real and 100 synthetic images by 2 clinical experts. DL networks for glaucoma detection were trained with real or synthetic images and evaluated on independent internal and external test data sets of 140 and 300 real images, respectively.MAIN OUTCOMES AND MEASURES Evaluations of the clinical set between the experts were compared. Glaucoma detection performance of the DL networks was assessed using area under the curve (AUC) analysis. Class activation maps provided visualizations of the regions contributing to the respective classifications.RESULTS A total of 990 normal and 862 glaucomatous eyes were analyzed. Evaluations of the clinical set were similar for gradeability (expert 1: 92.0%; expert 2: 93.0%) and authenticity (expert 1: 51.8%; expert 2: 51.3%). The best-performing DL network trained on synthetic images had AUC scores of 0.97 (95% CI, 0.95-0.99) on the internal test data set and 0.90 (95% CI, 0.87-0.93) on the external test data set, compared with AUCs of 0.96 (95% CI, 0.94-0.99) on the internal test data set and 0.84 (95% CI, 0.80-0.87) on the external test data set for the network trained with real images. An increase in the AUC for the synthetic DL network was observed with the use of larger synthetic data set sizes. Class activation maps showed that the regions of the synthetic images contributing to glaucoma detection were generally similar to that of real images.CONCLUSIONS AND RELEVANCE DL networks trained with synthetic OCT images for glaucoma detection were comparable with networks trained with real images. These results suggest potential use of generative models in the training of DL networks and as a means of data sharing across institutions without patient information confidentiality issues.

show abstract