Modeling and Enhancing Low-Quality Retinal Fundus Images

Shen, Ziyi; Fu, Huazhu; Shen, Jianbing; Shao, Ling

doi:10.1109/tmi.2020.3043495

Cited by 117 publications

(110 citation statements)

References 47 publications

Supporting

Mentioning

110

Contrasting

Order By: Relevance

“…Operating a smartphone-based fundus camera requires some skills that non-specialists in ophthalmology may lack. Variability between operators of a medical device may compromise screening results [49]. Considering 3 EyeFundusScope operators, k of 0.8, k maximum amplitude of 0.2, diabetic retinopathy prevalence of 20% and 95% confidence intervals, the sample size estimate for reproducibility calculations is 200 individuals with diabetes for the study of interoperator reproducibility, and 200 individuals with diabetes for the study of intraoperator reproducibility.…”

Section: Inter and Intraoperator Agreement And Reliabilitymentioning

confidence: 99%

“…Furthermore, the proportion of ungradable images in people with central cataracts accounts for 57% of ungradable images, and the more the years after diagnosis of diabetes, the higher is the proportion of ungradable images [48]. Moreover, the skill of each operator influences the quality of images, and although we expect it to vary for each operator, we need to assess if these differences affect the AI classification for diabetic retinopathy [49].…”

Section: Introduction 1background and Rationalementioning

confidence: 99%

See 1 more Smart Citation

Implementation and Evaluation of a Mobile Retinal Image Acquisition System for Screening Diabetic Retinopathy: Study Protocol

et al. 2022

View full text Add to dashboard Cite

Screening diabetic retinopathy, a major cause of blindness, is time-consuming for ophthalmologists and has some constrains in achieving full coverage and attendance. The handheld fundus camera EyeFundusScope was recently developed to expand the scale of screening, drawing on images acquired in primary care and telescreening made by ophthalmologists or a computer-aided diagnosis (CADx) system. This study aims to assess the diagnostic accuracy of the interpretation of images captured using EyeFundusScope and perform its technical evaluation, including image quality, functionality, usability, and acceptance in a real-world clinical setting. Physicians and nurses without training in ophthalmology will use EyeFundusScope to take pictures of the retinas of patients with diabetes and the images will be classified for the presence or absence of diabetic retinopathy and image quality by a panel of ophthalmologists. A subgroup of patients will also be examined with the reference standard tabletop fundus camera. Screening results provided by the CADx system on images taken with EyeFundusScope will be compared against the ophthalmologists’ analysis of images taken with the tabletop fundus camera. Diagnostic accuracy measures with 95% confidence intervals (CIs) will be calculated for positive and negative test results. Proportion of each category of image quality will be presented. Usability and acceptance results will be presented qualitatively.

show abstract

Section: Inter and Intraoperator Agreement And Reliabilitymentioning

confidence: 99%

Section: Introduction 1background and Rationalementioning

confidence: 99%

Implementation and Evaluation of a Mobile Retinal Image Acquisition System for Screening Diabetic Retinopathy: Study Protocol

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Especially for those samples with under-exposure or slight over-exposure, it could lead to great improvement in overall visual quality that is well aligned to human perception, resulting in a better high-quality score. Some fundus degradation factors, such as light transmission disturbance [42] and absence of exposure, could impair observation and fovea localization. The illumination regularization is an appropriate way for overcoming degradation factors to a large extent, thereby improving fovea localization.…”

Section: F Ablation Studiesmentioning

confidence: 99%

Structure and Illumination Constrained GAN for Medical Image Enhancement

Liu

et al. 2021

IEEE Trans. Med. Imaging

Self Cite

111

View full text Add to dashboard Cite

The development of medical imaging techniques has greatly supported clinical decision making. However, poor imaging quality, such as non-uniform illumination or imbalanced intensity, brings challenges for automated screening, analysis and diagnosis of diseases. Previously, bi-directional GANs (e.g., CycleGAN), have been proposed to improve the quality of input images without the requirement of paired images. However, these methods focus on global appearance, without imposing constraints on structure or illumination, which are essential features for medical image interpretation. In this paper, we propose a novel and versatile bi-directional GAN, named Structure and illumination constrained GAN (StillGAN), for medical image quality enhancement. Our StillGAN treats low-and high-quality images as two distinct domains, and introduces local structure and illumination constraints for learning both overall characteristics and local details. Extensive experiments on three medical image datasets (e.g., corneal confocal microscopy, retinal color fundus and endoscopy images) demonstrate that our method performs better than both conventional methods and other deep learning-based methods. In addition, we have investigated the impact of the proposed method on different medical image analysis and clinical tasks such as nerve segmentation, tortuosity grading, fovea localization and disease classification.

show abstract

“…However, different equipments and ophthalmologists pose large variations to the quality of fundus images. A screening study of 5,575 patients found that about 12% of [13] and Cofe-Net [58], fail to recover the LQ images.…”

Section: Introductionmentioning

confidence: 99%

“…Recently, deep Convolutional Neural Networks (CNNs) have been applied to learn the end-to-end mapping function of fundus image restoration. Although impressive results have been achieved, the CNN-based methods [13,45,47,56,58,68,78] show limitations in capturing long-range dependencies. In recent years, the natural language processing (NLP) model, Transformer [63] has been introduced into computer vision and outperformed CNN methods in many tasks.…”

Section: Introductionmentioning

confidence: 99%

RFormer: Transformer-based Generative Adversarial Network for Real Fundus Image Restoration on A New Clinical Benchmark

Cai¹,

Chen²,

Gong³

et al. 2022

Preprint

View full text Add to dashboard Cite

Ophthalmologists have used fundus images to screen and diagnose eye diseases. However, different equipments and ophthalmologists pose large variations to the quality of fundus images. Low-quality (LQ) degraded fundus images easily lead to uncertainty in clinical screening and generally increase the risk of misdiagnosis. Thus, real fundus image restoration is worth studying. Unfortunately, real clinical benchmark has not been explored for this task so far. In this paper, we investigate the real clinical fundus image restoration problem. Firstly, We establish a clinical dataset, Real Fundus (RF), including 120 low-and high-quality (HQ) image pairs. Then we propose a novel Transformer-based Generative Adversarial Network (RFormer) to restore the real degradation of clinical fundus images. The key component in our network is the Window-based Self-Attention Block (WSAB) which captures non-local self-similarity and long-range dependencies. To produce more visually pleasant results, a Transformer-based discriminator is introduced. Extensive experiments on our clinical benchmark show that the proposed RFormer significantly outperforms the state-of-the-art (SOTA) methods. In addition, experiments of downstream tasks such as vessel segmentation and optic disc/cup detection demonstrate that our proposed RFormer benefits clinical fundus image analysis and applications. The dataset, code, and models will be released.

show abstract

Modeling and Enhancing Low-Quality Retinal Fundus Images

Cited by 117 publications

References 47 publications

Implementation and Evaluation of a Mobile Retinal Image Acquisition System for Screening Diabetic Retinopathy: Study Protocol

Implementation and Evaluation of a Mobile Retinal Image Acquisition System for Screening Diabetic Retinopathy: Study Protocol

Structure and Illumination Constrained GAN for Medical Image Enhancement

RFormer: Transformer-based Generative Adversarial Network for Real Fundus Image Restoration on A New Clinical Benchmark

Contact Info

Product

Resources

About