AI Article Synopsis
- The study aimed to develop a digital ray technique using GANs to improve the quality of postoperative cataract images for better detection of retinopathy.
- Researchers captured and analyzed preoperative and postoperative color fundus photographs (CFP) and ultra-wide field (UWF) images from cataract patients, comparing the performance of an original CycleGAN and a modified version.
- Results showed that the modified CycleGAN significantly enhanced image quality, leading to increased detection accuracy of retinopathy from 78% to 91% for CFP images, and even higher for UWF images, demonstrating the method's effectiveness for clinical diagnostics.
Article Abstract
Background/aims: The aim of this study was to develop and evaluate digital ray, based on preoperative and postoperative image pairs using style transfer generative adversarial networks (GANs), to enhance cataractous fundus images for improved retinopathy detection.
Methods: For eligible cataract patients, preoperative and postoperative colour fundus photographs (CFP) and ultra-wide field (UWF) images were captured. Then, both the original CycleGAN and a modified CycleGAN (CycleGAN) framework were adopted for image generation and quantitatively compared using Frechet Inception Distance (FID) and Kernel Inception Distance (KID). Additionally, CFP and UWF images from another cataract cohort were used to test model performances. Different panels of ophthalmologists evaluated the quality, authenticity and diagnostic efficacy of the generated images.
Results: A total of 959 CFP and 1009 UWF image pairs were included in model development. FID and KID indicated that images generated by CycleGAN presented significantly improved quality. Based on ophthalmologists' average ratings, the percentages of inadequate-quality images decreased from 32% to 18.8% for CFP, and from 18.7% to 14.7% for UWF. Only 24.8% and 13.8% of generated CFP and UWF images could be recognised as synthetic. The accuracy of retinopathy detection significantly increased from 78% to 91% for CFP and from 91% to 93% for UWF. For retinopathy subtype diagnosis, the accuracies also increased from 87%-94% to 91%-100% for CFP and from 87%-95% to 93%-97% for UWF.
Conclusion: Digital ray could generate realistic postoperative CFP and UWF images with enhanced quality and accuracy for overall detection and subtype diagnosis of retinopathies, especially for CFP.\ TRIAL REGISTRATION NUMBER: This study was registered with ClinicalTrials.gov (NCT05491798).
Download full-text PDF |
Source |
|---|---|
| http://www.ncbi.nlm.nih.gov/pmc/articles/PMC11503040 | PMC |
| http://dx.doi.org/10.1136/bjo-2024-325403 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Ultra‐widefield retinal and choroidal vascular architecture in mild cognitive impairment and normal cognition.
Alzheimers Dement
December 2024
Duke Eye Center, Durham, NC, USA
Background: Mild cognitive impairment (MCI) is currently a clinical diagnosis characterized by decline in memory and daily cognitive function from baseline. Exploratory studies using optical coherence tomography angiography have reported alterations in the retinal capillary plexus vessel density and attenuation of the retinal nerve fiber layer, but these results appear to be mixed. We used ultra‐widefield (UWF) imaging to evaluate retinal and choroidal vasculature and structure in individuals with mild cognitive impairment (MCI) compared to controls with normal cognition.
View Article and Find Full Text PDFA Convolutional Neural Network to Identify Parkinson’s Disease using Multimodal Retinal Images.
Alzheimers Dement
December 2024
Duke Eye Center, Durham, NC, USA
Background: Retinal structure and microvasculature may be used as a surrogate for parallel processes in the brain. Previous studies have revealed differences in retinal structure and microvasculature in patients with Parkinson’s disease (PD) compared to cognitively normal controls[1]. Previous work developed a convolutional neural network (CNN) trained on multimodal retinal images that was able to identify Alzheimer’s dementia with an area under the curve (AUC) of 0.
View Article and Find Full Text PDFQuantitative analysis of choroidal vascular structures and anatomical changes in pachychoroid spectrum diseases using ultra-widefield SS-OCTA.
Sci Rep
January 2025
The Department of Ophthalmology, General Hospital of Central Theater Command, No. 627 Wuluo Road, Wuchang District, Wuhan, 430000, Hubei, China.
This study used ultra-widefield swept-source optical coherence tomography angiography (UWF SS-OCTA) to analyze and compare choroidal blood flow and anatomical changes in eyes affected by central serous chorioretinopathy (CSC), pachychoroid neovasculopathy (PNV), and uncomplicated pachychoroid (UCP). The findings revealed distribution patterns of vortex veins across the three patient groups and provided initial findings insights into the origin of choroidal neovascularization (CNV) in PNV. A total of 44 patients with CSC, 38 with PNV, and 46 with UCP were included in the analysis.
View Article and Find Full Text PDFQuantitative ultra-widefield fluorescein angiography biomarkers in diabetic retinopathy and association with treatment and progression.
Clin Ophthalmol
December 2024
Department of Ophthalmology and Visual Sciences, W. K. Kellogg Eye Center, University of Michigan Medical School, Ann Arbor, MI, USA.
Article Synopsis
- This study aimed to explore how demographic factors and measurements of nonperfusion (NP) and neovascularization (NV) on ultra-widefield fluorescein angiography relate to intravitreal injections (IVIs) and panretinal photocoagulation (PRP) treatment in diabetic patients.
- It included 363 diabetic patients treated at the University of Michigan, analyzing the effectiveness of these treatments based on retinal image data from 2009 to 2018 while excluding those with previous PRP treatment or poor image quality.
- Results revealed that higher NP and NV areas were associated with increased IVIs and PRP treatments, and type 2 diabetes was linked to a greater risk of progression in diabetic retinopathy compared to type 1
Ultra-Widefield Optical Coherence Tomography Beyond the Ora Serrata in Retinopathy of Prematurity.
JAMA Ophthalmol
December 2024
Casey Eye Institute, Oregon Health and Science University, Portland.
Importance: Capturing high-quality images of the entire peripheral retina while minimizing the use of scleral depression could increase the quality of examinations for retinopathy of prematurity (ROP) while reducing neonatal stress.
Objective: To evaluate whether an investigational handheld ultra-widefield optical coherence tomography (UWF-OCT) device without scleral depression can be used to document high-quality images of the peripheral retina for use in ROP examinations.
Design, Setting, And Participants: This was a prospective, cross-sectional study in the neonatal intensive care unit at a single academic medical center.
Want AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!