Dataset Information

Generalizability of Deep Neural Networks for Vertical Cup-to-Disc Ratio Estimation in Ultra-Widefield and Smartphone-Based Fundus Images.

ABSTRACT:

Purpose

To develop and validate a deep learning system (DLS) for estimation of vertical cup-to-disc ratio (vCDR) in ultra-widefield (UWF) and smartphone-based fundus images.

Methods

A DLS consisting of two sequential convolutional neural networks (CNNs) to delineate optic disc (OD) and optic cup (OC) boundaries was developed using 800 standard fundus images from the public REFUGE data set. The CNNs were tested on 400 test images from the REFUGE data set and 296 UWF and 300 smartphone-based images from a teleophthalmology clinic. vCDRs derived from the delineated OD/OC boundaries were compared with optometrists' annotations using mean absolute error (MAE). Subgroup analysis was conducted to study the impact of peripapillary atrophy (PPA), and correlation study was performed to investigate potential correlations between sectoral CDR (sCDR) and retinal nerve fiber layer (RNFL) thickness.

Results

The system achieved MAEs of 0.040 (95% CI, 0.037-0.043) in the REFUGE test images, 0.068 (95% CI, 0.061-0.075) in the UWF images, and 0.084 (95% CI, 0.075-0.092) in the smartphone-based images. There was no statistical significance in differences between PPA and non-PPA images. Weak correlation (r = -0.4046, P < 0.05) between sCDR and RNFL thickness was found only in the superior sector.

Conclusions

We developed a deep learning system that estimates vCDR from standard, UWF, and smartphone-based images. We also described anatomic peripapillary adversarial lesion and its potential impact on OD/OC delineation.

Translational relevance

Artificial intelligence can estimate vCDR from different types of fundus images and may be used as a general and interpretable screening tool to improve community reach for diagnosis and management of glaucoma.

SUBMITTER: Yap BP

PROVIDER: S-EPMC10996969 | biostudies-literature | 2024 Apr

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Generalizability of Deep Neural Networks for Vertical Cup-to-Disc Ratio Estimation in Ultra-Widefield and Smartphone-Based Fundus Images.

Yap Boon Peng BP Kelvin Li Zhenghao LZ Toh En Qi EQ Low Kok Yao KY Rani Sumaya Khan SK Goh Eunice Jin Hui EJH Hui Vivien Yip Cherng VYC Ng Beng Koon BK Lim Tock Han TH

Translational vision science & technology 20240401 4

<h4>Purpose</h4>To develop and validate a deep learning system (DLS) for estimation of vertical cup-to-disc ratio (vCDR) in ultra-widefield (UWF) and smartphone-based fundus images.<h4>Methods</h4>A DLS consisting of two sequential convolutional neural networks (CNNs) to delineate optic disc (OD) and optic cup (OC) boundaries was developed using 800 standard fundus images from the public REFUGE data set. The CNNs were tested on 400 test images from the REFUGE data set and 296 UWF and 300 smartph ...[more]

PMID: 38568608

Similar Datasets

Project description:Glaucoma is the leading cause of irreversible blindness worldwide. Often asymptomatic for years, this disease can progress significantly before patients become aware of the loss of visual function. Critical examination of the optic nerve through ophthalmoscopy or using fundus images is a crucial component of glaucoma detection before the onset of vision loss. The vertical cup-to-disc ratio (VCDR) is a key structural indicator for glaucoma, as thinning of the superior and inferior neuroretinal rim is a hallmark of the disease. However, manual assessment of fundus images is both time-consuming and subject to variability based on clinician expertise and interpretation. In this study, we develop a robust and accurate automated system employing deep learning (DL) techniques, specifically the YOLOv7 architecture, for the detection of optic disc and optic cup in fundus images and the subsequent calculation of VCDR. We also address the often-overlooked issue of adapting a DL model, initially trained on a specific population (e.g., European), for VCDR estimation in a different population. Our model was initially trained on ten publicly available datasets and subsequently fine-tuned on the REFUGE dataset, which comprises images collected from Chinese patients. The DL-derived VCDR displayed exceptional accuracy, achieving a Pearson correlation coefficient of 0.91 (P = 4.12 × 10-412) and a mean absolute error (MAE) of 0.0347 when compared to assessments by human experts. Our models also surpassed existing approaches on the REFUGE dataset, demonstrating higher Dice similarity coefficients and lower MAEs. Moreover, we developed an optimization approach capable of calibrating DL results for new populations. Our novel approaches for detecting optic discs and optic cups and calculating VCDR, offers clinicians a promising tool that significantly reduces manual workload in image assessment while improving both speed and accuracy. Most importantly, this automated method effectively differentiates between glaucoma and non-glaucoma cases, making it a valuable asset for glaucoma detection.

Project description:PurposeTo compare the relative number of retinal pixels and retinal area imaged using the Optos P200DTx (Optos PLC) and Zeiss Clarus 500 (Carl Zeiss Meditec AG) ultra-widefield (UWF) fundus cameras.DesignSingle-center retrospective cross-sectional analysis.ParticipantsSeventy-eight eyes of 46 patients.MethodsEyes were imaged with Optos P200DTx, single-capture, and Zeiss Clarus 500, 2 capture montages when possible, UWF fundus cameras. Relative number of pixels encompassing all foveal-centered retinal quadrants were measured. Retinal area was measured with Zeiss Clarus 500 images that were registered to the Optos P200DTx images. Patients and technicians were asked for preferences between the machines. Imaging session times were recorded.Main outcome measuresRelative number of retinal pixels and retina area captured by each fundus camera.ResultsOptos P200DTx consistently captured more relative pixels compared with Zeiss Clarus 500: 510.4 versus 355.6 (P < 0.001) in total with a similarly statistically significant trend in all 4 quadrants (P < 0.001 for each). For area calculation, 70 of the 78 images achieved successful registration. Optos captured a larger total retinal area: 765.6 versus 566.5 mm2 (P < 0.001) with a similarly statistically significant trend in all 4 quadrants. In the subset of 52 of 70 registered and montaged Zeiss Clarus 500 images, similar results were found. For peripheral pathologic features, Optos P200DTx captured unique findings in 28 images, and Zeiss Clarus 500 captured unique findings 1 image (P < 0.001). Among the 48 imaging sessions in which technicians preferred Optos P200DTx for 28 sessions (58%) and Zeiss Clarus 500 for 20 (42%; P = 0.15). Among patients who responded with a preference, 24 preferred Optos P200DTx and 20 preferred Zeiss Clarus 500 (P = 0.52). Average imaging session time was 4.6 minutes (standard deviation, 3.0 minutes) for Optos P200DTx and 5.2 minutes (standard deviation, 3.0 minutes) for Zeiss Clarus 500 (P = 0.17).ConclusionsIn the current study, the Optos P200DTx captured statistically significantly more retinal area in all 4 quadrants compared with the Zeiss Clarus 500. No statistically significant difference was found in patient or technician preference or image acquisition time between devices.

Project description:IntroductionCompared with traditional fundus examination techniques, ultra-widefield fundus (UWF) images provide 200° panoramic images of the retina, which allows better detection of peripheral retinal lesions. The advent of UWF provides effective solutions only for detection but still lacks efficient diagnostic capabilities. This study proposed a retinal lesion detection model to automatically locate and identify six relatively typical and high-incidence peripheral retinal lesions from UWF images which will enable early screening and rapid diagnosis.MethodsA total of 24,602 augmented ultra-widefield fundus images with labels corresponding to 6 peripheral retinal lesions and normal manifestation labelled by 5 ophthalmologists were included in this study. An object detection model named You Only Look Once X (YOLOX) was modified and trained to locate and classify the six peripheral retinal lesions including rhegmatogenous retinal detachment (RRD), retinal breaks (RB), white without pressure (WWOP), cystic retinal tuft (CRT), lattice degeneration (LD), and paving-stone degeneration (PSD). We applied coordinate attention block and generalized intersection over union (GIOU) loss to YOLOX and evaluated it for accuracy, sensitivity, specificity, precision, F1 score, and average precision (AP). This model was able to show the exact location and saliency map of the retinal lesions detected by the model thus contributing to efficient screening and diagnosis.ResultsThe model reached an average accuracy of 96.64%, sensitivity of 87.97%, specificity of 98.04%, precision of 87.01%, F1 score of 87.39%, and mAP of 86.03% on test dataset 1 including 248 UWF images and reached an average accuracy of 95.04%, sensitivity of 83.90%, specificity of 96.70%, precision of 78.73%, F1 score of 81.96%, and mAP of 80.59% on external test dataset 2 including 586 UWF images, showing this system performs well in distinguishing the six peripheral retinal lesions.ConclusionFocusing on peripheral retinal lesions, this work proposed a deep learning model, which automatically recognized multiple peripheral retinal lesions from UWF images and localized exact positions of lesions. Therefore, it has certain potential for early screening and intelligent diagnosis of peripheral retinal lesions.

Project description:ImportanceThere is no widespread effective treatment to halt the progression of retinitis pigmentosa. Consequently, adequate assessment and estimation of residual visual function are important clinically.ObjectiveTo examine whether deep learning can accurately estimate the visual function of patients with retinitis pigmentosa by using ultra-widefield fundus images obtained on concurrent visits.Design, setting, and participantsData for this multicenter, retrospective, cross-sectional study were collected between January 1, 2012, and December 31, 2018. This study included 695 consecutive patients with retinitis pigmentosa who were examined at 5 institutions. Each of the 3 types of input images-ultra-widefield pseudocolor images, ultra-widefield fundus autofluorescence images, and both ultra-widefield pseudocolor and fundus autofluorescence images-was paired with 1 of the 31 types of ensemble models constructed from 5 deep learning models (Visual Geometry Group-16, Residual Network-50, InceptionV3, DenseNet121, and EfficientNetB0). We used 848, 212, and 214 images for the training, validation, and testing data, respectively. All data from 1 institution were used for the independent testing data. Data analysis was performed from June 7, 2021, to December 5, 2022.Main outcomes and measuresThe mean deviation on the Humphrey field analyzer, central retinal sensitivity, and best-corrected visual acuity were estimated. The image type-ensemble model combination that yielded the smallest mean absolute error was defined as the model with the best estimation accuracy. After removal of the bias of including both eyes with the generalized linear mixed model, correlations between the actual values of the testing data and the estimated values by the best accuracy model were examined by calculating standardized regression coefficients and P values.ResultsThe study included 1274 eyes of 695 patients. A total of 385 patients were female (55.4%), and the mean (SD) age was 53.9 (17.2) years. Among the 3 types of images, the model using ultra-widefield fundus autofluorescence images alone provided the best estimation accuracy for mean deviation, central sensitivity, and visual acuity. Standardized regression coefficients were 0.684 (95% CI, 0.567-0.802) for the mean deviation estimation, 0.697 (95% CI, 0.590-0.804) for the central sensitivity estimation, and 0.309 (95% CI, 0.187-0.430) for the visual acuity estimation (all P < .001).Conclusions and relevanceResults of this study suggest that the visual function estimation in patients with retinitis pigmentosa from ultra-widefield fundus autofluorescence images using deep learning might help assess disease progression objectively. Findings also suggest that deep learning models might monitor the progression of retinitis pigmentosa efficiently during follow-up.

Dataset Information

Generalizability of Deep Neural Networks for Vertical Cup-to-Disc Ratio Estimation in Ultra-Widefield and Smartphone-Based Fundus Images.

Purpose

Methods

Results

Conclusions

Translational relevance

Publications

Generalizability of Deep Neural Networks for Vertical Cup-to-Disc Ratio Estimation in Ultra-Widefield and Smartphone-Based Fundus Images.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets