Evolution of optic nerve photography for glaucoma screening: a review

Myers, Jonathan S.; Fudemberg, Scott J.; Lee, Dong Hoon

doi:10.1111/ceo.13138

Cited by 47 publications

(39 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the present study, the PPV for fundus photography with two or more glaucomatous optic features in the detection of glaucoma was 83.9% (78/93), consistent with previous reports that the overall accuracy of non‐mydriatic fundus photography for the detection of glaucoma in diabetic patients was about 80% . Availability of portable devices create opportunities for practical and affordable glaucoma screening in underserved regions by non‐physicians . Recent advances in artificial intelligence algorithms and telemedicine models also offer the promise of more rapid and accurate evaluation of optic nerve photographs in a large populations, potentially further improving access to glaucoma care .…”

Section: Discussionsupporting

confidence: 90%

“…25 Availability of portable devices create opportunities for practical and affordable glaucoma screening in underserved regions by non-physicians. 26 Recent advances in artificial intelligence algorithms Healthcare system perspective was used for the cost assessment. All costs are in US dollars at the average 2016 exchange rate (1 US dollar = 6.64 RMB) and changed into 2019 cost using an inflation rate of 2% base on the 2016-2019 Consumer Price Index of China.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Integrating opportunistic glaucoma screening into general health examinations in China: A pilot study

Zhang

Sun

Liu

et al. 2019

Clinical Exper Ophthalmology

View full text Add to dashboard Cite

Importance Under‐detection and late diagnosis are major causes of glaucoma‐related visual impairment. Cost‐effective opportunistic glaucoma screening is of great interest in the early identification and prevention of glaucoma. Background To describe the results of a health examination centre‐based opportunistic glaucoma screening and referral model. Design This single centre cross‐sectional study was conducted in a health examination centre affiliated to a tertiary hospital in Shenyang, northeastern China. Participants From 21 March to 30 September 2016, 14 367 individuals aged ≥ 30 years undergoing routine physical examinations were invited for this glaucoma screening. Methods Presenting visual acuity, non‐contact pneumotonometry and non‐mydriatic fundus photography were evaluated. Fundus photographs were classified as non‐glaucoma, possible, probable and definitive glaucoma. Participants with probable and definite glaucomatous discs or intraocular pressure ≥ 24 mmHg were referred for definitive examinations. Main Outcome Measures Detection rate of glaucoma suspects and ocular hypertension (OHT). Cost to identify a single case with suspected and diagnosed glaucoma was also calculated. Results Altogether, 277 glaucoma suspects and 327 ocular hypertension suspects were identified. Among 190 participants with probable/definite glaucomatous discs, 93 (48.9%) accepted further examination. Among these, 78 were diagnosed as glaucoma, seven as suspects and eight were excluded. Only 98 ocular hypertension suspects (30.0%) accepted further examinations: eight had primary angle closure and 23 had confirmed ocular hypertension. The cost to identify a single glaucoma suspect and definite glaucoma case were US$135 and US$857, respectively. Conclusions and Relevance This novel screening model provides opportunities to improve glaucoma detection at low cost. Interventions to improve follow‐up are needed.

show abstract

Section: Discussionsupporting

confidence: 90%

Section: Discussionmentioning

confidence: 99%

Integrating opportunistic glaucoma screening into general health examinations in China: A pilot study

Zhang

Sun

Liu

et al. 2019

Clinical Exper Ophthalmology

View full text Add to dashboard Cite

show abstract

“…However, on those original ODPs, a large percentage (142/600, 23.7%) of images was misdiagnosed by the ophthalmologists. This might have been owed to the fact that, even with high-resolution, good-quality ODPs, there is often limited detectability of small and indistinct pathologies due to insufficient time, fatigue, and/or lack of experience [7]. We demonstrated that by use of enhanced ODPs, detection accuracy for ONH pathology can be greatly improved.…”

Section: Plos Onementioning

confidence: 94%

“…In real clinical practice, limited patient cooperation, small pupils, or media opacities can limit the performance of ODP [7]. As a result, ODPs can have several limitations, such as insufficient resolution, low color contrast, and inconsistency of image quality (especially in cases of media opacity due to cataracts).…”

Section: Introductionmentioning

confidence: 99%

Deep-learning-based enhanced optic-disc photography

Sun

Kim

et al. 2020

PLoS ONE

View full text Add to dashboard Cite

Optic-disc photography (ODP) has proven to be very useful for optic nerve evaluation in glaucoma. In real clinical practice, however, limited patient cooperation, small pupils, or media opacities can limit the performance of ODP. The purpose of this study was to propose a deep-learning approach for increased resolution and improved legibility of ODP by contrast, color, and brightness compensation. Each high-resolution original ODP was transformed into two counterparts: (1) down-scaled 'low-resolution ODPs', and (2) 'compensated high-resolution ODPs' produced via enhancement of the visibility of the optic disc margin and surrounding retinal vessels using a customized image post-processing algorithm. Then, the differences between these two counterparts were directly learned through a super-resolution generative adversarial network (SR-GAN). Finally, by inputting the high-resolution ODPs into SR-GAN, 4-times-up-scaled and overall-color-and-brightness-transformed 'enhanced ODPs' could be obtained. General ophthalmologists were instructed (1) to assess each ODP's image quality, and (2) to note any abnormal findings, at 1-month intervals. The image quality score for the enhanced ODPs was significantly higher than that for the original ODP, and the overall optic disc hemorrhage (DH)-detection accuracy was significantly higher with the enhanced ODPs. We expect that this novel deep-learning approach will be applied to various types of ophthalmic images.

show abstract

“…Optic disc photography remains the fundamental approach in glaucoma screening [8,9]. Disc photography reviewed by experienced specialists can achieve similar diagnostic performance as that of advanced imaging technology such as optical coherence tomography (OCT) [10], and combining other modalities with fundus images did not necessarily increase the diagnostic power in teleglaucoma [7].…”

Section: Introductionmentioning

confidence: 99%

Deep learning assisted detection of glaucomatous optic neuropathy and potential designs for a generalizable model

Wey

Chen

et al. 2020

PLoS ONE

View full text Add to dashboard Cite

To evaluate ways to improve the generalizability of a deep learning algorithm for identifying glaucomatous optic neuropathy (GON) using a limited number of fundus photographs, as well as the key features being used for classification. Methods A total of 944 fundus images from Taipei Veterans General Hospital (TVGH) were retrospectively collected. Clinical and demographic characteristics, including structural and functional measurements of the images with GON, were recorded. Transfer learning based on VGGNet was used to construct a convolutional neural network (CNN) to identify GON. To avoid missing cases with advanced GON, an ensemble model was adopted in which a support vector machine classifier would make final classification based on cup-to-disc ratio if the CNN classifier had low-confidence score. The CNN classifier was first established using TVGH dataset, and then fine-tuned by combining the training images of TVGH and Drishti-GS datasets. Class activation map (CAM) was used to identify key features used for CNN classification. Performance of each classifier was determined through area under receiver operating characteristic curve (AUC) and compared with the ensemble model by diagnostic accuracy. Results In 187 TVGH test images, the accuracy, sensitivity, and specificity of the CNN classifier were 95.0%, 95.7%, and 94.2%, respectively, and the AUC was 0.992 compared to the 92.8% accuracy rate of the ensemble model. For the Drishti-GS test images, the accuracy of the CNN, the fine-tuned CNN and ensemble model was 33.3%, 80.3%, and 80.3%,

show abstract

Evolution of optic nerve photography for glaucoma screening: a review

Cited by 47 publications

References 64 publications

Integrating opportunistic glaucoma screening into general health examinations in China: A pilot study

Integrating opportunistic glaucoma screening into general health examinations in China: A pilot study

Deep-learning-based enhanced optic-disc photography

Deep learning assisted detection of glaucomatous optic neuropathy and potential designs for a generalizable model

Contact Info

Product

Resources

About