The purpose of this study was to explore the use of fluorescein angiography (FA) images in a convolutional neural network (CNN) in the management of retinopathy of prematurity (ROP). Methods: The dataset involved a total of 835 FA images of 149 eyes (90 patients), where each eye was associated with a binary outcome (57 "untreated" eyes and 92 "treated"; 308 "untreated" images, 527 "treated"). The resolution of the images was 1600 and 1200 px in 20% of cases, whereas the remaining 80% had a resolution of 640 and 480 px. All the images were resized to 640 and 480 px before training and no other preprocessing was applied. A CNN with four convolutional layers was trained on 90% of the images (n = 752) randomly chosen. The accuracy of the prediction was assessed on the remaining 10% of images (n = 83). Keras version 2.2.0 for R with Tensorflow backend version 1.11.0 was used for the analysis. Results: The validation accuracy after 100 epochs was 0.88, whereas training accuracy was 0.97. The receiver operating characteristic (ROC) presented an area under the curve (AUC) of 0.91. Conclusions: Our study showed, we believe for the first time, the applicability of artificial intelligence (CNN) technology in the ROP management driven by FA. Further studies are needed to exploit different fields of applications of this technology. Translational Relevance: This algorithm is the basis for a system that could be applied to both ROP as well as experimental oxygen induced retinopathy.
Purpose
To study the effect of statin exposure on the progression from non-exudative to exudative age-related macular degeneration (AMD).
Methods
Retrospective cohort study of commercially insured patients diagnosed with non-exudative AMD (n = 231,888) from 2007 to 2015. Time-to-event analysis of the association between exposure to lipid-lowering medications and time from non-exudative AMD to exudative AMD diagnosis was conducted. Outcome measures included progression to exudative AMD, indicated by diagnosis codes for exudative AMD or procedural codes for intravitreal injections.
Results
In the year before and after first AMD diagnosis, 11,330 patients were continuously prescribed lipid-lowering medications and 31,627 patients did not take any lipid-lowering medication. Of those taking statins, 21 (1.6%) patients were on very-high-dose lipophilic statins, 644 (47.6%) on high-dose lipophilic statins, and 689 (50.9%) on low-dose lipophilic statins. We found no statistically significant relationship between exposure to low (HR 0.89, 95% CI 0.83 to 1.38) or high-dose lipophilic statins (HR 1.12, 95% CI 0.86 to 1.45) and progression to exudative AMD. No patients taking very-high-dose lipophilic statins converted from non-exudative to exudative AMD, though this difference was not statistically significant due to the subgroup size (p = .23, log-rank test).
Conclusions
No statistically significant relationship was found between statin exposure and risk of AMD progression. Interestingly, no patients taking very-high-dose lipophilic statins progressed to exudative AMD, a finding that warrants further exploration.
BACKGROUND AND OBJECTIVE:
To investigate whether surgical management of newly diagnosed rhegmatogenous retinal detachment (RRD) varies if patients are diagnosed during a major national ophthalmology conference.
PATIENTS AND METHODS:
This retrospective cohort study included 34,759 patients with incident RRD, of whom, 1,246 (3.6%) were diagnosed during dates of three national ophthalmology conferences, and 1,170 (3.4%) underwent surgery during conference dates. The authors identified patients with primary repair with cryotherapy, laser, scleral buckle, pneumatic retinopexy, or pars plana vitrectomy. Multinomial logistic regression models were used to determine patients' likelihood of receiving each type of repair within 30 days of their diagnosis depending on whether they were diagnosed during a national ophthalmology conference. Linear regression models were used to determine the relationship between the date patients were diagnosed and how long they waited to receive a repair. Main outcome measures included days between diagnosis with RRD and RRD repair, receiving repair on the same day of diagnosis, and reoperation rate within 30 days of the primary repair.
RESULTS:
Mean time from diagnosis to repair was 1.5 days (standard deviation: ± 2.4 days), and 71% of patients underwent repair within a day of diagnosis. Repairs were followed by a second surgery within 30 days in 11.1% of patients. Patients diagnosed during conferences waited 0.23 days longer between diagnosis and repair compared with patients diagnosed outside of conference dates (
P
= .001). Patients diagnosed with RRD during conferences were less likely to receive surgical repair within a day of diagnosis compared to patients diagnosed during non-conference dates (
P
= .037). Patients who were diagnosed with RRD during a conference date and also received surgery during a conference date were more likely to undergo a second surgery within 30 days of the primary procedure (
P
= .006)
CONCLUSIONS:
Patients diagnosed with RRD during national ophthalmology conference dates waited slightly longer for surgery, were slightly less likely to receive surgery within a day, and were more likely to undergo a second surgery within 30 days of the primary procedure. The “national meeting effect” phenomenon is present in ophthalmology, albeit to a lesser degree that is likely not visually significant.
[
Ophthalmic Surg Lasers Imaging Retina
. 2020;51:328–337.
]
BACKGROUND AND OBJECTIVE:
To compare morphologic and functional status at age 4 years for patients treated in one eye with laser photocoagulation and the other eye with intravitreal bevacizumab (IVB) injection for Type 1 retinopathy of prematurity (ROP).
PATIENTS AND METHODS:
In this single-center, randomized, controlled trial, best-corrected visual acuity (BCVA) in logMAR was obtained along with spherical equivalent refraction (SER), fluorescein angiography (FA), optical coherent tomography (OCT), and OCT angiography (OCTA).
RESULTS:
Eighteen babies (36 eyes) were selected for this study. BCVA and SER were similar in the two groups, but six patients had anisometropia of 4 diopters or more. IVB-treated eyes tended to have thinner foveal thickness than laser-treated eyes (mean difference: −5.33 pixels; 95% confidence interval, −9.62 to −1.05).
CONCLUSION:
Although the differences found here are minimal between the IVB-treated and laser-treated groups, further long-term evaluation of not only FA, but also OCT and OCTA, are needed in larger studies.
[
Ophthalmic Surg Lasers Imaging Retina
. 2020;51:180–186.]
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.