Sensitivity and Specificity of Ultrawide-Field Fundus Photography for the Staging of Sickle Cell Retinopathy in Real-Life Practice at Varying Expertise Level

Proliferative Sickle Cell Retinopathy in the Retinal Periphery Detected by Ultra-Widefield Imaging: A Case Report

Sickle cell proliferative retinopathy usually presents first in the peripheral retina and the ability to extend and enhance our visualization of the peripheral retina would allow for superior clinical decision-making. In our practice, we had a 28-year-old patient diagnosed with major sickle cell disease of the homozygous type SS (HbSS) that presented with sickle cell proliferative retinopathy detected by ultra-widefield imaging in the nasal side of the left fundus. At follow-up, neovascularization was detected in the extreme nasal periphery of the left eye by ultra-widefield imaging fluorescein angiography with right gaze. The case was graded as Goldberg stage 3, and the patient was administered photocoagulation treatment. With further advancements in the quality and modality of peripheral retinal imaging, novel proliferative lesions can be detected and appropriately managed much earlier than was previously possible. Ultra-widefield imaging allows for the visualization of the central 200 degrees of the retina but, with gaze view, peripheral retina beyond 200 degrees can be reached.

Comparison of Ultra-Wide Field Photography to Ultra-Wide Field Angiography for the Staging of Sickle Cell Retinopathy

Sickle cell retinopathy (SCR) is classified by Goldberg based on peripheral vascular changes. Ultra-wide field (UWF) imaging has enhanced visualization of the peripheral retina. However, there is no consensus on the optimal imaging technique for the screening of SCR. We performed a monocentric observational cross-sectional study to compare UWF fundus photography (UWF-FP) with UWF angiography (UWF-FA). All patients who underwent UWF-imaging (Optos, PLC, Scotland, UK) for screening of sickle cell retinopathy between January 2016 and December 2019 were retrospectively included. Eyes with previous laser treatment or concomitant retinal disease were excluded. UWF-FP images were graded based on the Goldberg classification by four graders with various degrees of experience. UWF-FA pictures were reviewed by an independent retina specialist. Differences in Goldberg staging across UWF-FP and UWF-FA were assessed. A total of 84 eyes of 44 patients were included. Based on UWF-FA, most eyes were stage 2 (77.4%) and 19 were stage 3 (22.6%). The pre-retinal neovascularization detection sensitivity on UWF-FP was 52.6 to 78.9%, depending on the graders. UWF-FA led to a later Goldberg stage of retinopathy, in most cases from stage 1 to stage 2. Neovascularization (stage 3) was not detected by our graders on UWF-FP in 21.1 to 57.9% of eyes. UWP-FP tends to underestimate Goldberg stages of retinopathy compared with UWF-FA and is less accurate when detecting neovascularization in sickle cell retinopathy, which has a direct impact on therapeutic management and prognosis.

Deep learning-based classification of retinal vascular diseases using ultra-widefield colour fundus photographs

Objective

To assess the ability of a deep learning model to distinguish between diabetic retinopathy (DR), sickle cell retinopathy (SCR), retinal vein occlusions (RVOs) and healthy eyes using ultra-widefield colour fundus photography (UWF-CFP).

Methods and Analysis

In this retrospective study, UWF-CFP images of patients with retinal vascular disease (DR, RVO, and SCR) and healthy controls were included. The images were used to train a multilayer deep convolutional neural network to differentiate on UWF-CFP between different vascular diseases and healthy controls. A total of 224 UWF-CFP images were included, of which 169 images were of retinal vascular diseases and 55 were healthy controls. A cross-validation technique was used to ensure that every image from the dataset was tested once. Established augmentation techniques were applied to enhance performances, along with an Adam optimiser for training. The visualisation method was integrated gradient visualisation.

Results

The best performance of the model was obtained using 10 epochs, with an overall accuracy of 88.4%. For DR, the area under the receiver operating characteristics (ROC) curve (AUC) was 90.5% and the accuracy was 85.2%. For RVO, the AUC was 91.2% and the accuracy 88.4%. For SCR, the AUC was 96.7% and the accuracy 93.8%. For healthy controls, the ROC was 88.5% with an accuracy that reached 86.2%.

Conclusion

Deep learning algorithms can classify several retinal vascular diseases on UWF-CPF with good accuracy. This technology may be a useful tool for telemedicine and areas with a shortage of ophthalmic care.

Artificial intelligence for improving sickle cell retinopathy diagnosis and management

Sickle cell retinopathy is often initially asymptomatic even in proliferative stages, but can progress to cause vision loss due to vitreous haemorrhages or tractional retinal detachments. Challenges with access and adherence to screening dilated fundus examinations, particularly in medically underserved areas where the burden of sickle cell disease is highest, highlight the need for novel approaches to screening for patients with vision-threatening sickle cell retinopathy. This article reviews the existing literature on and suggests future research directions for coupling artificial intelligence with multimodal retinal imaging to expand access to automated, accurate, imaging-based screening for sickle cell retinopathy. Given the variability in retinal specialist practice patterns with regards to monitoring and treatment of sickle cell retinopathy, we also discuss recent progress toward development of machine learning models that can quantitatively track disease progression over time. These artificial intelligence-based applications have great potential for informing evidence-based and resource-efficient clinical diagnosis and management of sickle cell retinopathy.

Deep Learning Detection of Sea Fan Neovascularization From Ultra-Widefield Color Fundus Photographs of Patients With Sickle Cell Hemoglobinopathy

Importance

Adherence to screening for vision-threatening proliferative sickle cell retinopathy is limited among patients with sickle cell hemoglobinopathy despite guidelines recommending dilated fundus examinations beginning in childhood. An automated algorithm for detecting sea fan neovascularization from ultra-widefield color fundus photographs could expand access to rapid retinal evaluations to identify patients at risk of vision loss from proliferative sickle cell retinopathy.

Objective

To develop a deep learning system for detecting sea fan neovascularization from ultra-widefield color fundus photographs from patients with sickle cell hemoglobinopathy.

Design, Setting, and Participants

In a cross-sectional study conducted at a single-institution, tertiary academic referral center, deidentified, retrospectively collected, ultra-widefield color fundus photographs from 190 adults with sickle cell hemoglobinopathy were independently graded by 2 masked retinal specialists for presence or absence of sea fan neovascularization. A third masked retinal specialist regraded images with discordant or indeterminate grades. Consensus retinal specialist reference standard grades were used to train a convolutional neural network to classify images for presence or absence of sea fan neovascularization. Participants included nondiabetic adults with sickle cell hemoglobinopathy receiving care from a Wilmer Eye Institute retinal specialist; the patients had received no previous laser or surgical treatment for sickle cell retinopathy and underwent imaging with ultra-widefield color fundus photographs between January 1, 2012, and January 30, 2019.

Interventions

Deidentified ultra-widefield color fundus photographs were retrospectively collected.

Main Outcomes and Measures

Sensitivity, specificity, and area under the receiver operating characteristic curve of the convolutional neural network for sea fan detection.

Results

A total of 1182 images from 190 patients were included. Of the 190 patients, 101 were women (53.2%), and the mean (SD) age at baseline was 36.2 (12.3) years; 119 patients (62.6%) had hemoglobin SS disease and 46 (24.2%) had hemoglobin SC disease. One hundred seventy-nine patients (94.2%) were of Black or African descent. Images with sea fan neovascularization were obtained in 57 patients (30.0%). The convolutional neural network had an area under the curve of 0.988 (95% CI, 0.969-0.999), with sensitivity of 97.4% (95% CI, 86.5%-99.9%) and specificity of 97.0% (95% CI, 93.5%-98.9%) for detecting sea fan neovascularization from ultra-widefield color fundus photographs.

Conclusions and Relevance

This study reports an automated system with high sensitivity and specificity for detecting sea fan neovascularization from ultra-widefield color fundus photographs from patients with sickle cell hemoglobinopathy, with potential applications for improving screening for vision-threatening proliferative sickle cell retinopathy.

Retinal ultra-wide-field colour imaging versus dilated fundus examination to screen for sickle cell retinopathy

PURPOSE

To compare ultra-wide-field colour fundus imaging (UWFI) to dilated fundus examination (DFE) for the screening of sickle cell retinopathy (SCR).

DESIGN

This study is a prospective, blinded, multicentre case series.

PARTICIPANTS

This study included two groups: an adult group (n=268 eyes) and a paediatric group (n=168 eyes). Sickle cell disease (SCD) types included haemoglobin S homozygous (HbSS), haemoglobin S and C (HbSC) and Hb S with β-thalassaemia (HbSß-Thal).

METHODS

Participants underwent DFE and UWFI. Each eye received three independent grades (1-4), documented by three graders: clinical grader, image grader 1 and image grader 2. Three clinically relevant diagnostic thresholds were determined. Based on these thresholds, the sensitivity, specificity, positive predictive value and negative predictive value for all three graders were calculated relative to each other as reference tests.

RESULTS

HbSC was associated with the most advanced SCR grades. When compared to the clinical grader, image grader 1 and image grader 2 consistently detected more SCR and higher SCR grades in both adult and paediatric groups. In both groups, image grader 1 and image grader 2 identified twice as many cases of capillary occlusion/anastomosis than clinical grader. To detect the presence of any proliferative SCR, image grader 1 and image grader 2 had a sensitivity of 82%, 71% in the paediatrics group and 90% and 72% in the adult group. The clinical grader sensitivity was 52% in the paediatrics group and 53% in the adult group.

CONCLUSION

The UWFI is a sensitive tool to screen for SCR. It is superior to DFE in detecting capillary occlusion or anastomosis.