Search Results (3)

Hyperfluorescence (HF) and reduced autofluorescence (RA) are important biomarkers in fundus autofluorescence images (FAF) for the assessment of health of the retinal pigment epithelium (RPE), an important indicator of disease progression in geographic atrophy (GA) or central serous chorioretinopathy (CSCR). Autofluorescence images have been annotated by human raters, but distinguishing biomarkers (whether signals are increased or decreased) from the normal background proves challenging, with borders being particularly open to interpretation. Consequently, significant variations emerge among different graders, and even within the same grader during repeated annotations. Tests on in-house FAF data show that even highly skilled medical experts, despite previously discussing and settling on precise annotation guidelines, reach a pair-wise agreement measured in a Dice score of no more than 63–80% for HF segmentations and only 14–52% for RA. The data further show that the agreement of our primary annotation expert with herself is a 72% Dice score for HF and 51% for RA. Given these numbers, the task of automated HF and RA segmentation cannot simply be refined to the improvement in a segmentation score. Instead, we propose the use of a segmentation ensemble. Learning from images with a single annotation, the ensemble reaches expert-like performance with an agreement of a 64–81% Dice score for HF and 21–41% for RA with all our experts. In addition, utilizing the mean predictions of the ensemble networks and their variance, we devise ternary segmentations where FAF image areas are labeled either as confident background, confident HF, or potential HF, ensuring that predictions are reliable where they are confident (97% Precision), while detecting all instances of HF (99% Recall) annotated by all experts. Full article

Background: Central serous chorioretinopathy (CSCR) is a chorioretinal disease affecting mostly middle age males. It is marked by the serous detachment of the neurosensory layer at the macula. This review of the literature provides a framework of the current characteristic/relevant imaging findings of CSCR. Although the pathogenesis of CSCR is unclear, the choroid plays a major role and its changes are fundamental to the diagnosis and treatment of CSCR. Methods: A systematic literature search focusing on current multimodal imaging for CSCR was performed. Only articles reporting on original clinical data were selected, studies in a language other than English were included only if an English abstract was provided. Additional sources included articles cited in the references list of the first selected articles. We deduced imaging findings based on current and relevant literature on the topic. Results: We found that sub foveal choroidal thickness (SFCT) and choroidal vascularity index (CVI) were greater in eyes with acute CSCR than in eyes with chronic CSCR or normal eyes. There was increased choroidal thickness (CT) in the macula compared to peripapillary region. In healthy eyes, the highest CVI was found in the nasal region followed by the inferior, temporal, and superior quadrant. The area with the least CVI was the macula. In eyes with CSCR, 100% had asymmetric dominant vortex veins compared to 38% in normal eyes. Conclusion: Choroidal imaging has advanced the diagnosis of CSCR. This has led to numerous imaging biomarkers like CVI, CT, and hyper-reflective dots for early detection and possible prognostication of CSCR. More techniques like wide field scans and en face imaging are being employed to characterize the choroid in CSCR. Full article

The diagnosis and treatment of choroidal neovascularization (CNV) in eyes with chronic central serous chorioretinopathy (CSCR) can be challenging. The purpose of this study was to classify eyes with suspected CNV using multimodal imaging. The effect of intravitreal anti-vascular endothelial growth factor (VEGF) was assessed and compared to controls. This retrospective study included chronic CSCR patients with suspected secondary CNV who received intravitreal bevacizumab. Eyes were divided into “definite CNV” and “no CNV” based on optical coherence tomography angiography (OCTA). Eyes that did not undergo OCTA imaging were considered as “presumed CNV”. One-year outcome in visual acuity (VA) and central foveal thickness (CFT) were investigated and compared to non-treated control patients to assess the response to anti-VEGF. Logistic regression analysis was used to explore predictive biomarkers of CNV detection and improvement after anti-VEGF. Ninety-two eyes with chronic CSCR from 88 participants were included in this study. Sixty-one eyes received bevacizumab and 31 eyes were non-treated control subjects. The presence of subretinal hyperreflective material (SHRM) and shallow irregular retinal pigment epithelium (RPE) elevation (SIRE) with sub-RPE hyperreflectivity on OCT was associated with a significantly increased risk of detecting CNV on OCTA. Intravitreal anti-VEGF caused significant functional and anatomical improvement in patients with neovascular CSCR as compared to non-treated eyes. In contrast, VA and CFT changes were not significantly different between treated and non-treated CSCR with no evidence of CNV on OCTA. No clinical or anatomical biomarkers were found to be associated with response to treatment. In conclusion, OCTA should be used to confirm the presence CNV in suspected chronic CSCR patients. Intravitreal anti-VEGF treatment resulted in a significantly better one-year outcome in patients with definitive OCTA evidence of CNV. Full article