Vitrectomy versus scleral buckle for retinal detachment without posterior vitreous detachment.

To compare the effectiveness and safety of scleral buckling and pars plana vitrectomy in treating retinal detachment without posterior vitreous detachment. A total of 88 eyes of 83 patients with retinal detachment without prior posterior vitreous detachment were investigated retrospectively. Group A comprised patients who underwent scleral buckling (n = 47) and Group B (n = 36) patients who were treated with pars plana vitrectomy. Anatomical success, postoperative visual acuity, and ocular adverse events were evaluated. The primary and final anatomical success rate showed a nonsignificant difference (p = 0.465 and p = 0.37 respectively). No significant difference was observed in the reoperation rate or development of epiretinal membrane between the groups (p = 0.254 and p = 0.254 respectively). However, scleral buckling resulted in significantly better visual acuity at the last follow-up (0.12 ± 0.23) compared to pars plana vitrectomy (0.37 ± 0.46, p = 0.001). The incidence of cataract progression was also significantly higher in the pars plana vitrectomy group (46%) compared to the scleral buckling group (10%, p < 0.001). Scleral buckling and pars plana vitrectomy show similar success rates in treating retinal detachment without vitreous detachment. However, due to less cataract progression and better visual acuity outcomes, scleral buckling is recommended for these cases. Determining vitreous status before surgery is crucial for optimal outcomes.

Quality assessment of colour fundus and fluorescein angiography images using deep learning.

BACKGROUND/AIMS: Image quality assessment (IQA) is crucial for both reading centres in clinical studies and routine practice, as only adequate quality allows clinicians to correctly identify diseases and treat patients accordingly. Here we aim to develop a neural network for automated real-time IQA in colour fundus (CF) and fluorescein angiography (FA) images. METHODS: Training and evaluation of two neural networks were conducted using 2272 CF and 2492 FA images, with binary labels in four (contrast, focus, illumination, shadow and reflection) and three (contrast, focus, noise) modality specific categories plus an overall quality ranking. Performance was compared with a second human grader, evaluated on an external public dataset and in a clinical trial use-case. RESULTS: The networks achieved a F1-score/area under the receiving operator characteristic/precision recall curve of 0.907/0.963/0.966 for CF and 0.822/0.918/0.889 for FA in overall quality prediction with similar results in most categories. A clear relation between model uncertainty and prediction error was observed. In the clinical trial use-case evaluation, the networks achieved an accuracy of 0.930 for CF and 0.895 for FA. CONCLUSION: The presented method allows automated IQA in real time, demonstrating human-level performance for CF as well as FA. Such models can help to overcome the problem of human intergrader and intragrader variability by providing objective and reproducible IQA results. It has particular relevance for real-time feedback in multicentre clinical studies, when images are uploaded to central reading centre portals. Moreover, automated IQA as preprocessing step can support integrating automated approaches into clinical practice.