Objective: With the scenario of limited labeled dataset, this paper introduces a deep learning-based approach that leverages Diabetic Retinopathy (DR) severity recognition performance using fundus images combined with wide-field swept-source optical coherence tomography angiography (SS-OCTA). Methods: The proposed architecture comprises a backbone convolutional network associated with a Twofold Feature Augmentation mechanism, namely TFA-Net. The former includes multiple convolution blocks extracting representational features at various scales. The latter is constructed in a two-stage manner, i.e., the utilization of weight-sharing convolution kernels and the deployment of a Reverse Cross-Attention (RCA) stream. Results: The proposed model achieves a Quadratic Weighted Kappa rate of 90.2% on the small-sized internal KHUMC dataset. The robustness of the RCA stream is also evaluated by the single-modal Messidor dataset, of which the obtained mean Accuracy (94.8%) and Area Under Receiver Operating Characteristic (99.4%) outperform those of the state-of-the-arts significantly. Conclusion: Utilizing a network strongly regularized at feature space to learn the amalgamation of different modalities is of proven effectiveness. Thanks to the widespread availability of multi-modal retinal imaging for each diabetes patient nowadays, such approach can reduce the heavy reliance on large quantity of labeled visual data. Significance: Our TFA-Net is able to coordinate hybrid information of fundus photos and wide-field SS-OCTA for exhaustively exploiting DR-oriented biomarkers. Moreover, the embedded feature-wise augmentation scheme can enrich generalization ability efficiently despite learning from small-scale labeled data.

We evaluated the incidence and risk factors for recurrence in patients with neovascular age-related macular degeneration (nAMD) who discontinued anti-vascular endothelial growth factor (VEGF) therapy under a modified treat-and-extend (TAE) protocol. A retrospective analysis of 68 patients was conducted. Therapy was discontinued after extending the treatment interval to ≥ 5 months and maintaining disease stability for 6 months. The modified TAE protocol included three phases: loading, observation, and TAE, with initial treatment intervals determined by the first recurrence interval. Recurrence rates were 22.2%, 42.2%, and 54.4% at 1-, 2-, and 3-year follow-ups, respectively. The median time to recurrence was 16 months, with patients receiving an average of 7.7 injections before discontinuation. Intraretinal cysts were significantly more prevalent in patients with recurrence. Rapid early response to treatment was associated with a lower risk of exudative recurrence. Vision loss of two or more lines occurred in five patients despite treatment resumption; all exhibited subretinal hemorrhages on baseline imaging. The modified TAE protocol allows for successful therapy discontinuation with fewer injections and reduced recurrence rates. Patients with a favorable early response to anti-VEGF therapy had a lower risk of recurrence.

Purpose: To evaluate the peripapillary microvasculature changes in patients with epiretinal membrane (ERM) following pars plana vitrectomy (PPV) with internal limiting membrane (ILM) peeling using swept-source optical coherence tomography angiography (SS-OCTA). Method: Medical records and multimodal imaging data of 33 eyes after PPV for ERM were retrospectively reviewed. Peripapillary SS-OCTA images of 6×6 mm² were recorded at at pre- and post-operatively every 6 months for 1 year. A semi-automated method was used to analyzed SS-OCTA images, excluding the optic disc area, using the MATLAB software. The peripapillary vessel density (pVD) of superficial capillary plexus (SCP) and deep capillary plexus (DCP) was quantified in four quadrants (superior, inferior, nasal and temporal). Result: The mean pVD in SCP and DCP decreased at 6- and 12-months follow-up. In sectoral analysis, superior, inferior, and temporal quadrants pVD in SCP and DCP were significantly reduced at 1 year after vitrectomy (all p < 0.05). Meanwhile, inferior quadrants pVD in SCP and DCP showed the earliest significant reduction at 6-months (p = 0.022 and 0.048, respectively). A reduction of post-operative mean pVD in DCP was significantly greater in patients with diabetic retinopathy (p = 0.043). Conclusion: Peripapillary capillary density significantly decreased after surgical removal of ERM. Vitrectomy with ILM peeling can cause peripapillary microvascular damage starting in inferior sector.

Background: Platinum-based combination chemotherapy, including cisplatin and carboplatin, are important cytotoxic anti-cancer agents that are widely used to treat various solid tumors. Carboplatin has a similar effect on survival in small cell lung cancer, but generally has a milder toxicity profile when compared with cisplatin. Both may cause moderate or severe neurotoxicity, but ocular neurotoxicity from carboplatin is rarely reported. Case presentation: A 79-year-old man underwent intravenous polychemotherapy (atezolizumab, etoposide, and carboplatin) for small cell lung cancer. One week after the second cycle of chemotherapy, he reported bilateral visual loss as hand motion in both eyes. Dilated fundus examination showed retinal arterial narrowing without hemorrhage, and diffuse choroidal and retinal thinning was observed in an optical coherence tomography scan. Fluorescein angiography revealed significantly delayed circulation without evidence of obstructive lesions. 30-Flicker electroretinogram testing showed a complete absence of cone response in both eyes. The patient’s visual acuity aggravated to no light perception in both eyes, even after the cessation of chemotherapy. Conclusions: Carboplatin combination chemotherapy administered at therapeutic doses can result in irreversible visual loss, a side effect that is not widely acknowledged. When using carboplatin, physicians should be aware of its potential ocular toxicity.