Submacular hemorrhage (SMH), which may arise from age-related macular degeneration, retinal arterial macroaneurysm, and other causes, can result in severe vision loss and central visual field impairment. Although tissue plasminogen activator (tPA) is used off-label to treat SMH in many countries, no formulation has been approved for this indication. Because early intervention is critical when tPA is used for SMH, limited access to centers that can provide this treatment may delay care and reduce treatment opportunities. Decentralized clinical trials (DCTs) reduce or eliminate the need for participants to travel to trial sites. We therefore designed the investigator-initiated SACLA trial to evaluate subretinal tPA for SMH. The DCT framework is intended to reduce logistical barriers related to the disease severity and rarity. The SACLA trial is a phase II multicenter, open-label, single-arm surgical study with a pre-post comparison design (jRCT2071250003). Twenty eligible participants will undergo pars plana vitrectomy followed by subretinal injection of 0.1 mL (8,000 IU) of tPA. Participants will remain hospitalized at the trial site until the primary outcome, change in central foveal thickness (CFT) from baseline at Week 1, is assessed. Thereafter, follow-up visits will be conducted at either the trial site or partner sites within the DCT framework. Secondary efficacy outcomes include change in CFT from baseline, presence of a foveal hemorrhage measuring at least 1 disc diameter, best-corrected visual acuity (BCVA), and change in BCVA from baseline at Weeks 4 and 12. Adverse events will be collected throughout the 12-week observation period to assess safety. The protocol and related study documents were reviewed and approved by the Saga University Hospital Institutional Review Board. This study is designed to generate prospective evidence on the feasibility, short-term anatomical response, and safety of subretinal tPA for SMH.

Diabetic Retinopathy (DR) is a medical condition in which high blood sugar levels damage the retina's blood vessels. Existing Solutions for multi-class DR identification are computationally intensive and also suffer from low accuracy. There is an immense need for an automated, computationally efficient approach for monitoring DR progression in diabetic patients. The study proposed Random Forest (RF), Logistic Regression (LR), Decision Tree (DT), and Gaussian Naive Bayes (GNB) models for the classification of retinal images into five DR classes (No, Mild, Moderate, Proliferate, and Severe) using spatial features extracted through a Convolutional Neural Network (CNN), textural features extracted through a Grey Level Co-occurrence Matrix (GLCM), and hybrid features by combining these features. The CNN, EfficientNet, PyramidCNN, and Pyramid Vision Transformer (PVT) were also evaluated for the classification of DR stages. The results revealed that the RF model with hybrid features outperformed, with an accuracy of 98.00% and high performance across all evaluation metrics, with a 1.00% increase over existing approaches. The EfficientNet model also performs competitively with 97.00% accuracy. The ML models also emerged as computationally efficient in terms of training and inference time for deployment in low-resource clinical environments for automated monitoring of DR progression in diabetic patients.