BackgroundThe visual outcome of open globe injury (OGI)-no light perception (NLP) eyes is unpredictable traditionally. This study aimed to develop a model to predict the visual outcomes of vitrectomy surgery in OGI-NLP eyes using a machine learning algorithm and to provide an interpretable system for the prediction results.MethodsClinical data of 459 OGI-NLP eyes were retrospectively collected from 19 medical centres across China to establish a training data set for developing a model, called ‘VisionGo’, which can predict the visual outcome of the patients involved and compare with the Ocular Trauma Score (OTS). Another 72 cases were retrospectively collected and used for human–machine comparison, and an additional 27 cases were prospectively collected for real-world validation of the model. The SHapley Additive exPlanations method was applied to analyse feature contribution to the model. An online platform was built for real-world application.ResultsThe area under the receiver operating characteristic curve (AUC) of VisionGo was 0.75 and 0.90 in previtrectomy and intravitrectomy application scenarios, which was much higher than the OTS (AUC=0.49). VisionGo showed better performance than ophthalmologists in both previtrectomy and intravitrectomy application scenarios (AUC=0.73 vs 0.57 and 0.87 vs 0.64). In real-world validation, VisionGo achieved an AUC of 0.60 and 0.91 in previtrectomy and intravitrectomy application scenarios. Feature contribution analysis indicated that wound length-related indicators, vitreous status and retina-related indicators contributed highly to visual outcomes.ConclusionsVisionGo has achieved an accurate and reliable prediction in visual outcome after vitrectomy for OGI-NLP eyes.

Background To investigate the associations between retinal/choroidal microvasculature and carotid plaque in patients with CHD assessed by optical coherence tomography angiography (OCTA). Methods This study included 127 CHD patients with and 79 without carotid plaque. Each patient had both OCTA taken and digitized to determine retinal/choroidal thickness, vessel density and flow area and carotid ultrasound for carotid plaque size and stability measurement. SCP, DCP, out retina and choriocapillaris vessel density, out retina and choriocapillaris flow area, and full retina thickness were analyzed in the fovea centered 6 × 6 mm area. The association between OCTA measurements and carotid plaque characteristics in patients with CHD were evaluated. Results The duration of hypertension and DM was significantly longer in CHD patients with carotid plaque than that without (p < 0.001). The mean values for vessel density SCP and DCP (except fovea zone), and choriocapillaris nasal zone were significantly lower in plaque group (p < 0.05). Negative correlations between the carotid plaque width and vessel density SCP and DCP (except fovea zone) (p < 0.05) were also found in this study. Conclusions In patients with CHD, carotid plaque, a risk factor and marker of atherosclerosis and stenosis, is significantly and independently associated with retinal and choroidal microvascular changes by OCTA.