Introduction To design and evaluate a deep learning model based on ultra-widefield images (UWFIs) that can detect several common fundus diseases. Methods Based on 4574 UWFIs, a deep learning model was trained and validated that can identify normal fundus and eight common fundus diseases, namely referable diabetic retinopathy, retinal vein occlusion, pathologic myopia, retinal detachment, retinitis pigmentosa, age-related macular degeneration, vitreous opacity, and optic neuropathy. The model was tested on three test sets with data volumes of 465, 979, and 525. The performance of the three deep learning networks, EfficientNet-B7, DenseNet, and ResNet-101, was evaluated on the internal test set. Additionally, we compared the performance of the deep learning model with that of doctors in a tertiary referral hospital. Results Compared to the other two deep learning models, EfficientNet-B7 achieved the best performance. The area under the receiver operating characteristic curves of the EfficientNet-B7 model on the internal test set, external test set A and external test set B were 0.9708 (0.8772, 0.9849) to 1.0000 (1.0000, 1.0000), 0.9683 (0.8829, 0.9770) to 1.0000 (0.9975, 1.0000), and 0.8919 (0.7150, 0.9055) to 0.9977 (0.9165, 1.0000), respectively. On a data set of 100 images, the total accuracy of the deep learning model was 93.00%, the average accuracy of three ophthalmologists who had been working for 2 years and three ophthalmologists who had been working in fundus imaging for more than 5 years was 88.00% and 94.00%, respectively. Conclusion High performance was achieved on all three test sets using our UWFI multidisease classification model with a small sample size and fast model inference. The performance of the artificial intelligence model was comparable to that of a physician with 2–5 years of experience in fundus diseases at a tertiary referral hospital. The model is expected to be used as an effective aid for fundus disease screening.

To determine possible correlations between different levels of blood pressure (BP) control and retinal microvascular changes in the macula and optic nerve head, using coherence tomography angiograp...

Purpose: To compare the choroidal vascularity of large- and middle-sized choroidal vessels and choriocapillaris (CC) perfusion in patients with different degrees of myopia using swept-source optical coherence tomography angiography (SS-OCTA). Methods: One hundred and thirteen people with myopia were enrolled. SS-OCTA was performed to analyze the choroidal vascularity and CC perfusion. Three-dimensional (3D) choroidal vascularity index (CVI) and choroidal luminal volumes (LV) were obtained by artificial intelligence segmentation of the choroidal lumen in Volume OCT images. CC perfusion was assessed by flow signal voids (FSVs). Results: In the macular, multiple linear regression model showed that choroidal thickness (CT), total choroidal volume, LV, and choroidal stromal volume were negatively correlated with axis length (AL), respectively (all p < 0.001). Three dimensional CVI was negatively associated with AL (p < 0.05). FSV% was positively correlated with age only (p < 0.001). Additionally, after adjustment for age and AL, FSV% had a significant negative correlation with CT (p < 0.05). Conclusion: Choroidal vascularity decreases gradually with increasing severity of myopia. The decrease of CC blood perfusion was related to a higher severity of myopia and the thinning of choroid.

Chemotherapy is an important part of retinoblastoma (RB) treatment. However, the development of drug resistance increases the likelihood of treatment failure. Therefore, increasing the sensitivity of chemotherapeutic drugs is very important. Recent research has explored the relationship between the expression level of gasdermin E (GSDME) and drug sensitivity in RB. Our study found that GSDME expression was significantly reduced in human RB tissues and cell lines. Downregulation of GSDME expression reduced the sensitivity of cells to chemotherapeutic drugs. Decitabine treatment and transfection with GSDME‐overexpressing lentivirus (LV‐GSDME) upregulated GSDME expression in Y79 and WERI‐RB‐1 cell lines. The half maximal inhibitory concentrations (IC₅₀) for carboplatin‐induced cell death were significantly reduced. Low‐dose carboplatin could achieve the IC₅₀, and no significant difference was found in the production of prodeath‐activating proteins, but the mode of cell death changed from apoptosis to pyroptosis. Increased GSDME expression can reduce the required dose of chemotherapeutic drugs. After inhibition of caspase‐3 activation, the IC₅₀ of carboplatin‐induced cell death was significantly increased in cells with high GSDME expression, and the method of cell death switched from pyroptosis to apoptosis, which increased the concentration of chemotherapeutic drugs. Furthermore, the sensitivity of cells to carboplatin was reduced. The in vivo xenograft tumor model further confirmed that GSDME upregulation could promote carboplatin‐induced tumor cell death. Therefore, the sensitivity of cells to chemotherapeutic drugs can be predicted by detecting the GSDME expression level, and we used pyroptosis induction as a new method for promoting tumor death.

Purpose: To determine possible impacts on retinal microvasculature in healthy young adults during exercise with a face mask, using optical coherence tomography angiography (OCTA). Methods: Twenty-three healthy participants (23 eyes, 17 women and 6 men) performed the incremental continuous running test (ICRT) with different masks. OCTA of the macula and optic nerve head were performed before and after ICRT to detect changes in retinal vessel density (VD). All participants were in groups A, B, and C (before ICRT) and groups A′, B′, and C′ (after ICRT), which comprised data from volunteers without a mask, with a surgical mask, and with an N95 mask, respectively. Results: Before ICRT, group C showed significantly reduced VD in the superficial plexus (SP), except foveal VD, compared with group A (P < 0.05). After ICRT, groups B′ and C′ showed significantly shorter maximum running time, lower oxygen saturation, and lower perifoveal VD of SP compared with group A′ (P < 0.05). Conclusions: Use of an N95 mask reduced VD in SP even under quiescent conditions, which might have clinical implications for protecting healthy workers and indoor manual labor workers from potential risks of retinal damage due to long-term mask use. Moreover, mask use while exercising might lead to attenuated exercise ability and lower VD in SP, which should be investigated in additional studies. Translational Relevance: Retina vascular perfusion dynamics could be monitored in vivo by OCTA, which would be valuable to study physiologic retinal blood flow redistribution and potential impacts on retinal vascular perfusion during exercise with face masks.

Artesunate (ART) is a derivative of artemisinin. Compared with artemisinin, ART has excellent water solubility, high stability and oral bioavailability. In this review, the application of ART in cl... Plain language summary A medicine called artesunate (ART), which comes from a plant, is sometimes used to treat malaria. It was first introduced by Chinese scientist Tu Youyou, who won the Nobel Pr...