Multimodal deep transfer learning to predict retinal vein occlusion macular edema recurrence after anti-VEGF therapy.

Purpose: To develop a multimodal deep transfer learning (DTL) fusion model using optical coherence tomography angiography (OCTA) images to predict the recurrence of retinal vein occlusion (RVO) and macular edema (ME) after three consecutive anti-VEGF therapies. Methods: This retrospective cross-sectional study consisted of 2800 B-scan OCTA macular images collected from 140 patients with RVO-ME. The central macular thickness (CMT) > 250 µm was used as a criterion for recurrence in the three-month follow-up after three injections of anti-VEGF therapy. The qualified OCTA image preprocessing and the lesion area segmentation were performed by senior ophthalmologists. We developed and validated the clinical, DTL, and multimodal fusion models based on clinical and extracted OCTA imaging features. The performance of the models and experts predictions were evaluated using several performance metrics, including the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity. Results: The DTL models exhibited higher prediction efficacy than the clinical models and experts' predictions. Among the DTL models, the Vgg19 performed better than that of the other models, with an AUC of 0.968 (95 % CI, 0.943-0.994), accuracy of 0.913, sensitivity of 0.922, and specificity of 0.902 in the validation cohort. Moreover, the fusion Vgg19 model showed the highest prediction efficacy among all the models, with an AUC of 0.972 (95 % CI, 0.946-0.997), accuracy of 0.935, sensitivity of 0.935, and specificity of 0.934 in the validation cohort. Conclusions: Multimodal fusion DTL models showed robust performance in predicting RVO-ME recurrence and may be applied to assist clinicians in determining patients' follow-up time after anti-VEGF therapy.

S100A8 promotes migration and infiltration of inflammatory cells in acute anterior uveitis.

Uveitis, the pathologic condition of inflammation of the uvea, frequently leads to severe vision loss and blindness. S100A8 is a calcium-binding protein which mainly expresses in granulocytes and monocytes and plays a prominent role in the regulation of inflammatory processes and immune response. Here, we determined the role of S100A8-positive cells in acute anterior uveitis (AAU) and keratitis. In rat models of endotoxin (lipopolisaccharide, LPS) -induced uveitis (EIU) and keratitis, S100A8-positive granulocytes and monocytes increased significantly in the iris-ciliary body and cornea as well as in the blood. Interestingly, Glucocorticoids slightly increased S100A8 levels in leukocytes, but reduced its presence significantly in the iris-ciliary body after LPS injection. Moreover, inhibition of NF-kB activation remarkably suppressed both progression of AAU and total S100A8 levels in leukocytes and the iris-ciliary body after LPS administration. Additionally, S100A8 protein level was also found to be elevated in the serum of AAU patients parallel with the progression of AAU through the designated clinical stages. Thus, S100A8 plays a pivotal role in the processes of AAU through involvement in migration and infiltration of S100A8-positive cells. Our findings suggest that serum levels of S100A8 protein can be used to monitor inflammatory activity in AAU.