Purpose:The alteration of the exosomal proteins in the aqueous humor (AH) is linked to the development of eye diseases. The goal of this study was to examine the exosomal protein profile of patients with age-related macular degeneration (AMD) to better understand their role in the pathogenesis of AMD. Methods: Exosomes were isolated from the AH of 28 AMD and 25 control eyes. The quality, concentration, and size distribution of exosomes were measured using a nanoparticle tracking analysis system (NTA). Total exosomal proteins from each sample were purified and digested with trypsin for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. Results: Based on LC-MS/MS analysis, we got 105 exosomal peptides from AMD and control patients. Gene ontology (GO) analysis in the biology process revealed that exosomal proteins of AMD were enriched in the lipoprotein metabolic process. T-test analysis revealed six exosomal proteins in patients with AMD were significantly different from controls. Comparing the exosomal protein profile of AMD patients who were receiving anti-VEGF therapy, we observed the amount of two proteins decreased with the duration of the anti-VEGF treatment time. Conclusions: In this study, we successfully isolated and purified AH exosomes. Our results provide pioneering findings for the exosomal protein profile in AMD development and under therapy. These unique proteins could be the new targets for drug discovery or biological markers for evaluating therapeutic efficacy.

The aim of the study is to determine the effects of monocular 0.125% atropine daily treatment on the longer axial length (AL) eyes in children with pediatric anisometropia. This was a retrospective cohort study. The charts of children with anisometropia (aged 6–15 years) who had a > 0.2-mm difference in AL between the two eyes were reviewed. Children who received monocular treatment of 0.125% atropine in the eye with longer AL were included for final analysis. The main outcome measure was the difference in AL between the two eyes after treatment. Regression analysis was used to model the changes in AL according to the time of treatment in both eyes. Finally, forty eyes in 20 patients (mean age 10.2 years) were included in the analyses. During the treatment period, AL was controlled in the treated eyes (p = 0.389) but elongated significantly in the untreated eyes (p < 0.001). The difference in AL between the treated and untreated eyes decreased from 0.57 to 0.22 mm (p < 0.001) after the 1-year treatment period. In the regression model, the best fit for the relationship between changes in AL and time during the treatment period in the treated eyes was the quadratic regression model with a concave function. In conclusion, these data suggest that 0.125% atropine daily is an effective treatment to reduce the interocular difference of AL in eyes with axial anisometropia. This pilot study provides useful information for future prospective and larger studies of atropine for the treatment of pediatric axial anisometropia.

Deep learning (DL) algorithms are used to diagnose diabetic retinopathy (DR). However, most of these algorithms have been trained using global data or data from patients of a single region. Using different model architectures (e.g., Inception-v3, ResNet101, and DenseNet121), we assessed the necessity of modifying the algorithms for universal society screening. We used the open-source dataset from the Kaggle Diabetic Retinopathy Detection competition to develop a model for the detection of DR severity. We used a local dataset from Taipei City Hospital to verify the necessity of model localization and validated the three aforementioned models with local datasets. The experimental results revealed that Inception-v3 outperformed ResNet101 and DenseNet121 with a foreign global dataset, whereas DenseNet121 outperformed Inception-v3 and ResNet101 with the local dataset. The quadratic weighted kappa score (κ) was used to evaluate the model performance. All models had 5–8% higher κ for the local dataset than for the foreign dataset. Confusion matrix analysis revealed that, compared with the local ophthalmologists’ diagnoses, the severity predicted by the three models was overestimated. Thus, DL algorithms using artificial intelligence based on global data must be locally modified to ensure the applicability of a well-trained model to make diagnoses in clinical environments.

No abstract