Epidemiological studies suggest that antidepressants use may increase the risk of cataract, but the results are inconclusive. We aimed to examine this association by performing a systematic review and meta-analysis. Relevant studies were identified by searching PubMed and Web of Science databases through June 2017. We included studies that reported risk estimates for the association between antidepressants use and cataract risk. A random-effects model was used to calculate the summary odds ratio (OR) with its 95% confidence interval (CI). We identified seven studies of antidepressants use and risk of cataract involving 447,672 cases and 1,510,391 controls. Overall, the combined ORs (95% CIs) of cataract for selective serotonin reuptake inhibitors (SSRIs), serotonin noradrenalin reuptake inhibitors (SNRIs), and tricyclic antidepressants (TCAs) were 1.12 (1.06–1.19), 1.13 (1.04–1.24), and 1.19 (1.11–1.28), respectively. A certain degree of heterogeneity was observed across studies (P < 0.001, I 2 = 92.2% for SSRIs, P = 0.026, I 2 = 67.5% for SNRIs, and P = 0.092, I 2 = 58.0% for TCAs). This meta-analysis provides evidence of a significant positive association between antidepressants use and risk of cataract. Because of the heterogeneity and limited eligible studies, further prospective studies are warranted to confirm the preliminary findings of our study.
Purpose. Iris biological features such as surface features and profile characteristics reflected the development of iris stroma and microvessels. Iris vessels and microcirculation are still lack of effective detection methods, and we can directly observe only the iris surface biological characteristics. This cross-sectional study evaluated the association between iris surface biological features and corneal biomechanics in young adults with myopia. Methods. We recruited 152 patients with myopia aged ≥18 years, from the Eye Hospital of Wenzhou Medical University, who had complete Corneal Visualization Scheimpflug Technology (Corvis ST) data and graded iris surface features. Iris surface features included crypts, furrows, and color measured from digital slit lamp images. The biomechanical properties of the cornea were assessed using Corvis ST. Only 1 eye of each participant was randomly selected for analysis. Associations between the iris surface features and corneal biomechanics were analyzed using linear regression models. The grade of iris crypts, furrows, and color and corneal biomechanical parameters measured with Corvis ST was the main outcome measures. Results. The iris crypts were significantly associated with deflection amplitude at the first applanation (A1 DLA, , ), A1 delta arc length (A1 dArcL) ( , ), maximum delta arc length (dArcLM) ( , ), and stiffness at the first applanation (SP-A1) ( , ). The iris furrows were only associated with integrated radius ( , ). Iris color was found not related with corneal biomechanical parameters measured via Corvis ST. Conclusions. Iris surface features were associated with corneal biomechanical properties in myopic eyes; more iris crypts were associated with lower corneal stiffness while more extensive furrows were related with higher corneal stiffness. Iris crypts and furrows may provide useful information on corneal biomechanical properties in myopic eyes.
28 April 2025 in Computer Methods and Programs in Biomedicine
Artificial intelligence (AI) models are effective for analyzing high-quality slit-lamp images but often face challenges in real-world clinical settings due to image variability. This study aims to develop and evaluate a hybrid AI-based image quality control system to classify slit-lamp images, improving diagnostic accuracy and efficiency, particularly in telemedicine applications. Cross-sectional study. Our Zhejiang Eye Hospital dataset comprised 2982 slit-lamp images as the internal dataset. Two external datasets were included: 13,554 images from the Aier Guangming Eye Hospital (AGEH) and 9853 images from the First People's Hospital of Aksu District in Xinjiang (FPH of Aksu). We developed a Hybrid Prior-Net (HP-Net), a novel network that combines a ResNet-based classification branch with a prior knowledge branch leveraging Hough circle transform and frequency domain blur detection. The two branches' features are channel-wise concatenated at the fully connected layer, enhancing representational power and improving the network's ability to classify eligible, misaligned, blurred, and underexposed corneal images. Model performance was evaluated using metrics such as accuracy, precision, recall, specificity, and F1-score, and compared against the performance of other deep learning models. The HP-Net outperformed all other models, achieving an accuracy of 99.03 %, precision of 98.21 %, recall of 95.18 %, specificity of 99.36 %, and an F1-score of 96.54 % in image classification. The results demonstrated that HP-Net was also highly effective in filtering slit-lamp images from the other two datasets, AGEH and FPH of Aksu with accuracies of 97.23 % and 96.97 %, respectively. These results underscore the superior feature extraction and classification capabilities of HP-Net across all evaluated metrics. Our AI-based image quality control system offers a robust and efficient solution for classifying corneal images, with significant implications for telemedicine applications. By incorporating slightly blurred but diagnostically usable images into training datasets, the system enhances the reliability and adaptability of AI tools for medical imaging quality control, paving the way for more accurate and efficient diagnostic workflows.