Predictive and Diagnostic Approaches for Systemic Disorders Using Ocular Assessment

Editors

Kai Jin

Eye Center, Second Affiliated Hospital, School of Medicine, Zhejiang University

Jiong Zhang

University of Southern California

Andrzej Grzybowski

University of Warmia and Mazury in Olsztyn

Impact

Study Protocol

15 October 2024

Association between central corneal thickness and systemic lupus erythematosus: a cross-sectional study protocol

Juan David Saldaña-Garrido

, 3 more and

Francisca Sivera

1,092 views

2 citations

An overview of the study approach and methodology. The multi-source heterogeneous fundus (MSHF) dataset is collected, and then serves as an input to train the local and global attention aggregated deep neural network (LGAANet). The output is the image quality of each image based on three metrics, and a heap map is created to show the interpretability.

Original Research

07 August 2024

Enhancing the ophthalmic AI assessment with a fundus image quality classifier using local and global attention mechanisms

Shengzhan Wang

, 10 more and

Juan Ye

1,168 views

0 citations

Framework of the proposed method. Our net includes three components: the DR grading module based on transfer learning, adversarial multi-lesion masks generate network, and lesion external attention module. The input data consists of a very small set of unlabeled target images and a large set of annotated source images. The adversarial multi-lesion masks generate network is used to learn multi-lesion masks, where target lesion predicted will with the original images as inputs for training DR grading module based on transfer learning. At the same time, the lesion external attention module aims to force classification network to pay those lesions for improving the final disease grading performance.

Original Research

14 May 2024

Cross-modality transfer learning with knowledge infusion for diabetic retinopathy grading

Tao Chen

, 7 more and

Yitian Zhao

1,401 views

0 citations

Original Research

22 March 2024

Association of autoimmune and allergic diseases with senile cataract: a bidirectional two-sample Mendelian randomization study

Weichen Yuan

, 3 more and

Fangkun Zhao

Background: Many observational studies have been reported that patients with autoimmune or allergic diseases seem to have a higher risk of developing senile cataract, but the views are not consistent. In order to minimize the influence of reverse causality and potential confounding factors, we performed Mendelian Randomization (MR) analysis to investigate the genetic causal associations between autoimmune, allergic diseases and senile cataract.

Methods: Single nucleotide polymorphisms associated with ten common autoimmune and allergic diseases were obtained from the IEU Open genome-wide association studies (GWAS) database. Summary-level GWAS statistics for clinically diagnosed senile cataract were obtained from the FinnGen research project GWAS, which consisted of 59,522 individuals with senile cataracts and 312,864 control individuals. MR analysis was conducted using mainly inverse variance weighted (IVW) method and further sensitivity analysis was performed to test robustness.

Results: As for ten diseases, IVW results confirmed that type 1 diabetes (OR = 1.06; 95% CI = 1.05-1.08; p = 2.24×10^-12), rheumatoid arthritis (OR = 1.05; 95% CI = 1.02-1.08; p = 1.83×10^-4), hypothyroidism (OR = 2.4; 95% CI = 1.42-4.06; p = 1.12×10^-3), systemic lupus erythematosus (OR = 1.02; 95% CI = 1.01-1.03; p = 2.27×10^-3), asthma (OR = 1.02; 95% CI = 1.01-1.03; p = 1.2×10^-3) and allergic rhinitis (OR = 1.07; 95% CI = 1.02-1.11; p = 2.15×10^-3) were correlated with the risk of senile cataract. Celiac disease (OR = 1.04; 95% CI = 1.01-1.08; P = 0.0437) and atopic dermatitis (OR = 1.05; 95% CI = 1.01-1.10; P = 0.0426) exhibited a suggestive connection with senile cataract after Bonferroni correction. These associations are consistent across weighted median and MR Egger methods, with similar causal estimates in direction and magnitude. Sensitivity analysis further proved that these associations were reliable.

Conclusions: The results of the MR analysis showed that there were causal relationships between type 1 diabetes, rheumatoid arthritis, hypothyroidism, systemic lupus erythematosus, asthma, allergic rhinitis and senile cataract. To clarify the possible role of autoimmune and allergy in the pathophysiology of senile cataract, further studies are needed.