English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

The status of retinal arteriovenous crossing is of great significance forclinical evaluation of arteriolosclerosis and systemic hypertension. As anophthalmology diagnostic criteria, Scheie's classification has been used tograde the severity of arteriolosclerosis. In this paper, we propose a deeplearning approach to support the diagnosis process, which, to the best of ourknowledge, is one of the earliest attempts in medical imaging. The proposedpipeline is three-fold. First, we adopt segmentation and classification modelsto automatically obtain vessels in a retinal image with the correspondingartery/vein labels and find candidate arteriovenous crossing points. Second, weuse a classification model to validate the true crossing point. At last, thegrade of severity for the vessel crossings is classified. To better address theproblem of label ambiguity and imbalanced label distribution, we propose a newmodel, named multi-diagnosis team network (MDTNet), in which the sub-modelswith different structures or different loss functions provide differentdecisions. MDTNet unifies these diverse theories to give the final decisionwith high accuracy. Our severity grading method was able to validate crossingpoints with precision and recall of 96.3% and 96.3%, respectively. Amongcorrectly detected crossing points, the kappa value for the agreement betweenthe grading by a retina specialist and the estimated score was 0.85, with anaccuracy of 0.92. The numerical results demonstrate that our method can achievea good performance in both arteriovenous crossing validation and severitygrading tasks. By the proposed models, we could build a pipeline reproducingretina specialist's subjective grading without feature extractions. The code isavailable for reproducibility.

Automated Grading System of Retinal Arterio-venous Crossing Patterns: A  Deep Learning Approach Replicating Ophthalmologist's Diagnostic Process of  Arteriolosclerosis