Assessment of Eyelid Topology and Kinetics Using Deep Learning in Eyelid Diseases

What this Trial Is About

Researchers are evaluating eyelid conditions using deep learning methods to automatically analyze eyelid shape and blinking patterns. The study aims to measure features like margin reflex distance, eyelid contour, corneal exposure area, and blinking frequency, velocity, and duration. It also seeks to predict different types of levator muscle function based on blinking characteristics. This observational research involves people with various eyelid diseases as well as normal volunteers. Participants include normal individuals and patients with blepharoptosis, blepharospasm, dry eye disease, or Graves' disease. Facial photographs and blinking videos are collected to capture eyelid topology and kinetics. These images and videos are then analyzed using deep learning techniques to extract one-dimensional and two-dimensional eyelid parameters and to classify levator muscle function subtypes. Throughout the study, which lasts up to 5 years, researchers will monitor eyelid features and blinking behavior using the collected visual data. The primary outcomes include detailed measures of eyelid shape and function assessed continuously during the study period. This long-term observation will help improve understanding of eyelid diseases and develop new methods for their assessment.

Assessment of Eyelid Topology and Kinetics Based on Deep Learning Method