This study aims to develop the deep learning-based system that integrates multimodal information to automatically diagnose, stage, and grade thyroid-associated ophthalmopathy (TAO), assist in treatment decision-making, and predict patient prognosis, thereby providing novel tools to support clinical management.
Study Type
OBSERVATIONAL
Enrollment
500
2D and 3D Facial photography, blinking videos, orbital CT and MRI images, fundus images and images of nine gaze positions are taken. Clinical information is recorded. Quality of life is assessed. Ocular examinations, such as visual acuity, levator function, exophthalmos measurement, strabismus test, visual field examination, axial length measurement, OCT, OCTA are made.
Juan Ye
Hangzhou, Zhejiang, China
RECRUITINGeye morphological parameters
Margin reflex distances(mm)
Time frame: through study completion, average of 5 years
Blinking feature of patients with thyroid-associated ophthalmopathy
Blinking frequency
Time frame: through study completion, average of 5 years
Ocular movement measurement
Ocular movement ability in nine gaze positions
Time frame: through study completion, average of 5 years
Automatic orbital structures segmentation and measurement using deep learning methods
Automatic segmentation and measurement of orbital structures,such as eyeball, extraocular muscle, orbital fat, bony orbital region, optic nerves, etc
Time frame: through study completion, average of 5 years
Automatic diagnosis, grading, staging and prediction of patients with thyroid-associated ophthalmopathy
Patients with thyroid-associated ophthalmopathy(TAO) are classified as TAO or non-TAO. Grading is classified as mild, moderate, severe and sight-threatening. Staging is classified as active and stable.
Time frame: through study completion, average of 5 years
three-dimensional eye parameters
eyelid volume
Time frame: through study completion, average of 5 years
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.