AI-Empowered Fundus Platform

AI-Empowered Fundus Imaging: Building an Intelligent Platform for Diagnosis, Treatment, and Early Warning of Fundus Diseases

According to estimates by the World Health Organization, approximately half of the global adult population currently suffers from varying degrees of myopia. In urban areas of China, the myopia rate among primary school students has reached about 40%, and can exceed 80% in high school students. The proportion of patients with high myopia (refractive error greater than 600 degrees) is also increasing year by year. This group is more prone to severe visual problems, including an increased risk of complications such as cataracts, glaucoma, and macular degeneration. Pathological myopia is the main risk factor for vision loss caused by high myopia. In such patients, the elongation of the ocular axis leads to local thinning and protrusion of the posterior pole of the eyeball, forming a posterior scleral staphyloma. Posterior scleral staphyloma is the most representative pathological feature of pathological myopia. Local dilation of the choroid leads to thinning and stretching of the retina, which may ultimately cause problems such as macular retinal schisis, seriously affecting vision and even leading to blindness.

This project aims to develop a multimodal intelligent screening system by combining the excellent imaging technology and advanced artificial intelligence (AI) of the Optos non-mydriatic ultra-wide-angle laser scanning ophthalmoscope, B-scan ultrasonography ophthalmic diagnostic instrument, and optical coherence tomography (OCT). The system is designed to achieve precise identification and assessment of ocular fundus diseases, especially pathological myopia, and particularly the core condition of posterior scleral staphyloma (PSS). Simultaneously, a vision prediction model will be constructed to assist doctors in formulating personalized diagnosis and treatment strategies, predicting the trend of vision deterioration, and enhancing the effectiveness of early intervention. This system is expected to significantly improve the prevention and control of pathological myopia, reduce the risk of blindness, and play a pivotal role in telemedicine services, benefiting people in remote areas and promoting public health equity and service quality.

Study Overview

• Status: Recruiting
• Conditions: High Myopia
• Intervention / Treatment: Diagnostic test: Multimodal Ocular Imaging and Biometry with Aritifical Intelligence Analysis

• Study Type: Observational
• Enrollment (Estimated): 1000

Contacts and Locations

• Study Contact: Name: Yiwen Hu; Phone Number: +86 18019320181; Email: 1006108590@qq.com

Study Locations

• China
  • Shanghai, China
    • Recruiting
    • Shanghai 10th People's Hospital
    • Contact: Yiwen Hu; Phone Number: +86 18019320181; Email: 1006108590@qq.com

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult; Older Adult
• Accepts Healthy Volunteers: No

• Sampling Method: Non-Probability Sample

Study Population

This retrospective study enrolled patients diagnosed with high myopia and posterior staphyloma (PSS) who visited the Department of Ophthalmology, Shanghai Tenth People's Hospital before October 2024.

Sample Size：A total of 1,000 patients (600 for training, 200 for validation, and 200 for internal testing) from the hospital cohort, contributing approximately 6,000 multimodal images.

Description

Inclusion Criteria:

Patients with a diagnosis of cataract and high myopia and posterior staphyloma, who underwent comprehensive ophthalmic examinations including ultra-widefield fundus photography (Optos), B-scan ultrasonography, and optical coherence tomography (OCT). Baseline data, including demographics, health status, ophthalmic history, and ocular biometric parameters, were completely documented.

Exclusion Criteria:

Patients with poor-quality images affecting PSS identification, severe coexisting ocular diseases including glaucoma, diabetic retinopathy that cause media opacities, or severe systemic diseases that could interfere with the study outcomes.

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort	Intervention / Treatment
Cataract patients with posterior staphyloma due to high myopia	Diagnostic test: Multimodal Ocular Imaging and Biometry with Aritifical Intelligence Analysis; The patient underwent a complete set of fundus color photography, B-scan ultrasonography, and OCT imaging during the consultation, and preoperative baseline data, including demographic information, health status, ophthalmic history, and ocular biometric parameters, were collected and recorded in detail

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Data collection outcome	Demographic data clasification of posterior scleral staphyloma visual acuity and imaging data	From October 2024 to April 2026 (Data collection from patients who visited the hospital before October 2024)

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
AI modeling Analysis	Diagnostic performance of the AI model for posterior scleral staphyloma. Area under the receiver operating characteristic curve (AUC-ROC) of the AI model. Accuracy, sensitivity, and specificity of the model in detecting posterior scleral staphyloma. Comparison of AI performance and doctors' performance	From April 2026 to October 2026

Collaborators and Investigators

• Sponsor: Shanghai 10th People's Hospital

Study record dates

Study Major Dates

• Study Start (Actual): October 1, 2024
• Primary Completion (Estimated): March 1, 2026
• Study Completion (Estimated): December 1, 2026

Study Registration Dates

• First Submitted: February 14, 2026
• First Submitted That Met QC Criteria: February 14, 2026
• First Posted (Actual): February 20, 2026

Study Record Updates

• Last Update Posted (Actual): February 20, 2026
• Last Update Submitted That Met QC Criteria: February 14, 2026
• Last Verified: February 1, 2026

More Information

Terms related to this study

• Other Study ID Numbers: SHSY-LYZX-413

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No