Pivotal Trial of an Automated AI-based System for Early Diagnosis and Prediction of Late Age Related Macular Degeneration in Primary Care Settings.
试验速览
- 阶段
- 不适用
- 入组人数
- 1,000
- 试验地点
- 1
- 主要终点
- Specificity of identification of referable and non-referable AMD for early diagnosis of AMD using the iPredict's AI-based AMD screening software utilizing color fundus imaging.
研究概览
简要总结
This study seeks to evaluate a system for the automated early detection of Age-Related Macular Degeneration (AMD). AMD is a condition in which there is breakdown of the macula of the eye, the part of the retina that is responsible for sharp, central vision. We will take pictures of subjects' eyes using an automated camera. These photographs will be securely transmitted and and then analyzed by a computer program which has been developed in other studies. The outcome of the computer program analysis will be compared with human analysis of these same pictures. If the computer analysis is has good enough accuracy, then this computer system could be used for wide-scale screening for AMD.
详细描述
iPredict,an AI and telemedicine based software which used individual's color fundus image for early diagnosis of AMD and predict if an individual is at risk of progression to late AMD. iPredict platform integrates the server-side programs (the image analysis and deep-learning modules for AMD severity screening and prediction) and local remote computer/mobile devices (for collecting patient data and images). DRS plus camera will be used in the doctor's office. The remote devices will upload images and data to the server to analyze and screen AMD automatically. The telemedicine platform has been developed for web-based platform. The automatic analysis will be performed on the server, and a report will be sent to the patient/remote devices with an individual's AMD stage as referable or non-referable AMD, and a risk prediction score of developing late AMD (within a minute), and further recommendations to visit a nearby ophthalmologist.
研究设计
- 研究类型
- Observational
- 观察模型
- Cohort
- 时间视角
- Prospective
入排标准
- 年龄范围
- 50 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 是
入选标准
- •Subjects will be recruited if willing and able to comply with clinic visit and study-related procedures, and provide signed informed consent
- •Gender of Subjects: Both males and females will be invited to participate.
- •Age of Subjects: Patients will be over 50 years and older
排除标准
- •Unable to provide informed consent.
- •Other retinal degenerations and retinal vascular diseases such as diabetic retinopathy or macular edema, prior retinal surgery.
结局指标
主要结局
Specificity of identification of referable and non-referable AMD for early diagnosis of AMD using the iPredict's AI-based AMD screening software utilizing color fundus imaging.
时间窗: 2 years
Using the gold standard (i.e., the ophthalmologist's grading), the sensitivity and specificity are calculated as: Sens=TP/(TP+FN) Spec=TN/(TN+FP) Where TP is the number of true positives (referable AMD subjects correctly classified), FN is the number of false negatives (referable AMD subjects incorrectly classified as non-referable), TN is the number of true negatives (non-referable subjects correctly classified), and FP is the number of false positives (non-referable AMD subjects incorrectly classified as referable AMD).
Sensitivity of identification of referable and non-referable AMD for early diagnosis of AMD
时间窗: 2 years
Sensitivity of identification of referable and non-referable AMD for early diagnosis of AMD using the iPredict's AI-based AMD screening software utilizing color fundus imaging.
次要结局
未报告次要终点