Diabetes, Diabetic Retinopathy
Conditions
Keywords
Early Diagnosis, Screening
Brief summary
In this pivotal trial, we aim to perform a prospective study to find the efficacy of iPredict-DR, an artificial intelligence (AI) based software tool on early diagnosis of Diabetic Retinopathy (DR) in the primary care and endocrinology clinics. DR is one of the leading causes of blindness in the United States and other developed countries. Every individual with diabetes is at risk of DR. It does not show any symptoms until the disease is progressed to advanced stages. If the disease is caught at an early stage, it can be prevented, managed, or treated effectively. Currently, screening for DR is done by the Ophthalmologists, which is limited to areas with limited availability. This is also time-consuming and expensive. All of these can be complemented by automated screening and set up the screening in the primary care clinics.
Detailed description
In this pivotal trial, we aim to perform a prospective study to find the efficacy of iPredict-DR, an artificial intelligence (AI) based software tool on early diagnosis of Diabetic Retinopathy (DR) in the primary care and endocrinology clinics. DR is one of the leading causes of blindness in the United States and other developed countries. Every individual with diabetes is at risk of DR. It does not show any symptoms until the disease is progressed to advanced stages. If the disease is caught at an early stage, it can be prevented, managed, or treated effectively. Currently, screening for DR is done by the Ophthalmologists, which is limited to areas with limited availability. This is also time-consuming and expensive. All of these can be complemented by automated screening and set up the screening in the primary care clinics. American Academy of Ophthalmology has suggested a 5-level DR disease severity scale (No DR, Mild DR, Moderate DR, Severe DR or Proliferative DR) based on the abnormalities in the retina such as microaneurysms, exudates, hemorrhages, intraretinal microvascular abnormalities (IRMA), and neovascularization. Automated screening for Diabetic Retinopathy has a potential to identify people at risk of developing sight-threatening disease and save millions of dollars in healthcare costs. To accomplish this, it is crucial to perform large scale population screening to identify the individuals with mild or early DR and better predict those at risk of developing late stage DR. A system that takes advantage of telemedicine with automated DR screening in reaching the mass populations in both urban and rural areas with the patient convenience is currently not widely available. Considering this urgent need, iHealthScreen has developed an automated software tool for DR screening which is based on artificial intelligence (AI) and make it widely available in both urban and remote/rural areas and for large-scale screening through a telemedicine platform, and thereby have the potential to prevent blindness in diabetic patients.
Interventions
DEVICENo intervention
No intervention. Evaluate the automated DR screening software.
Sponsors
National Institutes of Health (NIH)
iHealthScreen Inc
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
22 Years to No maximum
Healthy volunteers
Yes
Inclusion criteria
* Age of Subjects: Patients ≥ 22 years of age. * Gender of Subjects: Both males and females will be invited to participate. * Subjects with diabetes (A1C level ≥ 6.5). * Subjects must be willing and are able to comply with clinic visit, understand the study-related procedures/provisions, and provide signed informed consent.
Exclusion criteria
* Unable to understand the study, Our unable to or unwilling to sign the informed consent * Previously diagnosed with macular edema, any form of diabetic retinopathy, radiation retinopathy, or retinal vein occlusion * participants who are experiencing persistent vision loss, blurred vision, or other vision problems that should be evaluated by an eye care provider * subjects whose retinal images were used in training, validating, or developing the device * Currently participating in another investigational eye study or actively receiving investigational product for DR or DME. * A condition that, in the opinion of the investigator, would preclude participation in the study; * Contraindicated for imaging by fundus imaging systems used in the study because of hypersensitivity to light, recently underwent photodynamic therapy, or was taking medication that causes photosensitivity.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| mtmDR detected (Referable DR) OR mtmDR not detected (non-referable DR) | 1-year or 2-year | Sensitivity and specificity of identification of referable and non-referable DR for early diagnosis of DR using the iPredict-DR's AI-based DR screening software utilizing color fundus imaging. iPredict-DR can detect non-referable DR (normal retina or mild DR) and referable DR (moderate or severe DR including moderate non-proliferative, proliferative DR and diabetic macular edema) at a similar level of expert ophthalmologists. For this, the healthcare workers will be taking the disc and macula center 45-degree field view images using DRSPlus camera (from iCare Inc.). The output of AI model and ground truth (produced by graders from reading centers) will be compared for image level and subject level accuracy measurements. The worst eye will be considered to define a subject's referability or non-referability to an ophthalmologist. Using the ground truth/gold standard, the sensitivity, specificity, precision, recall, accuracy, F-measure, positive predictive value and negative predictive |
| The accuracy of the iPredict-DR software developed by iHealthScreen system in early diagnosis of DR using color retinal photos vs. that of human expert graders | 1-year or 2-year | The accuracy of the iPredict-DR software developed by iHealthScreen system in early diagnosis of DR using color retinal photos vs. that of human expert graders for DR. Performance thresholds were defined at 80.0% for sensitivity and 80.0% for specificity |
Countries
United States
Contacts
Primary ContactAlauddin Bhuiyan
Backup ContactFariha Nuha, BS (Comp. Biology)
Outcome results
None listed