Simple, Mobile-based Artificial Intelligence AlgoRithms in the Detection of Diabetic ReTinopathy (SMART) Study

Diabetic Retinopathy Clinical Trial

— SMART  

Official title:

Simple, Mobile-based Artificial Intelligence AlgoRithms in the Detection of Diabetic ReTinopathy (SMART) Study

Clinical Trial Details — Status: Recruiting

Administrative data

• NCT number: NCT03572699
• Other study ID #: SMART1
• Status: Recruiting
• Start date: July 11, 2018
• Est. completion date: October 2018

Study information

• Verified date: October 2018
• Source: Medios Technologies Pte. Ltd
• Contact: Srikanth Y N, MS
• Phone: 9886450550
• Email: srikanthyn[@]yahoo.com
• Is FDA regulated: No
• Study type: Observational

Clinical Trial Summary

This is an observational cross sectional study aimed to evaluate the performance of the artificial intelligence algorithm in detecting any grade of diabetic retinopathy using retinal images from patients with diabetes.

Recruitment information / eligibility

• Status: Recruiting
• Enrollment: 900
• Est. completion date: October 2018
• Est. primary completion date: October 2018
• Accepts healthy volunteers: No
• Gender: All
• Age group: 18 Years and older

Eligibility

Inclusion Criteria:

1. Patients with type 1 or type 2 diabetes mellitus

2. Ages 18 and above

3. Male and female

Exclusion Criteria:

1. Persistent visual impairment in one or both eyes;

2. Subjects with corneal opacities and advanced cataract.

3. History of retinal vascular (vein or artery) occlusion;

4. Subject is contraindicated for fundus photography (for example, has light sensitivity);

5. Subject is currently enrolled in an interventional study of an investigational device or drug;

6. Subject has a condition or is in a situation which in the opinion of the Investigator, might confound study results, may interfere significantly with the subject's participation in the study, or may result in ungradable clinical reference standard photographs.

Related Conditions & MeSH terms

• Diabetic Retinopathy
• Retinal Diseases

Intervention

Other:
• diabetic retinopathy
This is an observational study of patients with diabetes. Patients undergoing routine care will undergo retinal imaging using a nonmydriatic fundus camera. The images will be run on an artificial intelligence (AI) algorithm. The diagnosis of the artificial intelligence algorithm will be compared to the image diagnosis given by the ophthalmologists. The ophthalmologists will be blinded to the diagnosis of the AI and vice versa. The data will be analyzed to evaluate the performance of the AI.

Locations

Country	Name	City	State
India	Diacon Hospital	Bangalore

Sponsors (2)

Lead Sponsor	Collaborator
Medios Technologies Pte. Ltd	Diacon Hospital

Country where clinical trial is conducted

India, 

Outcome

Type	Measure	Description	Time frame	Safety issue
Primary	Sensitivity and specificity of the AI in detecting any grade of diabetic retinopathy		3 months
Secondary	Sensitivity and specificity of the AI in detecting referable diabetic retinopathy (referable retinopathy defined as moderate non proliferative retinopathy or greater)		3 months
Secondary	Sensitivity and specificity of the AI in detecting sight threatening diabetic retinopathy (referable retinopathy defined as severe non proliferative retinopathy or greater)		3 months