Computer Aided Diagnosis of Multiple Eye Fundus Diseases From Color Fundus Photograph

NCT04723160 | Completed DMC

• 1 other identifier: BLG-20200101
• Study Type: observational
• Target: 748
• Locations: 1 country
• Sites: 5

Brief Summary

Blindness can be caused by many ocular diseases, such as diabetic retinopathy, retinal vein occlusion, age-related macular degeneration, pathologic myopia and glaucoma. Without timely diagnosis and adequate medical intervention, the visual impairment can become a great burden on individuals as well as the society. It is estimated that China has 110 million patients under the attack of diabetes, 180 million patients with hypertension, 120 million patients suffering from high myopia and 200 million people over 60 years old, which suggest a huge population at the risk of blindness. Despite of this crisis in public health, our society has no more than 3,000 ophthalmologists majoring in fundus oculi disease currently. As most of them assembling in metropolitan cities, health system in this field is frail in primary hospitals. Owing to this unreasonable distribution of medical resources, providing medical service to hundreds of millions of potential patients threatened with blindness is almost impossible. To solve this problem, this software (MCS) was developed as a computer-aided diagnosis to help junior ophthalmologists to detect 13 major retina diseases from color fundus photographs. This study has been designed to validate the safety and efficiency of this device.

Conditions

Diabetic Retinopathy; Retinal Vein Occlusion; Retinal Artery Occlusion; Central Serous Chorioretinopathy; Pathologic Myopia; Retinitis Pigmentosa; Epiretinal Membrane; Macular Holes; Nonexudative Age-related Macular Degeneration; Exudative Age Related Macular Degeneration; Suspect Glaucoma; Optic Atrophy; Retinal Detachment

Keywords

multiple eye fundus diseases

Outcome Measures

Primary Outcomes (1)

• consistent rate of diagnoses

  Formula for calculation: consistent rate of diagnoses=number of images with consistent diagnosis/ total number of images × 100%. Method: the diagnoses from the test group and the control group were compared with diagnoses from the gold standard. For each image, if one or more diagnoses were consistent with those of the gold standard, which means at least one label existed in the intersection of diagnoses from the test group(or the control group)and those from the gold standard, it would be classified as "image with consistent diagnosis". Otherwise, it would be classified as "image without consistent diagnosis". After above-mentioned steps, the investigators had obtained the number of images with consistent diagnosis in each group. As images with 1-2 labels account for the majority in actual work, the investigators stipulated that each image in both groups could be marked with 3 labels at most in case of invalid improvement in consistent rate owing to multiple selections.

  through study completion, an average of 1 year

Secondary Outcomes (3)

• sensitivity and specificity of software's diagnoses for each diseases

  through study completion, an average of 1 year

• PPV and NPV of software's diagnoses for each diseases

  through study completion, an average of 1 year

• full coincidence rate of software's diagnoses

  through study completion, an average of 1 year

Study Arms (2)

Test Group

ophthalmologists read images applying the assistant software

Diagnostic Test: Software assisted imaging diagnosis

Control Group

ophthalmologists read images independently

Interventions

Software assisted imaging diagnosis DIAGNOSTIC_TEST

In the test group, diagnoses are given with the help of the software.

Test Group

Eligibility Criteria

• Age: 18 Years - 75 Years
• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Adult (18-64), Older Adult (65+)
• Sampling Method: Non-Probability Sample

Study Population

Any patients who meet the eligibility criteria.

You may qualify if:

• Age between 18 and 75.
• Anyone need to take fundus photograph in clinical.
• Understand the study and volunteer to sign the informed consent.
• For fundus images of participants, the optic disc, fovea, the upper and lower vessel bow should be included in the fundus field.

You may not qualify if:

• Participants has any eye that cannot take fundus photos.
• Participants have joined or is participating in other clinical trial within one month.
• Participants who have any other issue that cannot be enrolled.
• Participants with cloudy refractive media that cannot take fundus photos or get clouding fundus photos.
• Participants with low quality fundus photos like incompetent vision field, overexposed/underexposed, out of focus, too many shadow or dirties and so on.

Sponsors & Collaborators

• Visionary Intelligence Ltd.: lead
• Peking University First Hospital: collaborator
• Beijing Municipal Science & Technology Commission: collaborator

Study Sites (5)

Peking Union Medical College Hospital, Chinese Academy of Medical Sciences

Beijing, Beijing Municipality, 100730, China

Location

The Second Hospital of Hebei Medical University

Shijiazhuang, Hebei, China

Location

West China Hospital of Sichuan University

Chengdu, Sichuan, China

Location

Tianjin Medical University Eye Hospital

Tianjin, Tianjin Municipality, China

Location

Eye Hospital, WMU Zhejiang Eye Hospital

Wenzhou, Zhejiang, China

Location

Related Publications (1)

• Li B, Chen H, Yu W, Zhang M, Lu F, Ma J, Hao Y, Li X, Hu B, Shen L, Mao J, He X, Wang H, Ding D, Li X, Chen Y. The performance of a deep learning system in assisting junior ophthalmologists in diagnosing 13 major fundus diseases: a prospective multi-center clinical trial. NPJ Digit Med. 2024 Jan 11;7(1):8. doi: 10.1038/s41746-023-00991-9.

  PMID: 38212607 DERIVED

MeSH Terms

Conditions

Diabetic Retinopathy; Retinal Vein Occlusion; Retinal Artery Occlusion; Central Serous Chorioretinopathy; Retinitis Pigmentosa; Epiretinal Membrane; Retinal Perforations; Ocular Hypertension; Optic Atrophy; Retinal Detachment

Condition Hierarchy (Ancestors)

Retinal Diseases; Eye Diseases; Diabetic Angiopathies; Vascular Diseases; Cardiovascular Diseases; Diabetes Complications; Diabetes Mellitus; Endocrine System Diseases; Venous Thrombosis; Thrombosis; Embolism and Thrombosis; Arterial Occlusive Diseases; Eye Diseases, Hereditary; Retinal Dystrophies; Retinal Degeneration; Genetic Diseases, Inborn; Congenital, Hereditary, and Neonatal Diseases and Abnormalities; Optic Nerve Diseases; Cranial Nerve Diseases; Nervous System Diseases

Study Officials

• You xin Chen, PHD

  Peking Union Medical College

  STUDY CHAIR

Study Design

• Study Type: observational
• Observational Model: CASE CONTROL
• Time Perspective: PROSPECTIVE
• Sponsor Type: INDUSTRY
• Responsible Party: SPONSOR

Study Record Dates

• First Submitted: January 17, 2021
• First Posted: January 25, 2021
• Study Start: August 10, 2020
• Primary Completion: March 10, 2021
• Study Completion: May 30, 2021
• Last Updated: December 30, 2021

Record last verified: 2021-12

Locations

CN (5)