Skip to content

Screening and identification of gastrointestinal diseases by deep learning using fundus images: a prospective multicenter study

Screening and identification of gastrointestinal diseases by deep learning using fundus images: a prospective multicenter study

Status
Recruiting
Phases
Early Phase 1
Study type
Observational
Source
ChiCTR
Registry ID
ChiCTR2100050640
Enrollment
Unknown
Registered
2021-09-01
Start date
2021-09-06
Completion date
Unknown
Last updated
2022-05-09

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

gastrointestinal diseases

Interventions

Gold Standard:Endoscopic imaging and biopsy of digestive tract
Index test:Artificial&#32
intelligence&#32
system&#32
based&#32
on&#32

Sponsors

Tongji Hospital, Tongji Medical College of HUST
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to 80 Years

Inclusion criteria

Inclusion criteria: 1.Aged >= 18 years. 2.normal eye position and good central fixation point for fundus photography. 3.more than 90% of the fundus image areas included the four main areas (optic disc, macular, and upper and lower retinal vascular arch), which were easy to read and discriminate.

Exclusion criteria

Exclusion criteria: 1.Inability to cooperate with fundus photography; 2.Images with eyelid occlusion, with light leakage (>10% of the area), lens flare or smudge, and overexposure were excluded from further analysis; 3.A serious and uncontrollable systemic disease.

Design outcomes

Primary

MeasureTime frame
area under the receiver operating characteristic curve of the deep learning system;

Secondary

MeasureTime frame
sensitivity and specificity of the deep learning system;

Countries

China

Contacts

Public ContactSun Xufang

Tongji Hospital, Tongji Medical College of HUST

s_1862777@163.com+86 27 83663223

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Feb 4, 2026