Evaluation of AI-assisted diagnosis software for colonoscopy images.

Evaluation of computer-aided diagnosis system for colonic narrow-band imaging using artificial intelligence - Evaluation of AI for colonic NBI images

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

JPRN

Registry ID

JPRN-UMIN000044609

Enrollment

400

Registered

2021-08-01

Start date

2021-06-01

Completion date

Unknown

Last updated

2026-06-29

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

Conditions

Colonic epithelial lesion

Interventions

None listed

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: 1. 20 years old or older at the time of examination 2. Lesions with clear NBI images 3. Lesions that have been resected and pathological diagnosis has been obtained

Exclusion criteria

Exclusion criteria: 1. Patients who refuse to participate. 2. Lesions in patients with inflammatory bowel disease in a broad sense 3. Lesions not captured by NBI imaging

Design outcomes

Primary

Measure	Time frame
Sensitivity of AI for colonic adenoma (=<10mm)	—

Secondary

Measure	Time frame
1. Specificity, accuracy, NPV and PPV of AI for colonic adenoma (=<10mm) 2. Sensitivity, specificity, accuracy, NPV and PPV of AI for colonic non-neoplasms (=<10mm) 3. Sensitivity, specificity, accuracy, NPV and PPV of AI for colonic sesile serrated adenoma/polyp (=<10mm) 4. Sensitivity, specificity, accuracy, NPV and PPV of AI by the morphology of the lesions.	—

Countries

Japan

Contacts

Public ContactMasashi Misawa

Showa University Northern Yokohama Hospital Digestive Disease Center

mmisawa@med.showa-u.ac.jp045-949-7000

Outcome results