| Unique ID issued by UMIN | UMIN000062517 |
|---|---|
| Receipt number | R000071549 |
| Scientific Title | The study of AI based Vascular healing in disease monitoring of UC: a multi-center prospective cohort study |
| Date of disclosure of the study information | 2026/08/07 |
| Last modified on | 2026/08/07 17:59:15 |
The study of AI based Vascular healing in disease monitoring of UC: a multi-center prospective cohort study
AI-VISION Trial
The study of AI based Vascular healing in disease monitoring of UC: a multi-center prospective cohort study
AI-VISION Trial
| Japan |
Ulcerative colitis
| Gastroenterology |
Others
NO
To evaluate the predictive performance and risk stratification ability of an artificial intelligence based narrow band imaging system for relapse within 12 months as a clinically important outcome
Efficacy
Exploratory
Time from colonoscopy to clinical relapse within 12 months
Clinical relapse is defined using the patient reported outcome measure PRO 2 as a rectal bleeding score of 1 or higher and a stool frequency score of 2 or higher
The predictive performance and risk stratification ability of vascular healing assessment by the artificial intelligence narrow band imaging system and histological assessment using the Geboes grade for clinical relapse within 12 months will be evaluated using the C statistic for time to event data
The 12 month relapse rate will be estimated using the Kaplan Meier method and stratified by vascular healing assessment using the artificial intelligence narrow band imaging system Geboes score and Nancy Histological Index Additional subgroup analyses will be performed according to clinical characteristics including treatment disease extent and duration of remission
An artificial intelligence based quantitative score will be calculated using white light endoscopic images and its association with relapse within 12 months will be evaluated The correlation between this score and established endoscopic scores including the Mayo Endoscopic Subscore and Ulcerative Colitis Endoscopic Index of Severity will also be evaluated
An artificial intelligence based digital pathology system will be applied to histopathological specimens to quantify neutrophil infiltration and colonic mucin content The associations between these quantitative measures and relapse within 12 months will be evaluated
Observational
| 18 | years-old | <= |
| Not applicable |
Male and Female
1. Age 18 years or older
2. Confirmed diagnosis of ulcerative colitis
3. Steroid-free clinical remission, defined as a partial Mayo score of 1 or less, maintained for at least 6 months
4. Scheduled to undergo colonoscopy based on clinical necessity
1. Unable to provide informed consent
2. Contraindication to colonoscopy or biopsy
3. Inadequate bowel preparation
4. Serious comorbidities
5. Pregnancy or breastfeeding
6. Change or escalation of treatment within the previous 3 months
7. Ulcerative proctitis
8. Considered unsuitable for participation by the investigator
300
| 1st name | Kazuo |
| Middle name | |
| Last name | Ohtsuka |
Showa Medical University Northern Yokohama Hospital
Digestive Disease Center
224-8503
35-1 Chigasaki-chuo, Tsuzuki, Yokohama, Kanagawa, Japan
045-949-7000
ohts@med.showa-u.ac.jp
| 1st name | Yasuharu |
| Middle name | |
| Last name | Maeda |
Showa Medical University Northern Yokohama Hospital
Digestive Disease Center
224-8503
35-1 Chigasaki-chuo, Tsuzuki, Yokohama, Kanagawa, Japan
045-949-5840
yasuharu@med.showa-u.ac.jp
Showa Medical University
Self funding
Self funding
Ethics Committee for Research Involving Human Subjects at Showa Medical University
1-5-8 Hatanodai, Shinagawa-ku, Tokyo 142-8555, Japan
03-3784-8129
m-rinri@ofc.showa-u.ac.jp
NO
昭和医科大学横浜市北部病院(神奈川県)、横浜市立市民病院(神奈川県)、東京科学大学病院(東京都)、旭川医科大学(北海道)、北里大学北里研究所病院(東京都)、大船中央病院(神奈川県)、昭和医科大学江東豊洲病院(東京都)
| 2026 | Year | 08 | Month | 07 | Day |
Unpublished
Open public recruiting
| 2026 | Year | 05 | Month | 07 | Day |
| 2026 | Year | 07 | Month | 31 | Day |
| 2026 | Year | 08 | Month | 07 | Day |
| 2029 | Year | 03 | Month | 31 | Day |
This is a multicenter prospective observational cohort study. At baseline colonoscopy performed as part of routine clinical care, white-light and NBI images, histological specimens, patient characteristics, treatment information, and laboratory data are collected. After colonoscopy, vascular healing is assessed using the AI-NBI system, an AI-based quantitative score is calculated from white-light images, and histological activity is assessed using the Geboes score and Nancy Histological Index. Participants are followed for 12 months from baseline colonoscopy. PRO-2, treatment changes, hospitalization, surgery, and clinical relapse are assessed during routine clinical visits. The predictive performance of AI-NBI vascular healing assessment and histological assessment for clinical relapse within 12 months is compared.
| 2026 | Year | 08 | Month | 07 | Day |
| 2026 | Year | 08 | Month | 07 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000071549