UMIN-ICDS Clinical Trial

Public title

Analysis of Endoscopist's Visual Gaze Pattern during Artificial Intelligence Assisted Colorectal Polyp Detection

Acronym

Analysis of endoscopist's VGP during AI-assisted colorectal polyp detection

Scientific Title

Analysis of Endoscopist's Visual Gaze Pattern during Artificial Intelligence Assisted Colorectal Polyp Detection

Scientific Title:Acronym

Analysis of endoscopist's VGP during AI-assisted colorectal polyp detection

Region

Japan

Condition

Colorectal polyp

Classification by specialty

Gastroenterology

Classification by malignancy

Others

Genomic information

NO

Narrative objectives1

The aim of this study is to reveal the endoscopist's visual gaze pattern (VGP) in real time during the detection of colorectal polyps with computer-aided detection (CADe).
The detail targets for the analysis are followings;

1. Percentage of lesions automatically detected by CADe that were actually recognized by the endoscopist.
2. Time required for the endoscopist to recognize a lesion that CADe was able to detect automatically.
3. Characteristics of lesions automatically detected by CADe and recognized by the endoscopist.
4. Characteristics of lesions that CADe could automatically detect and the endoscopist could not recognize.
5. Characteristics of lesions that CADe could not automatically detect and endoscopist could recognize.
6. Probability that the endoscopist gazed at the area where CADe falsely detected bubbles or remaining stools.

The main objective of the study is to determine how many of the lesions identified by CADe can be followed by the endoscopist through these analyses. We also aim to characterize which lesions are or are not detectable by CADe. We also aim to evaluate the performance of currently available CADe systems.

Basic objectives2

Efficacy

Basic objectives -Others

Trial characteristics_1

Exploratory

Trial characteristics_2

Explanatory

Developmental phase

Not applicable

Primary outcomes

Percentage of lesions that CADe was able to automatically detect that were actually recognized by the endoscopist

Key secondary outcomes

1. Time taken for the endoscopist to recognize lesions that were automatically detected by CADe
2. Analysis of features of lesions automatically detected by CADe but not recognized by the endoscopist
3. Analysis of lesion features that the endoscopist could recognize but CADe could not automatically detect
4. Percentage of endoscopists gazing at sites where CADe falsely detected bubbles or remaining stools (percentage of endoscopists gazing at CADe falsely detected sites)
5. Analysis of patients background in relation to the percentage that CADe falsely detected sites
6. Analysis of VGP of endoscopists during lesion detection by CADe

Study type

Interventional

Basic design

Single arm

Randomization

Non-randomized

Randomization unit

Blinding

Open -no one is blinded

Control

Uncontrolled

Stratification

Dynamic allocation

Institution consideration

Blocking

Concealment

No. of arms

1

Purpose of intervention

Diagnosis

Type of intervention

Device,equipment

Interventions/Control_1

During colonoscopy, CADe (CAD-EYE, Fujifilm Corporation) and the VGP real-time analysis system we developed are operated simultaneously to record and store inspection movies with the endoscopist's VGP displayed on the endoscopic image.

Interventions/Control_2

Interventions/Control_3

Interventions/Control_4

Interventions/Control_5

Interventions/Control_6

Interventions/Control_7

Interventions/Control_8

Interventions/Control_9

Interventions/Control_10

Age-lower limit

20

years-old

<=

Age-upper limit

90

years-old

>

Gender

Male and Female

Key inclusion criteria

1. Patients who are undergoing colonoscopy.
2. Patients whose age is between 20 and 90 years old.
3. Patients with a performance status (ECOG) of 0 (no limitation in daily activities), 1 (able to perform light tasks but not physical labor), or 2 (able to walk and perform personal activities but not light tasks)
4. Patient's participation in the study has been fully explained to him/her and his/her written consent has been obtained.

Key exclusion criteria

1. Patients with a history of inflammatory bowel disease (ulcerative colitis, Crohn's disease, Behcet's disease).
2. Patients with hereditary or non-hereditary gastrointestinal polyposis. 3) Patients with hereditary non-polyposis colorectal cancer (Lynch syndrome).
3. Patients with hereditary non-polyposis colorectal cancer (Lynch syndrome).
4. Patients with known severe diverticular disease of the colon that makes colonoscopy difficult and dangerous.
5. Patients who are unable to take colonoscopy pretreatment medication (laxatives).
6. Patients who are allergic to colonoscopy pretreatment drugs or sedatives.
7. Pregnant women.
8. Those who are breast-feeding.
9. Those who have not obtained consent to participate in the study
10. Those who are deemed inappropriate by the study investigator.

Target sample size

60

Name of lead principal investigator

1st name	Fumiaki
Middle name
Last name	Ishibashi

Organization

International University of Health and Welfare Ichikawa Hospital

Division name

Department of Gastroenterology

Zip code

272-0827

Address

6-1-14, Konodai, Ichikawa-shi, Chiba

TEL

047-375-1111

Email

ishibashi-gast@iuhw.ac.jp

Name of contact person

1st name	Fumiaki
Middle name
Last name	Ishibashi

Organization

International University of Health and Welfare Ichikawa Hospital

Division name

Department of Gastroenterology

Zip code

272-0827

Address

6-1-14, Konodai, Ichikawa-shi, Chiba

TEL

047-375-1111

Homepage URL

Email

ishibashi-gast@iuhw.ac.jp

Institute

International University of Health and Welfare

Institute

Department

Personal name

Organization

None

Organization

Division

Category of Funding Organization

Other

Nationality of Funding Organization

Co-sponsor

Name of secondary funder(s)

Organization

International University of Health and Welfare, Chiba Area Ethics Committee

Address

4-3, Kozunomori, Narita-shi, Chiba

Tel

0476-20-7708

Email

rinri_md@iuhw.ac.jp

Secondary IDs

NO

Study ID_1

Org. issuing International ID_1

Study ID_2

Org. issuing International ID_2

IND to MHLW

Institutions

Date of disclosure of the study information

2023

Year

11

Month

25

Day

URL releasing protocol

Publication of results

Unpublished

URL related to results and publications

Number of participants that the trial has enrolled

Results

Results date posted

Results Delayed

Results Delay Reason

Date of the first journal publication of results

Baseline Characteristics

Participant flow

Adverse events

Outcome measures

Plan to share IPD

IPD sharing Plan description

Recruitment status

No longer recruiting

Date of protocol fixation

2023

Year

08

Month

17

Day

Date of IRB

2023

Year

10

Month

31

Day

Anticipated trial start date

2023

Year

11

Month

01

Day

Last follow-up date

2024

Year

03

Month

30

Day

Date of closure to data entry

2024

Year

03

Month

30

Day

Date trial data considered complete

Date analysis concluded

Other related information

Registered date

2023

Year

11

Month

24

Day

Last modified on

2024

Year

03

Month

30

Day

Link to view the page

Value
https://center6.umin.ac.jp/cgi-open-bin/icdr_e/ctr_view.cgi?recptno=R000060350