Unique ID issued by UMIN | UMIN000028843 |
---|---|
Receipt number | R000032802 |
Scientific Title | Multicenter validation study on computer-aided diagnostic system for endocytoscopic image in diagnosis of colrectal lesions |
Date of disclosure of the study information | 2017/09/04 |
Last modified on | 2017/11/10 18:29:51 |
Multicenter validation study on computer-aided diagnostic system for
endocytoscopic image in diagnosis of colrectal lesions
EndoBRAIN STUDY
Multicenter validation study on computer-aided diagnostic system for
endocytoscopic image in diagnosis of colrectal lesions
EndoBRAIN STUDY
Japan |
Colorectal lesion
Gastroenterology |
Others
NO
Compare the diagnostic capability with the endoscopist (non specialist) about the pathological diagnosis prediction ability of the endoscopic diagnosis support software EndoBRAIN.
Efficacy
1.Confirm the accuracy of diagnosis in the case of using a ultra-high magnified staining image in distinguishing neoplasm from non-neoplasm in colorectal lesion of 10 mm or less.(Verification that EndoBRAIN is superior to non-specialist in diagnostic accuracy)
2.Confirm the specificity of diagnosis in the case of using a ultra-high magnified staining image in distinguishing neoplasm from non-neoplasm in colorectal lesion of 10 mm or less.(Verification that EndoBRAIN is superior to non-specialist in diagnostic accuracy)
3.Confirm the accuracy of diagnosis in the case of using a ultra-high magnified narrow-banding image(NBI) in distinguishing neoplasm from non-neoplasm in colorectal lesion of 10 mm or less.(Verification that EndoBRAIN is superior to non-specialist in diagnostic accuracy)
4.Confirm the specificity of diagnosis in the case of using a ultra-high magnified narrow-banding image (NBI) in distinguishing neoplasm from non-neoplasm in colorectal lesion of 10 mm or less.(Verification that EndoBRAIN is superior to non-specialist in diagnostic accuracy)
1.Accuracy of distinguishing neoplasm from non-neoplasm of 10 mm or less colorectal lesion.
2.Whether either of diagnosis accuracy, sensitivity, specificity, PPV or NPV of EndoBRAIN exceeds 90% in distinguishing neoplasm from non-neoplasm in 10 mm or less lesion.
3.Whether either of diagnosis accuracy, sensitivity, specificity, PPV or NPV of EndoBRAIN exceeds 90% in distinguishing neoplasm from non-neoplasm in 5 mm or less lesion.
4.In case of diagnostic confidence is high, whether the negative predictive value for diagnosing 5 mm or less lesions in the rectum and sigmoid colon is more than 90%.
5.In case of diagnostic confidence is high, whether the each of accuracy, sensitivity, specificity, PPV or NPV value for diagnosing colorectal lesion using EndoBRAIN for distinguishing neoplasm from non-neoplasm in 10 mm or less lesion exceeds 90%.
6.In case of diagnostic confidence is high, whether the each of accuracy, sensitivity, specificity, PPV or NPV for diagnosing colorectal lesion using EndoBRAIN for distinguishing neoplasm from non-neoplasm in 5 mm or less lesion exceeds 90%.
7.Time required for diagnosis.
8.Reproducibility in EndoBRAIN diagnosis
9.Verification that EndoBRAIN is superior to non-specialist in results of specificity, PPV and NPV for distinguishing neoplasm from non-neoplasm in colorectal lesion of 10 mm or less.
10.Verification that EndoBRAIN has better diagnostic results than non-specialists in each of accuracy, sensitivity, specificity, PPV, NPV in distinguishing neoplasm from non-neoplasm in colorectal lesion of 5 mm or less.
11.In the case of diagnosis confidence is high, verification that EndoBRAIN has better diagnostic results than non-specialists in each of accuracy, sensitivity, specificity, PPV, NPV of distinguishing neoplasm from non-neoplasm of colorectal lesion of 5 mm or less.
*In the secondary outcomes, the diagnostic abilities of both the staining and narrow-band unltra-high magnifying image are each evaluated.
Observational
30 | years-old | <= |
80 | years-old | >= |
Male and Female
Colorectal lesions with a maximum diameter of 10 mm or less which were observed and resected by CF-Y0058I at the digestive desease center of Showa University Northern Yokohama Hospital, and were capable of pathological evaluation.
Regarding the use of image data, a case in which written consent was obtained from the patient himself before the endoscopic examination.
At the time of endoscopic image photographing, patients who are 30 to 80 years of age.
The leision from which white light image, ultra-high magnifying image and ultra-high magnifying NBI images are all acquired.
Lesions from patients with inflammatory bowel disease(broad sense).
Non-epithelial lesion(carcinoid tumor, malignant lymphoma)
Sessile serrated adenoma/polyp
In case that, each ultra-high magnifying image, ultra-high magnifying NBI and white light image can not be evaluated because of unclear image.
Because of strong image artifacts, diagnosis is impossible with EndoBRAIN(output as Not a Good Sample or Low confidence).
Lesions with multiple different pathological components within the same lesion.
30000
1st name | |
Middle name | |
Last name | Shin-ei Kudo |
Showa University Northern Yokohama Hospital
Digestive Disease Center
35-1 Chigasaki-chuo, Tsuzuki, Yokohama 224-8503
045-949-7000
kudos@med.showa-u.ac.jp
1st name | |
Middle name | |
Last name | Yuichi Mori |
Showa University Northern Yokohama Hospital
Digestive Disease Center
35-1 Chigasaki-chuo, Tsuzuki, Yokohama 224-8503
045-949-7000
ibusiginjp@gmail.com
Showa University Northern Yokohama Hospital
Japanese Agency for Medical Research and Development
Japanese Governmental office
NO
2017 | Year | 09 | Month | 04 | Day |
Unpublished
Open public recruiting
2017 | Year | 08 | Month | 14 | Day |
2017 | Year | 11 | Month | 10 | Day |
This study is a retrospective study using a web-based test.
2017 | Year | 08 | Month | 26 | Day |
2017 | Year | 11 | Month | 10 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000032802
Research Plan | |
---|---|
Registered date | File name |
Research case data specifications | |
---|---|
Registered date | File name |
Research case data | |
---|---|
Registered date | File name |