A Single Center Study on the Effectiveness and Safety of Polyp Classification With Artificial Intelligence

• Conditions: Colonic Diseases; Artificial Intelligence

• Registration Number: NCT04216901
• Lead Sponsor: Renmin Hospital of Wuhan University

Brief Summary

This is an artificial intelligence-based optical endoscopic polyp diagnosis system that can assist endoscopic doctors in diagnosing polyps and improve the quality of training in clinical Settings.

Detailed Description

Large sessile and laterally spreading colorectal lesions(LSLs) are increasingly encountered during colonoscopy. LSLs have an increased risk of harbouring invasive cancer and can be challenging to excise endoscopically. Wide-field endoscopic mucosal resection (WF-EMR) is widely used in treating LSLs. In the East, meanwhile endoscopic submucosal dissection (ESD) is the dominant technique due to its ability to achieve en bloc resection in over 80% of cases. Many papers have demonstrated that selective-esd has the highest economic benefit. The key is to find a reliable way to select.

Study Timeline

• Study Start: 2019-12-24
• Study First Submitted: 2019-12-31
• Study First Submitted QC: 2019-12-31
• Study First Posted: 2020-01-03
• Status Verified: 2020-10
• Last Update Submitted: 2020-10-19
• Last Update Posted: 2020-10-20
• Primary Completion: 2020-12-31
• Study Completion: 2021-06-30

Recruitment & Eligibility

• Status: UNKNOWN
• Sex: All
• Target Recruitment: 70

Inclusion Criteria

1. male or female aged 18 or above;
2. colonoscopy and related examinations should be performed to further clarify the characteristics of digestive tract diseases;
3. be able to read, understand and sign the informed consent;
4. the researcher believes that the subject can understand the process of the clinical study, is willing and able to complete all the study procedures and follow-up visits, and cooperate with the study procedures;
5. patients with > 1cm lesion detected by colonoscopy, requiring magnification staining or surgical resection.

Read More

Exclusion Criteria

1. have participated in other clinical trials, signed the informed consent and have been in the follow-up period of other clinical trials;
2. drug or alcohol abuse or psychological disorder in the last 5 years;
3. pregnant or nursing women;
4. subjects with previous history of intestinal surgery;
5. the researcher considers that the subject is not suitable for colonoscopy and related examination;
6. high-risk diseases or other special conditions that the investigator considers inappropriate for the subject to participate in the clinical trial.

Read More

Study & Design

• Study Type: OBSERVATIONAL
• Study Design: Not specified

Primary Outcome Measures

Name	Time	Method
Accuracy of evaluating the feasibility of selective-ESD	2019.12.24-2020.12.31	Calculate the accuracy of ai's judgment on whether ESD should be implemented.Accuracy is: the machine over a period of time to judge the results are consistent with the pathological lesion number of molecules, all lesions detected by a period of time for the denominator expressed as a percentage.The gold standard is the pathological results of diagnostic treatment. After the specimen was removed, the area suspected by endoscopists of early cancer was marked with Indian ink for pathological recovery.The specimens were then placed in formalin and fixed for 24 hours until the ink was a little dry.Even if the specimen is cut into 2mm-wide shapes, the suspected area can be identified by Indian ink staining under a microscope.The doctor suspected cancer patients were followed up for 60 days.
Accuracy of Vision location	2019.12.24-2020.12.31	Calculate the accuracy of the machine in locating the field of vision.The accuracy was as follows: the visual field localization results of the machine on the ESD intraoperative lesion screen captures were the numerator consistent with the number of visual fields determined by multiple endoscopists, and the number of visual fields of all localization in the same operation was the denominator, and the result was expressed as a percentage.The consistent results of visual field positioning by multiple endoscopic physicians watching the operation video were the gold standard.Patients with suspected cancer were followed up for 60 days, and the most serious pathological diagnosis within 60 days was taken as the diagnosis of the patient's disease.Patients whose doctors deemed no risk were followed until the end of colonoscopy.

Secondary Outcome Measures

Name	Time	Method
Consistent of classification among different endoscopists	2019.12.24-2020.12.31	Consistent of classification among different endoscopists
Consistent of vision positioning between diagnostic system and endoscopists	2019.12.24-2020.12.31	Consistent of vision positioning between diagnostic system and endoscopists
Consistent of classification between diagnostic system and endoscopists	2019.12.24-2020.12.31	Consistent of classification between diagnostic system and endoscopists

Trial Locations

Locations (1)

Renmin Hospital of Wuhan University; 🇨🇳 Wuhan, Hubei, China