EndoScreener lowered the adenoma miss rate in China's review

In the maker's randomized tandem trial that China's CMDE reviewed, the first colonoscopy missed 20.12% of adenomas with EndoScreener, against 31.25% without it.

  • Not rated an organization

    The lower miss rate was measured in the maker's pre-market trial, in which every patient also had the other exam, and the report gives no result from use after approval.

Depth runs from 1, a convenience or time saved, to 5, a life or a livelihood preserved.

Status
Intended benefit.
Evidence

foundstudiedreportedclaimed

An independent regulatory document supports the benefit.

Compared with

Colonoscopy without the software as the first exam, with each patient also having the other exam in tandem. Measured.

Without the system
31.25% adenoma miss rate with the standard exam first
With it
20.12% adenoma miss rate with the software first
The system’s part
Patients were randomized to have the exam with the software first or the standard exam first, so the lower miss rate is attributed to the software. The trial was the maker's. The report says that the product uses a deep learning algorithm.
Not yet checked
LimitsCosts and harmsLearns from dataLater outcome

Download data Cite Suggest a correction

Benefit type
Improved Task Accuracy
System
EndoScreener (肠息肉电子结肠内窥镜图像辅助检测软件)
Made by
Chengdu Wision Medical Device Co., Ltd. (成都微识医疗设备有限公司, the applicant) and Shanghai Wision AI
Used by
Chengdu Wision Medical Device (registration trial with 223 patients at four US academic medical centers)
Where
Four US academic medical centers (CMDE's report says only that they were outside China)
Stage
In a pilot
Date
, when the source was published
Kind
Aggregate result
Broad benefit
Identifying Objects, Actions, and Events, Monitoring Processes, Materials, or Surroundings
Collections
Scrutiny and findings, Cross-border records
Last revised
1 November 2026

Sources

  1. Peer-reviewed study, by an evaluator that the makers or users commissioned.

    Glissen Brown JR, Mansour NM, Wang P, et al. Deep Learning Computer-aided Polyp Detection Reduces Adenoma Miss Rate: A United States Multi-center Randomized Tandem Colonoscopy Study (CADeT-CS Trial). Clinical Gastroenterology and Hepatology 20(7):1499-1507.e4 (2022), published online 14 September 2021. doi:10.1016/j.cgh.2021.09.009.

    Patients were enrolled across 4 U.S. academic medical centers from 2019 through 2020.

  2. Regulatory document, by authors with no tie to the system.

    Center for Medical Device Evaluation (CMDE), National Medical Products Administration, China. 医疗器械产品注册技术审评报告：肠息肉电子结肠内窥镜图像辅助检测软件 (technical review report for the registration of colonoscopy polyp image detection-aid software), acceptance no. CQZ2100679, applicant 成都微识医疗设备有限公司. 21 July 2022. Section 四、产品受益风险判定 (benefit-risk determination).

    根据申请人提供的申报资料，经综合评价，在目前认知水平上，认为该产品的上市为适用人群带来的受益大于风险。

Details

AI system

Reported configuration

Version V1

Workflow

The software took the video signal from the endoscope's image processor and marked suspected polyps on a separate monitor. Each patient had one colonoscopy with the software and one without, in random order. The trial ran at four clinical institutions outside China.

Follow-up

Conditions for replication

Patients are adults having colonoscopy. The colonoscope is one of the specified models, used in white-light mode. Endoscopists judge each polyp from the images and the patient's condition and do not decide on the marks alone.

Entry details

Sector

Healthcare.

Purpose

Chengdu Wision Medical Device built EndoScreener to show licensed endoscopists where suspected polyps are, in real time, during colonoscopy of adults.

29 more questions not yet checked

model, provider, role, configuration, inputs and outputs, autonomy, human oversight, beneficiaries, reach, distribution and access, duration, every measurement question, every money question, ownership and control, dependencies, cross-border connections, geopolitical implications, replication attempts, availability, institutional decision, grounds, action and result, success criterion, every evidence question, observed or intended, alternative assessments, related entries, other identifiers, category mappings.

Cite this record

Peter H. Vartanian. (2026). EndoScreener lowered the adenoma miss rate in China's review (AIBD-1074). Artificial Intelligence Benefits Database, edition of 1 November 2026. https://petervartanian.xyz/aibd/records/AIBD-1074.html (archived at https://doi.org/10.7910/DVN/AADSII, version 7.0)

BibTeX Data Field definitions