Truace
Subscribe

AI-assisted endoscopy in gastric cancer screening workflow

Source article: Impact of AI-Assisted Endoscopy on Secondary Review and Referral Decisions in Gastric Cancer Screening: A Multicenter Real-World Study

Objectives Artificial intelligence (AI)-assisted endoscopy may support gastric cancer screening, but its impact on structured secondary human review remains unclear. We evaluated secondary review modifications and referral decisions before and after AI implementation. Methods This retrospective multicenter study included 5013 screening esophagogastroduodenoscopy examinations performed before (n = 2021) and after (n = 2992) AI implementation. The primary outcome was secondary review modifications. Secondary outco…

By the Truvace desk2 min read1 sourcePeer-reviewedTRV-2026-1986· cite & verify
Impact of AI-Assisted Endoscopy on Secondary Review and Referral Decisions in Gastric Cancer Screening: A Multicenter Real-World Study
"Endoscopy" by Yuya Tamai is licensed under CC BY 2.0. To view a copy of this license, visit https://creativecommons.org/licenses/by/2.0/.
Trace impact readingContested
P 67The P score combines the specificity and measured human impact of the grounded problem claim with the strength of this Trace’s cited sources.

Both sides are scored from claims and sources, not community votes.

G 68The G score combines the specificity and measured human impact of the grounded gain claim with the strength of this Trace’s cited sources.

In brief

Researchers compared 5013 screening esophagogastroduodenoscopy examinations from multiple centers, 2021 done before AI was introduced and 2992 after. They looked at how often a structured second human review changed the initial reading, and at whether patients were referred for further diagnostic evaluation.

The findings matter because AI is often assumed to reduce the need for human correction, yet changes by second reviewers became more frequent and persisted, especially among less experienced endoscopists. It remains unknown whether those changes were correct or whether lower referral was appropriate, since no external reference standard existed.

Main points

  1. The study was retrospective and multicenter, covering 5013 screening endoscopies: 2021 before AI and 2992 after.
  2. The primary outcome was secondary review modifications; referral for further diagnostic evaluation was a secondary outcome.
  3. Secondary review modifications rose from 1.2% to 3.7% after AI implementation.
  4. Analyses adjusted for age, sex, Helicobacter pylori eradication status, prior-year endoscopy, and endoscopist experience, and included inverse probability of treatment weighting.
  5. AI use was associated with lower referral rates, while modifications persisted mainly among nonexpert endoscopists.

The gain

AI-assisted endoscopy was associated with lower referral rates for further diagnostic evaluation in gastric cancer screening.

The problem

Secondary review modifications rose from 1.2% before AI to 3.7% after AI and persisted, particularly among nonexpert endoscopists, with no reference standard to verify them.

The rundown

The design compared a period before AI implementation with a period after it. Multivariable logistic regression adjusted for patient factors such as age, sex, Helicobacter pylori eradication status and prior-year endoscopy, plus endoscopist experience.

Additional interaction analyses by endoscopist experience and inverse probability of treatment weighting were used to test the robustness of results. The interaction work is where the finding about nonexpert endoscopists emerges.

The authors conclude that the results describe workflow changes. They do not establish that the secondary review modifications were clinically valid or that referral decisions were appropriate.

What this doesn’t fix

No external reference standard was available to verify secondary-review modifications or referral decisions, so the clinical validity of the changes and the appropriateness of referrals are not established.

Sources

  1. Peer-reviewedJournal of Gastroenterology and Hepatology2026-10-15

The debate