Truace
Subscribe
64/100

Moderate evidence

Cite
More
Export CSVExport JSONEmailLinkedInX

Endoscopy ruler cut size errors for early stomach cancer in a 78-patient study

Study title: Measurement of Early Gastric Cancer Size Using a Computer Vision-Assisted Virtual Ruler: A Prospective Clinical Study

Evidence type
Controlled or validated comparison
Sample
n = 78
Sources
1 sourcePeer-reviewed
Published
2 min read
Truvace record
TRV-2026-2473cite or check it

Study title: Measurement of Early Gastric Cancer Size Using a Computer Vision-Assisted Virtual Ruler: A Prospective Clinical Study

A prospective study of 78 patients who had endoscopic submucosal dissection for early gastric cancer in 2024 found that a computer vision-assisted virtual ruler measured lesions more accurately than doctors' visual estimates. Mean absolute error was 1.00 mm with the ruler against 3.00 mm for visual estimation.

Read the study abstract

Background: Accurate measurement of early gastric cancer (EGC) size is important for clinical management. We evaluated a computer vision-assisted virtual ruler (VR) for measuring lesion size during endoscopy. Methods: This prospective clinical trial included 78 patients who underwent endoscopic submucosal dissection (ESD) from January to December 2024. The maximum diameter of the post-ESD specimen, measured with a physical ruler, was used as the reference. VR measurements obtained during ESD and preoperative vis…

The gain

In a prospective study of 78 patients, a computer vision virtual ruler measured early gastric cancer lesions with a mean absolute error of 1.00 mm versus 3.00 mm for visual estimation.

In brief

The reference was the maximum diameter of the removed specimen, measured with a physical ruler. Visual estimation underestimated lesion size overall and in both the high-grade intraepithelial neoplasia and adenocarcinoma groups, while the virtual ruler's readings did not differ significantly from the reference.

Accurate size matters for clinical management of early gastric cancer. The gap was most apparent for flat lesions and adenocarcinoma, though the reported results come from one prospective study with 78 patients.

Main points

  1. The study enrolled 78 patients who underwent endoscopic submucosal dissection between January and December 2024.
  2. Mean absolute error was 1.00 mm with the virtual ruler versus 3.00 mm with visual estimation (P=0.001).
  3. Percentage relative error was 3.28% with the ruler versus 7.89% with visual estimation (P=0.001).
  4. Bland-Altman limits of agreement were -4.23 to 4.19 mm for the ruler versus -3.85 to 9.49 mm for visual estimates.
  5. The intraclass correlation coefficient was 0.979 for the ruler versus 0.909 for visual estimation (P<0.001).
  6. The difference was most apparent for flat lesions and adenocarcinoma.

The full story

Researchers compared two ways of sizing lesions: virtual ruler measurements taken during endoscopic submucosal dissection and preoperative visual estimates. Both were checked against the maximum diameter of the post-resection specimen, measured with a physical ruler.

Agreement statistics favored the virtual ruler. Its Bland-Altman limits of agreement spanned -4.23 to 4.19 mm, while visual estimates spanned -3.85 to 9.49 mm, a much wider upper range that reflects underestimation of lesion size.

The team also compared lesion size between patients with high-grade intraepithelial neoplasia and adenocarcinoma. Visual estimation underestimated size in both groups, and the virtual ruler's advantage was largest for flat lesions and adenocarcinoma.

Talk it overFour questions for a class, a team or the dinner table

  1. Who gets this benefit, and who was left out of the study or story?

    The gain In a prospective study of 78 patients, a computer vision virtual ruler measured early gastric cancer lesions with a mean absolute error of 1.00 mm versus 3.00 mm for visual estimation.

  2. How do we know? What can this kind of evidence show, and what can’t it?

    The evidence This rests on a controlled comparison with a sample of 78, from Surgical Laparoscopy, Endoscopy & Percutaneous Techniques.

  3. What does it leave open? What else would you want to know before acting on it?

  4. Have you seen AI used this way where you study, work or live? Did it go the same way?

The questions are the same on every record. The details under them are taken from this record as written.

Sources

1 on file
  1. Surgical Laparoscopy, Endoscopy & Percutaneous Techniquesdoi:10.1097/sle.0000000000001526
    Peer-reviewed2026-11-11

The debate

More from Health