Artificial intelligence in upper gastrointestinal endoscopy: what detection rates cannot tell us

Joon Sung Kim · 2026

Read Artificial intelligence in upper gastrointestinal endoscopy: what detection rates cannot tell us by Joon Sung Kim (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13624766.txt.

Description

Joon Sung Kim writes a 2026 letter on AI help during stomach endoscopy. The letter comments on a study that found real time AI raised lesion detection, above all for very small spots. Kim praises the real world evidence from a low resource setting, then raises five cautions. Two headline rates rested on biopsy results, which carry their own sampling error and disagreement between pathologists.

Cancer counts were small, so a flat cancer rate may reflect chance rather than no effect. Fewer biopsies after AI could mean sharper choices, or missed subtle lesions. The single center, single doctor design, comparing an earlier and a later period, cannot fully separate AI's effect from growing experience. Background stomach risk also went unreported. Kim closes by backing the study's value while asking for stronger designs in future work.

Catalog record

CREATORJoon Sung Kim
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE2 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13624766

Browse the full library · About the Tower of Babel