A reproducible audit protocol shows that high confidence scores and class-activation heatmaps in colonoscopy classifiers often disagree about lesion location, and neither alone proves localisation.
A prospective multicenter randomized controlled trial on artificial intelligence assisted colonoscopy for enhanced polyp detection.Scientific Reports, 14(1):25453, 2024
1 Pith paper cite this work, alongside 16 external citations. Polarity classification is still indexing.
1
Pith paper citing it
16
external citations · OpenAlex
fields
eess.IV 1years
2026 1verdicts
ACCEPT 1representative citing papers
citing papers explorer
-
endoExplain: A reproducible protocol for auditing score-localisation discordance in colonoscopy image classifiers
A reproducible audit protocol shows that high confidence scores and class-activation heatmaps in colonoscopy classifiers often disagree about lesion location, and neither alone proves localisation.