Endoscopy
Signal score, how each part rates
ActionabilityPrevalenceStakesEvidence strengthNovelty
See the full rubric →Large language models' performance in answering common patient questions about colonoscopy: an expert-based evaluation.
Clinical takeawayConsider using ChatGPT-4 as a supplementary tool for patient education on colonoscopy, but ensure human oversight to verify accuracy and safety.
What it foundChatGPT-4 outperformed Google Bard in answering colonoscopy-related questions, with median overall index scores ranging from 4.1 to 4.7 vs 3.6 to 4.0, and significantly higher ratings in accuracy, applicability, comprehensiveness, and communication (p < 0.05, effect sizes 0.46-0.64).
ContextThis study provides evidence that ChatGPT-4 may be a more reliable resource for patient education compared to Google Bard, though human review remains critical.
No standard claimed for this paper
Every paper is compared to the standard governing its question. This one is not: either no standard in the corpus matches it, or the comparison did not hold up on review and was withdrawn rather than published unverified. Both are logged.
Güneş Aydemi̇r N … Aydemir AM · BMC Gastroenterology · IF 2.5 · PubMed ↗