ArticleAnnals of coloproctology2025
How appropriately can generative artificial intelligence platforms, including GPT-4, Gemini, Bing, and Wrtn, answer questions about colon cancer in the Korean language?
Article in Annals of coloproctology, 2025. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Cited by 2 papers.
What it found
Each row is one number read from the abstract, on the scale the paper reported it, with its interval. Left of the dashed line favours the treatment, right favours the comparator. Under each row is the sentence it came from. New to these charts? A ten-minute tutorial.
The abstract states no effect estimate the extractor could read, or names no intervention and outcome on the map, so this paper lights no cell and moves no belief. It is still indexed, cited and linked below.
The trial behind it
Trials whose registry record cites this paper, or whose number appears in the abstract. A trial that started after this paper was published is citing it as background, not reporting it.
Neither the registry nor the abstract names a trial number. If this is a trial report, that itself is worth knowing.
Who cites it
2 citing papers in PubMed.
- Temporal evolution of large language models (LLMs) in oncology.Journal of translational medicine · 2025Article
- Role of Medical Editors in the Age of Generative Artificial Intelligence.Healthcare informatics research · 2025Article
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
1 author.
Funding
Abstract
purposeThis study aims to assess the performance of 4 generative artificial intelligence (AI) platforms-Gemini (formerly Bard), Bing, GPT-4, and Wrtn-in answering questions about colon cancer in the Korean language. Two main research questions guided this study. First, which AI platform provides the most accurate answers? Second, can these AI-generated answers be reliably used to educate patients and their families about colon cancer?
methodsTen questions selected by the author were posed to the 4 generative AI platforms on February 22, 2024. Two colorectal surgeons in Korea, each with over 20 years of clinical experience, independently evaluated the answers provided by these generative AI platforms.
resultsThe generative AI platforms scored an average of 5.5 out of 10 points. Wrtn achieved the highest score at 6 points, followed by GPT-4 and Gemini, each with 5.5, and Bing, scoring 5 points. The weighted κ for inter-rater reliability was 0.597 (P<0.001). The generative AI platforms performed well in explaining the occult blood test for cancer screening, keyhole surgery, and dietary recommendations for cancer prevention. However, they demonstrated significant limitations in answering more complex topics, such as estimating survival rates following surgery, choosing targeted therapy after surgery, and accurately reporting the mortality rate due to colon cancer in Korea.
conclusionThe findings suggest that using these generative AI platforms as educational resources for patients and their families regarding colon cancer is premature. Further training on colorectal diseases is required before these AI platforms can be considered reliable information sources for the general public in Korea.
Indexed as
Identifiers
What OpenQuestion holds
Registered trials
Read under generation 80e0d062 · epoch 390. Bibliography from PubMed, PubMed Central and OpenAlex; grants from NIH RePORTER; trial links from ClinicalTrials.gov; estimates, votes and beliefs from the OpenQuestion graph.