ArticleANZ journal of surgery
Evaluation of ChatGPT's and Gemini's Ability to Find Studies About Neurosurgical Training and Trainees' Anatomy Understanding.
Article in ANZ journal of surgery. The graph could read no effect estimate from its abstract, so it casts no vote on the map. Not yet cited in PubMed.
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
0 citing papers in PubMed.
No citing paper in PubMed yet.
Corrections and comments
PubMed lists nothing against this paper. Absence here is not a guarantee, only a check that was made.
Authors and funding
6 authors.
Funding
No grant is acknowledged in the PubMed record.
Abstract
backgroundThere is a lack of research about artificial intelligence's potential to explore neurosurgical literature. We aimed to investigate ChatGPT's and Gemini's ability to identify and outline studies about neurosurgical training and trainees' anatomy understanding.
methodsWe asked ChatGPT 4.0 Turbo and Gemini 2.5 flash (in July 2025), and ChatGPT 5.3 and Gemini 3 flash (in March 2026) to list and summarize five papers: (1) about the use of virtual reality in neurosurgical training, (2) about the role of virtual reality in neurosurgery trainees' anatomy understanding, (3) that compared virtual reality with other neurosurgical training methods. We evaluated how many studies were successfully identified and accurately outlined.
resultsFor ChatGPT 4.0 Turbo, the successful identification and summarization percentages were 100%/60%, 40%/40%, and 60%/40% respectively. For ChatGPT 5.3, the respective percentages were 100%/0%, 20%/0%, and 40%/0%. For Gemini 2.5 flash, they were 80%/60%, 40%/20%, and 0%/0%. For Gemini 3 flash, they were 60%/0%, 20%/0%, and 0%/0%. There was a tendency towards reporting outcomes in favor of virtual reality. Gemini 2.5 flash and 3 flash hallucinated 5/15 and 6/15 papers respectively. Gemini 2.5 flash misattributed first authorship in 8/15 papers.
conclusionsApart from the excellent ChatGPT's ability to simply detect studies about the use of virtual reality in neurosurgical training, ChatGPT and Gemini did not provide reliable responses. Gemini performed worse and frequently hallucinated, while both platforms exhibited bias in favor of virtual reality. Newer versions performed generally worse than older ones. Ongoing development may improve these platforms' role in neurosurgical research.
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.