MétaCan
Menu
Back to cohort
Record W4410764776 · doi:10.2196/66429

Large Language Models as a Consulting Hotline for Patients With Breast Cancer and Specialists in China: Cross-Sectional Questionnaire Study

2025· article· en· W4410764776 on OpenAlexvenueno aff
Hui Liu, Jialun Peng, Ao Deng, Xiangxin Huang, Guobing Yin, Haojun Luo

Bibliographic record

VenueJMIR Medical Informatics · 2025
Typearticle
Languageen
FieldMedicine
TopicArtificial Intelligence in Healthcare and Education
Canadian institutionsnot available
Fundersnot available
KeywordsBreast cancerMedicineFamily medicineCancerHealth literacyLikert scalePsychologyHealth careInternal medicine

Abstract

fetched live from OpenAlex

Background: The disease burden of breast cancer is increasing in China. Guiding people to obtain accurate information on breast cancer and improving the public's health literacy are crucial for the early detection and timely treatment of breast cancer. Large language model (LLM) is a currently popular source of health information. However, the accuracy and practicality of the breast cancer-related information provided by LLMs have not yet been evaluated. Objective: This study aims to evaluate and compare the accuracy, practicality, and generalization-specificity of responses to breast cancer-related questions from two LLMs, ChatGPT and ERNIE Bot (EB). Methods: The questions asked to the LLMs consisted of a patient questionnaire and an expert questionnaire, each containing 15 questions. ChatGPT was queried in both Chinese and English, recorded as ChatGPT-Chinese (ChatGPT-C) and ChatGPT-English (ChatGPT-E) respectively, while EB was queried in Chinese. The accuracy, practicality, and generalization-specificity of each inquiry's responses were rated by a breast cancer multidisciplinary treatment team using Likert scales. Results: Overall, for both the patient and expert questionnaire, the accuracy and practicality of responses from ChatGPT-E were significantly higher than those from ChatGPT-C and EB (all Ps<.001). However, the responses from all LLMs are relatively generalized, leading to lower accuracy and practicality for the expert questionnaire compared to the patient questionnaire. Additionally, there were issues such as the lack of supporting evidence and potential ethical risks in the responses of LLMs. Conclusions: Currently, compared to other LLMs, ChatGPT-E has demonstrated greater potential for application in educating Chinese patients with breast cancer, and may serve as an effective tool for them to obtain health information. However, for breast cancer specialists, these LLMs are not yet suitable for assisting in clinical diagnosis or treatment activities. Additionally, data security, ethical, and legal risks associated with using LLMs in clinical practice cannot be ignored. In the future, further research is needed to determine the true efficacy of LLMs in clinical scenarios related to breast cancer in China.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.003
metaresearch head score (Gemma)0.006
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.035
Threshold uncertainty score0.069

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0030.006
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0010.002
Science and technology studies0.0010.001
Scholarly communication0.0010.001
Open science0.0010.001
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0020.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.033
GPT teacher head0.433
Teacher spread0.400 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations3
Published2025
Admission routes1
Has abstractyes

Explore more

Same venueJMIR Medical InformaticsSame topicArtificial Intelligence in Healthcare and EducationFrench-language works237,207