跳到正文
原文
Nature 健康护理主题·· 2 天前AI 评分35

可解释机器学习框架自动筛查患者端大语言模型咨询:医学美容咨询的比较性评估

An interpretable machine learning framework for automated screening of patient-facing LLM consultations: a comparative evaluation in medical aesthetic consultation

AI 导读

针对医学美容咨询场景,本研究开发并验证了基于生成元数据和可读性特征的可解释机器学习框架,用于筛选低质量LLM回复。DeepSeek-V3综合质量评分显著高于Kimi(3.89±0.45 vs 3.70±0.38,P<0.001),提示策略是决定质量的主导因素(Kendall's W=0.376)。机器学习模型对低质量的平均精度达0.802,但需标记83.2%的回复以确保检出90%的低质量内容。

正文

Scientific Reports (2026) Cite this article

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

The integration of large language models (LLMs) into patient-facing clinical decision support systems (CDSS) has outpaced the development of automated quality assurance mechanisms. In medical aesthetic consultation—where patient safety and informed decision-making are paramount—no validated framework exists for pre-deployment screening of LLM-generated content. To develop and evaluate an interpretable machine-learning framework that screens LLM-generated patient-facing consultations using automatically computable generation metadata and readability features—as a triage aid for prioritising expert review—and to comparatively evaluate DeepSeek-V3 and Kimi in medical aesthetic consultation. We generated 600 responses from 25 standardized queries across five aesthetic topics using DeepSeek-V3 (DeepSeek-AI, Hangzhou, China) and Kimi (Moonshot AI, Beijing, China) under six prompt strategies and two conversational modes. Two blinded dermatologists rated quality on six dimensions (accuracy, relevance, comprehensiveness, empathy, falsehood, safety). Readability was quantified using Flesch Reading Ease, Flesch-Kincaid Grade Level, Gunning Fog Index, and SMOG Index. Four machine learning models were trained to predict high-quality responses (composite score ≥ 4.0, pre-specified threshold) and evaluated with question-level grouped 5-fold cross-validation—in which all 24 responses of each seed question were confined to a single fold to prevent information leakage—with SHAP analysis to identify key predictive features. Between-model quality comparisons used paired tests reflecting the matched design. To align evaluation with the triage purpose, performance was additionally assessed with low-quality responses as the positive class (precision–recall analysis and review-burden operating points), and robustness was examined using alternative quality thresholds and an applicability-adjusted composite score. DeepSeek-V3 achieved higher composite quality than Kimi (3.89 ± 0.45 vs. 3.70 ± 0.38, paired Wilcoxon P < 0.001; robust under applicability-adjusted scoring). Prompt strategy was the dominant quality determinant (Friedman test, Kendall’s W = 0.376), with sixth-grade reading level prompts paradoxically yielding the lowest scores (3.25 ± 0.40). All ML models achieved moderate discrimination under question-level grouped cross-validation (AUC 0.714–0.732; response-level sensitivity analysis: 0.758–0.773), with findings robust to alternative quality thresholds (AUC 0.749–0.797 at cutoff 3.5 and 0.725–0.838 at cutoff 4.5) and to applicability-adjusted labelling (AUC 0.679–0.702); prompt strategy, sentence length, and model identity were the strongest SHAP-identified predictors. Aligned with the triage purpose, average precision for low-quality detection reached 0.802 (Logistic Regression), and capturing 90% of low-quality responses required flagging 83.2% of responses for expert review. The proposed framework demonstrates feasibility as an explainable triage aid that flags potentially low-quality LLM consultations for expert review using generation metadata and readability features; it does not assess semantic or clinical content, and expert review remains indispensable. Findings underscore that prompt engineering—not model selection alone—drives output quality, and that readability optimization must be decoupled from clinical completeness to ensure safe deployment in aesthetic medicine information systems.

Acknowledgements

The authors thank the two independent board-certified dermatologists who participated in the blinded quality assessment of LLM responses.

Funding

This work was supported by the Shandong Key Project of Digital Intelligent Medical Science and Technology Innovation Elite Program (Grant No. sdkldto20260006).

Author information

Author notes

  1. Shucheng Zhang, Xiaoyue Sun and Shunyi Lei contributed equally to this work.

Authors and Affiliations

  1. Department of Dermatology, The First Affiliated Hospital of Shandong First Medical University, 16766 Jingshi Road, Lixia District, Jinan, 250013, Shandong, China

    Shucheng Zhang, Xiaoyue Sun, Jiale Geng & Xiaoqing Si

  2. Department of Orthopedics, The First Affiliated Hospital of Harbin Medical University, Harbin, 150001, China

    Shunyi Lei

  3. Department of Obstetrics and Gynecology, Qilu Hospital of Shandong University, Jinan, 250012, China

    Zixian Shao

Authors

  1. Shucheng Zhang
  2. Xiaoyue Sun
  3. Shunyi Lei
  4. Jiale Geng
  5. Zixian Shao
  6. Xiaoqing Si

Corresponding author

Correspondence to Xiaoqing Si.

Ethics declarations

Competing interests

The authors declare no competing interests.

Ethics approval

This study did not require institutional ethical review board approval as it did not involve patient participants or animal subjects. Informed consent was obtained from the two independent board-certified dermatologists who participated in the blinded quality assessment for the use of their anonymized evaluation data in this study and publication.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhang, S., Sun, X., Lei, S. et al. An interpretable machine learning framework for automated screening of patient-facing LLM consultations: a comparative evaluation in medical aesthetic consultation. Sci Rep (2026). https://doi.org/10.1038/s41598-026-80689-6

Download citation

  • Received: 13 July 2026

  • Accepted: 10 November 2026

  • Published: 18 November 2026

  • DOI: https://doi.org/10.1038/s41598-026-80689-6

Keywords

来源:Nature 健康护理主题 · nature.com