文献速递牙周病口腔正畸口腔修复与种植

多专科牙科电子病历:基于椅旁音频的探索性研究

Multispecialty Dental EMRs from Chairside Audio: An Exploratory Study.
PubMed 文献 · J Dent Res · 2026-07-24

本研究开发并内部评估了一种模块化大语言模型系统,用于在声学干扰和专科异质性条件下从牙科椅旁咨询生成标准化电子病历。研究比较了基线端到端系统与增加证据增强模块的系统,使用100份来自牙周、正畸和修复科的脱敏门诊录音。结果显示系统2提高了鲁棒性,标准差降低33.8%至55.9%,且在复杂修复病例中增益最大。

要点

  • 证据增强与共识约束的LLM系统可提高电子病历生成的准确性和一致性。
  • 系统2在牙周、正畸和修复科的标准差分别降低33.8%、55.9%和42.6%。
  • AI评分与人类专家评分相关(r=0.823;R2=0.678),存在稳定正偏倚(平均差12.54分)。

核心结论

  • 基于证据和共识约束的LLM系统在从真实椅旁音频生成电子病历方面具有方法学可行性,提高了医学准确性和一致性,并减少了复杂临床工作流程中的变异性。

研究设计与样本

  • 研究开发了模块化LLM系统,比较基线端到端系统(系统1)与增加模块A(多源证据捕获与事实表构建)和模块B(多版本协作与共识投票;N=3,共识≥2/3)的证据增强系统(系统2)。
  • 评估了100份来自牙周科、正畸科和修复科的脱敏门诊录音,采用双盲临床医生评分和自动评分系统评估完整性、语言质量、医学准确性和结构标准化(每维度0-100分)。

主要发现

  • 系统1在所有模型中达到临床可接受性能,Kimi-K2表现出最高稳定性。
  • 系统2提高了鲁棒性,标准差降低:牙周科33.8%,正畸科55.9%,修复科42.6%,在复杂修复病例中增益最大。
  • 数据集特征和声学模式分析证实了语料的临床真实性,并确立了其用于性能评估的适用性。
  • AI生成评分与人类专家评分相关(r=0.823;R2=0.678),具有稳定的正偏倚(平均差12.54分)。

关键数值与推荐

  • 系统2的共识投票条件为N=3,共识≥2/3。
  • 输出质量评估维度包括完整性、语言质量、医学准确性和结构标准化,每维度评分0-100分。
  • AI评分与人类专家评分的相关系数r=0.823,决定系数R2=0.678,平均差异12.54分。

临床怎么用

  • 该系统可用于从牙科椅旁音频自动生成标准化电子病历,尤其在复杂修复病例中可能减少变异性。
  • 临床应用中需注意AI评分与人类专家评分存在稳定正偏倚(平均差12.54分),可能需要校准。

局限性

  • 本研究为探索性研究,样本量有限(100份录音),仅涉及牙周、正畸和修复三个专科。
  • 系统性能在声学干扰和专科异质性条件下评估,但未报告其他临床环境下的验证结果。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

The objective of this study was to develop and internally evaluate a modular large language model (LLM) system for generating standardized electronic medical records (EMRs) from dental chairside consultations under conditions of acoustic interference and specialty-specific heterogeneity. We built a controllable pipeline integrating multistage audio enhancement and local automatic speech recognition with a cascaded LLM generator. A baseline end-to-end system (system 1) was compared with an evidence-enhanced system (system 2) that added module A (multisource evidence capture and fact table construction) and module B (multiversion collaboration and consensus voting; N = 3, consensus ≥2/3). We evaluated 100 deidentified outpatient recordings from periodontics, orthodontics, and prosthodontics. Output quality was assessed by double-blind clinician scoring and an automated scoring system across completeness, language quality, medical accuracy, and structural standardization (0 to 100 per dimension). System 1 achieved clinically acceptable performance across models, with Kimi-K2 showing the highest stability. System 2 improved robustness with standard deviation reductions of 33.8% (periodontics), 55.9% (orthodontics), and 42.6% (prosthodontics), with the largest gains in complex prosthodontic cases. Analysis of dataset characteristics and acoustic patterns confirmed the clinical authenticity of the corpus and established its suitability for performance evaluation. Artificial intelligence-generated scores correlated with human expert ratings (r = 0.823; R2 = 0.678) with a stable positive bias (mean difference 12.54 points). Our evidence-grounded and consensus-constrained LLM system demonstrated methodological feasibility for generating EMRs from authentic chairside audio, with improvements in medical accuracy and consistency and reduced variability in complex clinical workflows.

来源:J Dent Res,查看原文。版权归原作者所有。

引用

  1. Wei L, Wang Y, Lu J, et al. Multispecialty Dental EMRs from Chairside Audio: An Exploratory Study. J Dent Res. 2026. PMID: 42499230. doi:10.1177/00220345261464508

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。