文献速递重症综合管理与镇静镇痛

大语言模型可提升临床医生对复杂危重症病例的诊断表现

A large language model improves clinicians' diagnostic performance in complex critical illness cases.
PubMed 文献 · Crit Care · 2025-06-06
一句话结论在复杂危重症病例中,DeepSeek-R1辅助使重症住院医师首选诊断准确率从27%升至58%,诊断时间从1920秒降至972秒。

这项前瞻性对照研究从文献中收集48例疑难危重症病例,将32名来自三级教学医院的重症住院医师随机分为无AI辅助组和AI辅助组,每组16人,每人平均处理3例。研究评估推理模型DeepSeek-R1的回答质量,并比较两组诊断准确率和效率。DeepSeek-R1回答的完整性、清晰性和有用性中位Likert评分分别为4.0、5.0和5.0;模型首选诊断准确率为60%(29/48),中位鉴别诊断质量评分为5.0。住院医师首选诊断准确率在无AI辅助组为27%(13/48),AI辅助组为58%(28/48);中位鉴别诊断质量评分无AI辅助为3.0,AI辅助为5.0。AI辅助下住院医师诊断时间中位数为972秒,显著短于无AI辅助的1920秒。作者认为推理模型是有前景的ICU诊断辅助工具。

要点

  • 在48例疑难危重症病例中,AI辅助组住院医师首选诊断准确率为58%(28/48),无AI辅助组为27%(13/48)。
  • AI辅助使住院医师诊断时间中位数从1920秒(IQR 1320-2640)降至972秒(IQR 570-1320)。
  • DeepSeek-R1模型自身首选诊断准确率为60%(29/48;95%CI 0.456-0.729),中位鉴别诊断质量评分为5.0。
  • AI辅助组中位鉴别诊断质量评分为5.0(IQR 3.0-5.0),无AI辅助组为3.0(IQR 0-5.0)。
  • DeepSeek-R1回答的完整性、清晰性、有用性中位Likert评分分别为4.0、5.0、5.0。

适用人群与场景

  • 从文献中收集的疑难危重症病例,共48例。
  • 来自三级教学医院的重症住院医师,共32名,随机分为无AI辅助组和AI辅助组,每组16名,每人平均处理3例。

研究设计与关键结果

  • 研究类型为前瞻性对照研究,纳入48例疑难危重症病例和32名重症住院医师,随机分为无AI辅助组(16人)和AI辅助组(16人),每人平均处理3例。
  • DeepSeek-R1回答的完整性中位Likert评分为4.0(IQR 4.0-5.0;95%CI 4.0-4.5),清晰性为5.0(IQR 4.0-5.0;95%CI 4.5-5.0),有用性为5.0(IQR 4.0-5.0;95%CI 4.0-5.0)。
  • DeepSeek-R1模型首选诊断准确率为60%(29/48;95%CI 0.456-0.729),中位鉴别诊断质量评分为5.0(IQR 4.0-5.0;95%CI 4.5-5.0)。
  • 住院医师首选诊断准确率:无AI辅助组为27%(13/48;95%CI 0.146-0.396),AI辅助组为58%(28/48;95%CI 0.438-0.729)。
  • 住院医师中位鉴别诊断质量评分:无AI辅助组为3.0(IQR 0-5.0;95%CI 2.0-4.0),AI辅助组为5.0(IQR 3.0-5.0;95%CI 3.0-5.0)。
  • 住院医师诊断时间中位数:无AI辅助组为1920秒(IQR 1320-2640;95%CI 1710-2370),AI辅助组为972秒(IQR 570-1320;95%CI 675-1200),AI辅助组显著缩短。

证据类型、等级与局限

  • 证据类型:前瞻性对照研究。
  • 局限:原文未报告局限性。

对床旁实践的改变

  • 对于诊断困难的危重症病例,作者认为推理模型是有前景的ICU诊断辅助工具。

提炼说明

  • 本条依据论文摘要整理;摘要未涉及「诊断标准与关键阈值」「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开

BACKGROUND: Large language models (LLMs) have demonstrated potential in assisting clinical decision-making. However, studies evaluating LLMs' diagnostic performance on complex critical illness cases are lacking. We aimed to assess the diagnostic accuracy and response quality of an artificial intelligence (AI) model, and evaluate its potential benefits in assisting critical care residents with differential diagnosis of complex cases.

METHODS: This prospective comparative study collected challenging critical illness cases from the literature. Critical care residents from tertiary teaching hospitals were recruited and randomly assigned to non-AI-assisted physician and AI-assisted physician groups. We selected a reasoning model, DeepSeek-R1, for our study. We evaluated the model's response quality using Likert scales, and we compared the diagnostic accuracy and efficiency between groups.

RESULTS: A total of 48 cases were included. Thirty-two critical care residents were recruited, with 16 residents assigned to each group. Each resident handled an average of 3 cases. DeepSeek-R1's responses received median Likert grades of 4.0 (IQR 4.0-5.0; 95% CI 4.0-4.5) for completeness, 5.0 (IQR 4.0-5.0; 95% CI 4.5-5.0) for clarity, and 5.0 (IQR 4.0-5.0; 95% CI 4.0-5.0) for usefulness. The AI model's top diagnosis accuracy was 60% (29/48; 95% CI 0.456-0.729), with a median differential diagnosis quality score of 5.0 (IQR 4.0-5.0; 95% CI 4.5-5.0). Top diagnosis accuracy was 27% (13/48; 95% CI 0.146-0.396) in the non-AI-assisted physician group versus 58% (28/48; 95% CI 0.438-0.729) in the AI-assisted physician group. Median differential quality scores were 3.0 (IQR 0-5.0; 95% CI 2.0-4.0) without and 5.0 (IQR 3.0-5.0; 95% CI 3.0-5.0) with AI assistance. The AI model showed higher diagnostic accuracy than residents, and AI assistance significantly improved residents' accuracy. The residents' diagnostic time significantly decreased with AI assistance (median, 972 s; IQR 570-1320; 95% CI 675-1200) versus without (median, 1920 s; IQR 1320-2640; 95% CI 1710-2370).

CONCLUSIONS: For diagnostically difficult critical illness cases, DeepSeek-R1 generates high-quality information, achieves reasonable diagnostic accuracy, and significantly improves residents' diagnostic accuracy and efficiency. Reasoning models are suggested to be promising diagnostic adjuncts in intensive care units.

来源:Crit Care,查看原文。版权归原作者所有。

参考来源

  1. Wu X, Huang Y, He Q, et al. A large language model improves clinicians' diagnostic performance in complex critical illness cases. Crit Care. 2025. PMID: 40481529. doi:10.1186/s13054-025-05468-7

供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。