相关实验视频

Updated: Sep 15, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness

Published on: December 6, 2024

将大型语言模型和文本嵌入模型进行比较,用于自动分类放射学报告中的文本,语义和关键变化

Maximilian Lindholz¹, Alina Burdenski¹, Richard Ruppel¹

¹Department of Radiology, Charité Universitätsmedizin Berlin, Berlin, Germany.

European journal of radiology

|July 17, 2025

概括

大型语言模型 (LLM) 在对放射学报告的变化进行分类方面表现有前途,与专家评估保持一致. 这些AI工具可以通过识别文本,语义或临床可行的报告修改来支持同行评审.

科学领域:

医疗信息学医疗信息学
医疗保健中的人工智能
放射学工作流的优化工作流的优化

背景情况:

放射学报告在工作流程中会发生变化,特别是在住院医生的草稿和主治医生的完成之间.
准确地分类这些变化对于质量保证和临床决策至关重要.

研究的目的:

探索大型语言模型 (LLM) 的有效性,并将技术嵌入到放射学报告中对变化的分类中.
将LLM的绩效与传统指标和人类专家评估进行比较.

主要方法:

评估了400个成年人CT报告,比较住院医生的草稿与主治医生的最终版本.
使用五分尺度评估变化 (没有变化至关键).
开源LLM (例如Qwen3-235B-A22B) 与传统指标 (词差异,Levenshtein,Jaccard) 和文本嵌入相似性的比较.

主要成果:

在人类评价者中,评价者之间的可靠性很好 (κ = 0.990).
Qwen3-235B-A22B LLM实现了与人类评估的最高一致性 (κ = 0.822),超过了传统方法.
与LLM和经典指标相比,嵌入模型的性能明显较低.

结论:

临床医学仪表 (LLM),特别是Qwen3-235B-A22B,显示出对精确评估放射学报告中变化的临床意义的巨大潜力.
在这个任务中,LLM优于嵌入方法和传统的文本比较方法.
这些发现表明,LLM可以作为一种有价值的工具,以加强放射学中的同行评审过程.

关键词:

人工智能的人工智能临床决策 - 临床决策医学教育医学教育自然语言处理自然语言处理.断层扫描 (Tomography) 是一个专业的技术.电脑计算的X射线

更多相关视频

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Machine Learning Algorithms for Early Detection of Bone Metastases in an Experimental Rat Model

Published on: August 16, 2020

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

相关概念视频

Improving Translational Accuracy

Improving Translational Accuracy

Base complementarity between the three base pairs of mRNA codon and the tRNA anticodon is not a failsafe mechanism. Inaccuracies can range from a single mismatch to no correct base pairing at all. The free energy difference between the correct and nearly correct base pairs can be as small as 3 kcal/ mol. With complementarity being the only proofreading step, the estimated error frequency would be one wrong amino acid in every 100 amino acids incorporated. However, error frequencies observed in...

Positron Emission Tomography

Positron Emission Tomography

Positron emission tomography (PET) is a medical imaging technique involving radiopharmaceuticals — substances that emit short-lived radiation. Although the first PET scanner was introduced in 1961, it took 15 more years before radiopharmaceuticals were combined with the technique and revolutionized its potential.
One of the main requirements of a PET scan is a positron-emitting radioisotope, which is produced in a cyclotron and then attached to a substance used by the part of the body...