利用深度学习与少样本学习改进全髋关节置换术手术记录的数据提取
该试点研究评估GPT-4在少样本学习提示下从全髋关节置换术(THA)手术记录中提取固定方式、技术使用和手术入路的可行性。测试集包含来自38位术者的240份独特记录,分类准确率分别为100%、98.9%和97.5%,并提供临床依据。模型引用原文细节的能力对广泛采用前的验证至关重要,有望替代人工病历审查。
要点
- GPT-4少样本学习分类THA固定方式、技术、入路准确率分别为100%、98.9%、97.5%。
- 模型为分类提供依据,可读性评分较高,与原文有直接引用匹配。
- 模型引用原文细节的能力对广泛采用前的验证至关重要。
核心结论
- 应用GPT-4配合自定义少样本提示对THA手术记录进行固定方式、技术和入路捕捉表现出优异性能。
- 模型引用原文细节的能力对广泛采用前的模型验证至关重要,有望成为人工病历审查的替代方案。
研究设计与样本
- 试点研究,使用GPT-4少样本学习提示,训练集包括固定方式4例、技术11例、手术入路13例金标准手术记录示例。
- 测试集为来自单中心38位术者2011年11月至2024年3月期间实施的初次THA的240份独特记录(固定方式60份、技术90份、入路120份)。
- GPT-4输出与人工病历审查比较准确性,并使用Flesch-Kincaid年级水平评分、self-BLEU评分和字符级序列匹配评估临床依据质量。
主要发现
- GPT-4分类固定方式、技术和入路的总体准确率分别为100%、98.9%和97.5%。
- 模型为分类提供依据的平均Flesch-Kincaid年级水平评分分别为固定方式17.9、技术16.2、入路24.4。
- 依据的平均self-BLEU评分均<0.1。
- 依据与原始记录的字符级序列匹配分别为固定方式87.6%、技术89.2%、入路96.5%,并直接引用原文。
临床怎么用
- 原文未据此给出推荐。
局限性
- 原文未报告局限性。
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
BACKGROUND: Annotating free-text clinical notes into structured data are critical for future large-scale data analysis in institutional and national orthopaedic registries. In total hip arthroplasty (THA), classifying implant fixation, use of technology, and especially surgical approach are particularly difficult for classical machine-learning techniques. In this pilot, we evaluated the feasibility of GPT-4 to capture and justify these common elements in THA operative notes using a custom few-shot learning prompt. METHODS: The GPT-4 was trained with a few-shot learning approach using plain language descriptions of various fixations, technologies, and approaches, along with examples from gold-standard operative notes-four for fixation, 11 for technology, and 13 for surgical approach. The test set comprised 240 unique notes (60 for fixation, 90 for technology, and 120 for approach) from primary THAs performed by 38 surgeons at a single institution (November 2011 to March 2024). The GPT-4's output was compared against manual chart reviews for accuracy. The quality of clinical justifications was assessed using Flesch-Kincaid Grade Level scores for readability, self-BLEU scores for logical diversity, and character-level sequence matches with original notes. RESULTS: The GPT-4 classified fixation, technology, and approach with an overall accuracy of 100, 98.9, and 97.5%, respectively. The model also provided justifications for classifications with average Flesch-Kincaid Grade Level scores of 17.9, 16.2, and 24.4 for fixation, technology, and approach, respectively, and average self-BLEU scores of < 0.1 each. Justifications had character-level sequence matches of 87.6, 89.2, and 96.5%, respectively, with direct note citations for fixation, technology, and approach. CONCLUSIONS: Applying GPT-4 with a custom few-shot prompt to THA operative notes demonstrated excellent performance in capturing fixation, technology, and approach methods. Moreover, the model's ability to cite details from the original notes is critical for model validation before widespread adoption, exhibiting a promising alternative to manual chart review for clinical data capture.
引用
- Attal K, Charalambous L, Di Gangi C, et al. Using Deep Learning With Few-Shot Learning to Improve Data Capture in Total Hip Arthroplasty Operative Notes. J Arthroplasty. 2025. PMID: 40484056. doi:10.1016/j.arth.2025.06.008
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。