[论文解读] Natural language processing of MIMIC-III clinical notes for identifying diagnosis and procedures with neural networks

Siddhartha Nuthakki, Sunil Neela+2 more|arXiv (Cornell University)|Dec 28, 2019

Machine Learning in Healthcare参考文献 29被引用 27

一句话总结

本研究将 ULMFiT 深度学习模型应用于 MIMIC-III 数据集中 120 万条急诊科临床记录，从非结构化文本中预测 ICD-9 诊断与操作编码。该模型在诊断和操作预测任务中分别实现了 80.3% 和 80.5% 的 top-10 准确率，表明其在自动化临床编码和减轻行政负担方面具有优异性能。

ABSTRACT

Coding diagnosis and procedures in medical records is a crucial process in the healthcare industry, which includes the creation of accurate billings, receiving reimbursements from payers, and creating standardized patient care records. In the United States, Billing and Insurance related activities cost around $471 billion in 2012 which constitutes about 25% of all the U.S hospital spending. In this paper, we report the performance of a natural language processing model that can map clinical notes to medical codes, and predict final diagnosis from unstructured entries of history of present illness, symptoms at the time of admission, etc. Previous studies have demonstrated that deep learning models perform better at such mapping when compared to conventional machine learning models. Therefore, we employed state-of-the-art deep learning method, ULMFiT on the largest emergency department clinical notes dataset MIMIC III which has 1.2M clinical notes to select for the top-10 and top-50 diagnosis and procedure codes. Our models were able to predict the top-10 diagnoses and procedures with 80.3% and 80.5% accuracy, whereas the top-50 ICD-9 codes of diagnosis and procedures are predicted with 70.7% and 63.9% accuracy. Prediction of diagnosis and procedures from unstructured clinical notes benefit human coders to save time, eliminate errors and minimize costs. With promising scores from our present model, the next step would be to deploy this on a small-scale real-world scenario and compare it with human coders as the gold standard. We believe that further research of this approach can create highly accurate predictions that can ease the workflow in a clinical setting.

研究动机与目标

• 自动化将非结构化临床记录映射到标准化 ICD-9 诊断与操作编码。
• 降低与医疗编码相关的高昂行政成本，2012 年美国医院支出中该部分达 4710 亿美元。
• 评估深度学习模型（特别是 ULMFiT）在大规模临床记录数据上执行临床编码任务的性能。
• 通过提升临床记录与计费流程的速度、准确性和效率，支持人工编码人员。

提出的方法

• 本研究采用在 MIMIC-III 数据集（包含 120 万条急诊科临床记录）上微调的 ULMFiT 迁移学习框架。
• 以入院时的现病史和症状文本作为输入，预测 ICD-9 诊断与操作编码。
• 通过微调嵌入表示和判别性微调，采用序列分类方法训练模型，以预测 top-10 和 top-50 的 ICD-9 编码。
• 使用 top-1 和 top-50 准确率指标对诊断与操作预测任务的模型性能进行评估。
• 该方法利用迁移学习，使通用语言模型在有限标注数据下适配临床领域。

实验结果

研究问题

• RQ1 像 ULMFiT 这类深度学习模型能否有效将非结构化临床记录映射到 ICD-9 诊断与操作编码？
• RQ2 ULMFiT 在临床编码任务中的表现与传统机器学习模型相比如何？
• RQ3 该模型在诊断与操作分类任务中，对 top-10 和 top-50 ICD-9 编码的预测准确率如何？
• RQ4 当与人工编码人员协同使用时，此类模型能否减少临床编码的时间与错误率？

主要发现

• 该模型在从非结构化临床记录中预测诊断编码时，实现了 80.3% 的 top-10 准确率。
• 该模型在预测操作编码时，实现了 80.5% 的 top-10 准确率，表明其在操作编码任务中表现优异。
• 对于 top-50 预测，诊断编码的准确率为 70.7%，而操作编码的准确率为 63.9%。
• 结果表明，像 ULMFiT 这类深度学习模型可显著协助人工编码人员，减少手动工作量并降低错误率。
• 本研究支持在真实临床环境中部署此类模型，以优化编码工作流程。