用自然语言处理技术和在重症监护室环境中的开放数据来预测扩展频谱β-乳糖酶产生细菌感染的预测模型:回顾性观察性研究
Genta Ito1, Shuntaro Yada1, Shoko Wakamiya1
1Department of Information Science, Nara Institute of Science and Technology, Ikoma City, Japan.
JMIR formative research
|July 10, 2024
概括
机器学习模型可以使用电子健康记录数据预测产生扩展光谱β-乳糖酶 (ESBL) 的细菌感染. 整合文本信息显著提高了预测准确性,证明了开放数据和NLP工具的价值.
科学领域:
- 医疗信息学 医疗信息学
- 计算生物学 计算生物学
- 医疗保健中的机器学习
背景情况:
- 机器学习模型越来越多地用于医疗事件预测,通常依赖私人数据集.
- 公共医疗信息中心重症监护III (MIMIC-3) 数据集为开发使用结构化和非结构化临床数据的强有力的预测模型提供了宝贵的资源.
研究的目的:
- 开发和评估一种机器学习模型,用于预测扩展光谱β-乳糖酶 (ESBL) 产生的细菌感染.
- 用命名实体识别 (NER) 来评估从电子病历 (EMR) 文本中提取的信息的有效性,以提高预测准确性.
- 证明开放数据源和可访问技术对临床上有意义的预测的有用性.
主要方法:
- 使用了MIMIC-3数据集,包括人口统计,生命体征,实验室结果和出院摘要.
- 专注于患有Klebsiella肺炎或大肠杆菌感染的患者,使用ESBL产生细菌标准进行预测.
- 我们比较了两个模型:L1-规则化的后勤回归 (仅结构化数据) 和LightGBM (结构化和文本数据),通过ROC-AUC和PR-AUC进行评估.
主要成果:
- 采用结构化数据和文本数据的LightGBM模型实现了ROC-AUC0.707和PR-AUC0.369,超过了仅使用结构化数据的模型 (ROC-AUC0.646,PR-AUC0.307).
- 关键预测因素包括患者年龄,入院时间和病史 (例如糖尿病).
- 通过包括文本病史来提高模型的性能,尽管注意到QuickUMLS的潜在不准确性.
结论:
- 通过使用MIMIC-3数据集,成功开发了ESBL产生细菌感染的预测模型.
- 该模型的透明度和对开放数据和NER技术的依赖是显著的优势.
- 未来的优化是有希望的先进的NLP工具,如BERT和GPT,用于从文本中提取增强的医疗数据.
相关概念视频
Acute Pyelonephritis II: Diagnostic Studies and Management
Introduction:For diagnosing acute pyelonephritis, a comprehensive patient history is collected to identify symptoms such as dysuria, frequent or urgent urination, flank pain, or costovertebral angle (CVA) tenderness that may suggest a kidney infection.Physical ExaminationDuring the physical examination, CVA tenderness is assessed. This involves gentle percussion over the costovertebral angle, where tenderness often indicates a kidney infection.Diagnostic TestsUrinalysis: Used to identify white...
Mechanism of Antibiotic Resistance in MRSA
Antibiotic resistance in bacteria arises when microorganisms evolve the ability to withstand drugs designed to kill them or inhibit their growth, rendering once-effective treatments useless. This phenomenon, driven by genetic change and selection under antibiotic exposure, poses a profound threat to modern medicine. Mechanisms include drug-inactivating enzymes (e.g., β-lactamases), efflux pumps that eject antibiotics, mutations altering antibiotic targets, decreased drug uptake, and acquisition...
Clinical Significance of Antibiotic Resistance
Methicillin-resistant Staphylococcus aureus (MRSA) presents a critical public health threat, arising from its capacity to resist β-lactam antibiotics due to acquisition of the mecA gene within the staphylococcal cassette chromosome mec (SCCmec). This gene encodes penicillin-binding protein 2a (PBP2a), which impairs binding efficacy of methicillin and other β-lactams. MRSA has evolved into distinct clonal lineages impacting humans and animals alike, reinforcing its significance within the One...


