使用大型语言模型在药物处方中使用零和少量射击命名实体识别和文本扩展
Natthanaphop Isaradech1, Andrea Riedel2, Wachiranun Sirikul3
1Department of Community Medicine, Faculty of Medicine, Chiang Mai University, Thailand; Institute for Medical Informatics, Statistics and Documentation, Medical University of Graz, Austria.
Artificial intelligence in medicine
|July 2, 2025
概括
大型语言模型 (LLM) 可以构建和扩展电子健康记录 (EHR) 的自由文本药物数据. 通过使用ChatGPT3.5进行几次拍摄的方法,显著提高了命名实体识别 (NER) 和文本扩展 (EX) 的准确性.
科学领域:
- 计算语言学 计算语言学
- 医疗信息学 医疗信息学
- 人工智能在医学中的应用
背景情况:
- 电子健康记录 (EHR) 中的药物信息通常是非结构化的自由文本.
- 这种自由文本格式包括多种语言,品牌名称和缩写,阻碍了解释.
- 大型语言模型 (LLM) 提供了处理和结构复杂文本数据的潜力.
研究的目的:
- 评估ChatGPT3.5在自动构建和扩展来自EHR出院摘要的药物陈述中的有效性.
- 通过使用零和少数拍摄提示策略来评估命名实体识别 (NER) 和文本扩展 (EX) 的性能.
- 为了比较ChatGPT3.5与其他先进的药物陈述处理LLM的性能.
主要方法:
- 在自由文本药物陈述上使用了ChatGPT3.5与NER和EX任务的零和少数拍摄提示策略.
- 手动注释和策划100种药物陈述进行评估.
- 通过严格和部分匹配测量NER性能,通过专家评估的语义等价性测量EX性能;计算F1分数.
主要成果:
- 对于NER,表现最好的提示实现了F1平均得分为0.94.
- 对EX的几次射击提示显示出卓越的性能,平均F1得分为0.87.
- 大多数测试的LLM (ChatGPT4o,Gemini 2.0 Flash,MedLM-1.5-Large,DeepSeekV3) 在NER和EX任务中都超过了ChatGPT3.5.
结论:
- 聊天GPT3.5显示了准确的NER和EX自由文本药物陈述的显著潜力.
- 在安全关键的药物数据处理中,采用短暂的学习方法对于预防幻觉至关重要.
- 高级LLM通常为这些临床NLP任务提供比ChatGPT3.5更好的性能.
相关概念视频
Drug Nomenclature
2.3K
During the development of a new pharmaceutical, the manufacturer initially assigns a code name to the drug. Once approved, the drug receives a United States Adopted Name (USAN)—a generic, nonproprietary designation. Upon being listed in the United States Pharmacopeia, this nonproprietary name becomes the drug's official name. Additionally, the manufacturer assigns a proprietary name or trademark, which serves as the brand name under which the drug is marketed. It is worth noting that...
2.3K
Prescription, Nonprescription and Orphan Drugs
857
Prescription drugs require a prescription from a medical practitioner and can only be obtained from a pharmacy. They have many applications, including treating pain, anxiety, and hypertension.
The misuse and addiction to prescription drugs is a growing problem that can affect people of all age groups, specifically teenagers. This can happen when prescription medications are used in ways not intended by the prescriber, such as taking someone else's prescription or using medication for...
The misuse and addiction to prescription drugs is a growing problem that can affect people of all age groups, specifically teenagers. This can happen when prescription medications are used in ways not intended by the prescriber, such as taking someone else's prescription or using medication for...
857
Targets for Drug Action: Overview
7.5K
Drugs target macromolecules to modify ongoing cellular processes. Primary drug targets include receptors, ion channels, transporters, and enzymes.
Receptors are either membrane-spanning or intracellular proteins, which upon binding a ligand, get activated and transmit the signal downstream to elicit a response. Drugs bind receptors, either mimicking the action of endogenous ligands or blocking the receptor activity to bring about a modified response. Nearly 35% of approved drugs target the G...
Receptors are either membrane-spanning or intracellular proteins, which upon binding a ligand, get activated and transmit the signal downstream to elicit a response. Drugs bind receptors, either mimicking the action of endogenous ligands or blocking the receptor activity to bring about a modified response. Nearly 35% of approved drugs target the G...
7.5K
Pharmacokinetic Models: Overview
1.2K
Pharmacokinetic models utilize mathematical analysis to achieve a detailed quantitative understanding of a drug's life cycle within the body. They are instrumental in simulating a drug's pharmacokinetic parameters, predicting drug concentrations over time, optimizing dosage regimens, linking concentrations with pharmacologic activity, and estimating potential toxicity.
There are three primary types of models: empirical, compartment, and physiological. Empirical models, with minimal...
There are three primary types of models: empirical, compartment, and physiological. Empirical models, with minimal...
1.2K
Drug Discovery: Overview
8.8K
Drug discovery is a multifaceted process involving extensive screening, testing, and optimization of lead compounds to identify potential new drugs for therapeutic use. It combines several approaches, including screening large numbers of natural products, chemical modification of known active molecules, identification of new drug targets, and rational design based on biological mechanisms and drug-receptor structure. These approaches are carried out in both academic research laboratories and...
8.8K
Drug Dependence
1.2K
Medications are typically administered to achieve therapeutic effects. Some drugs can modify an individual's mood and perception, frequently resulting in various enjoyable experiences. However, this can result in drug dependency, a condition marked by continuous drug use despite potential negative consequences. Drug dependency primarily falls into two categories: psychological and physical dependence. Psychological dependence occurs when the pleasurable feelings induced by the drug...
1.2K


