使用真实世界患者数据对人工智能平台和药物相互作用查数据库进行比较评估
Bálint Márk Domián1, Amir Reza Ashraf1, András Tamás Fittler1
1Department of Pharmaceutics, Faculty of Pharmacy, University of Pécs, Pécs, Hungary.
Exploratory research in clinical and social pharmacy
|December 22, 2025
概括
大型语言模型 (LLM) 显示了药物相互作用 (DDI) 查的潜力,但当前的AI工具缺乏可靠临床使用的精度和灵敏度. 专业验证LLM输出对于在多药店管理中的患者安全至关重要.
科学领域:
- 药理学 药理学是指药理学的学科.
- 人工智能的人工智能
- 临床药房 临床药房
背景情况:
- 多药制品增加了有害药物相互作用 (DDI) 的风险.
- 传统的DDI查工具通常覆盖范围有限,并产生过度的警报,导致警报疲劳.
- 大型语言模型 (LLM) 为DDI识别提供了一种新的方法,但它们在现实世界的临床实用性需要进一步研究.
研究的目的:
- 为了比较传统的DDI数据库与基于LLM的查的性能,使用真实世界的患者数据.
- 评估ChatGPT,谷歌Gemini和微软Copilot在识别临床相关DDI时的灵敏度,特异性,精度和F1分数.
主要方法:
- 一项探索性研究利用了来自风湿病患者的匿名药物清单.
- 通过使用Lexicomp,Medscape和Drugs.com.com建立了204个临床相关DDI的参考集合.
- 使用相同的提示来识别潜在的DDI,查询了ChatGPT,谷歌Gemini和微软Copilot.
主要成果:
- LLM 发现了大量潜在的 DDI,双子座发现了 1556 个和 Copilot 1813 个相互作用.
- 聊天GPT实现了最高的特异性 (0.868),而双子座具有最高的灵敏度 (0.697).
- 所有评估的LLM都表现出低精度,没有一个达到可靠临床决策所需的灵敏度和特异性的平衡,ChatGPT显示出最高的F1得分 (0.2520).
结论:
- 在识别真正的DDI方面,LLM是有前途的,但受限于产生临床不准确信息的"幻觉".
- 目前的LLM作为独立的DDI查工具是不可靠的,需要专业验证.
- 在管理多药房方面,LLM有可能支持临床药剂师,但人类监督对于患者安全至关重要.
相关概念视频
Bioequivalence of Drugs: Drugs with Multiple Indications
141
The concept of therapeutic equivalence (TE) in drugs with multiple indications is complex. A generic drug may be therapeutically equivalent to a brand-name product for one specific indication, but this doesn't necessarily mean it's equivalent for all other indications. Evidence of TE in one patient group and bioequivalence shown in healthy volunteers can support—but not confirm—TE for other indications. However, definitive proof requires individual clinical studies for each...
141
Quantitative Aspects of Drug-Receptor Interaction
1.7K
The receptor occupancy theory connects a drug's response to the number of occupied receptors. With higher drug concentrations, more receptors are occupied, leading to increased responses. The formation of drug-receptor complexes involves association and dissociation rates, which reach equilibrium when the forward and backward reactions are equal. The equilibrium association constant (Ka) and its inverse, the equilibrium dissociation constant (Kd), indicate drug affinity. Higher Ka and lower...
1.7K
Structure-Activity Relationships and Drug Design
1.6K
Drug design is a dynamic field that involves discovering and developing new medications based on specific biological targets. This process heavily relies on structure-activity relationships (SAR) and quantitative structure-activity relationships (QSAR) to guide the design and optimization of efficient drugs.
SAR studies the intricate relationship between a drug's chemical structure and biological activity. It focuses on understanding how modifications to a drug's structure can influence...
SAR studies the intricate relationship between a drug's chemical structure and biological activity. It focuses on understanding how modifications to a drug's structure can influence...
1.6K
Drug Products: Biologics, Biosimilars and Interchangeables
217
Body:Biologics, derived from living sources such as humans, animals, or microorganisms, represent a significant category of pharmaceuticals. These complex molecules, developed through advanced biotechnological methods or purified from natural sources, include essential medical treatments like insulin and growth hormones. The complexity of biologics arises from their large molecular structures and the intricate processes required for their production, making them distinct from conventional...
217


