使用大型语言模型检测焦虑和恶心/吐记录在癌症患者的临床笔记中
Nahid Zeinali1, Alaa Albashayreh2, Weiguo Fan3
1Department of Computer Science and Informatics, University of Iowa.
Computers, informatics, nursing : CIN
|December 23, 2025
概括
大型语言模型 (LLM) 增强了癌症患者的症状检测.
科学领域:
- 医疗保健中的计算语言学
- 自然语言处理 (NLP) 用于临床数据分析.
- 瘤学患者护理和症状管理
背景情况:
- 大型语言模型 (LLM) 在医疗保健中显示出命名实体识别 (NER) 的前景.
- 在电子健康记录 (EHR) 中准确的症状检测对于患者护理至关重要.
- 在癌症患者中识别心理和身体症状有助于治疗规划.
研究的目的:
- 在癌症患者的临床笔记中评估LLM在识别焦虑和恶心/吐症状方面的表现.
- 为了比较微调与基于提示的学习策略,用于症状检测.
- 评估基于LLM的症状检测对瘤护理团队的有用性.
主要方法:
- 使用命名实体识别 (NER) 来分析8,490名癌症患者的临床笔记.
- 在临床文本上预训练生物临床BERT和生物GPT模型.
- 实施症状-BERT和症状-GPT框架,使用微调和基于提示的学习.
- 使用F1分数对模型性能进行评估.
主要成果:
- 使用Symptom-BERT的微调实现了最高的F1分数:恶心/吐0.989和焦虑0.912.
- 微调优于基于提示的学习,特别是在记录良好的身体症状方面.
- 在28.69%的患者中检测到焦虑,在39.31%的患者中检测到恶心/吐.
结论:
- 微调的LLM,特别是Symptom-BERT,在临床叙述中为症状检测提供了更高的准确性.
- 基于LLM的症状检测可以帮助瘤护士进行早期识别和干预.
- 通过LLM的自动症状监测提高了瘤学中以患者为中心的护理.
相关概念视频
Chemotherapy-Induced Nausea and Vomiting: 5-HT3 Receptor Antagonists
564
5-HT3 receptor antagonists, such as dolasetron, granisetron (Kytril), ondansetron (Zofran), and palonosetron (Axoli), are crucial in managing chemotherapy-induced nausea and vomiting (CINV) and postoperative nausea. These drugs selectively block 5-HT3 receptors in the visceral vagal and spinal afferent nerves, chemoreceptor trigger zone, and the vomiting center. They have a rapid onset of action and can be given as a single dose before chemotherapy. Ondansetron and granisetron, in particular,...
564
Guidelines for Nursing Documentation I
1.7K
Quality documentation and reporting share essential characteristics that ensure they are practical and valuable resources for those who use them. These characteristics are:
Factual:
The following points emphasize the significance of upholding accurate and unbiased documentation in healthcare.
Factual:
The following points emphasize the significance of upholding accurate and unbiased documentation in healthcare.
1.7K
Chemotherapy-Induced Nausea and Vomiting: Neurokinin-1 Receptor Antagonists
516
Neurokinin 1 (NK1) receptors are distributed across the GI tract, vagal afferents, and key CNS regions including the central vomiting center and chemoreceptor trigger zone (CTZ) Chemotherapy agents stimulate enterochromaffin cells in the gastrointestinal (GI) tract to release large amounts of substance P (SP). SP is a neuropeptide released by specific sensory nerves in response to many different stressors, including those in the GI mucosa affected by chemotherapy. SP binds and activates...
516
Chemotherapy-Induced Nausea and Vomiting: Cannabinoids
649
Tetrahydrocannabinol (THC) is a phytocannabinoid that primarily interacts with the CB1 receptor, a type of G protein-coupled receptor (GPCR) predominantly in and around the chemoreceptor trigger zone (CTZ) and emetic center. THC also blocks the serotonin receptor activity in the dorsal vagal complex (DVC) by inhibiting serotonin release. THC exerts its anti-emetic effects through these interactions, which are beneficial for patients undergoing chemotherapy.
Two synthetic agonists of THC,...
Two synthetic agonists of THC,...
649
Chemotherapy-Induced Nausea and Vomiting: Dopamine Receptor Antagonists
785
Dopamine receptor antagonists, also known as antipsychotic agents, are critical in managing chemotherapy-induced vomiting. These antiemetic agents block dopamine receptors in the chemoreceptor trigger zone (CTZ), inhibiting signal transmission to the vomiting center. Antipsychotic agents encompass phenothiazines (PTZ), butyrophenones, benzamides, and thienobenzodiazepines (Zyprexa), which are utilized for their antiemetic and sedative properties.
Phenothiazines, such as prochlorperazine...
Phenothiazines, such as prochlorperazine...
785
Documentation of Nursing Diagnosis
1.6K
The nurse documents nursing diagnoses and enters them into the patient record. The identified patient's nursing diagnosis is either written out with a plan of care or entered into the electronic health record.
In some settings, data-driven computerized decision support systems are in place, allowing for more accurate nursing diagnoses. The database within one of these systems includes diagnostic labels defining characteristics, activities, and indicators for nursing. A nurse enters...
In some settings, data-driven computerized decision support systems are in place, allowing for more accurate nursing diagnoses. The database within one of these systems includes diagnostic labels defining characteristics, activities, and indicators for nursing. A nurse enters...
1.6K


