Clinical predictive models often rely on patients electronic health records (EHR), but integrating medical knowledge to enhance predictions and decision-making is challenging. This is because personalized predictions require personalized knowledge graphs (KGs), which are difficult to generate from patient EHR data. To address this, we propose GraphCare, an open-world framework that leverages external KGs to improve EHR-based predictions. Our method extracts knowledge from large language models (LLMs) and external biomedical KGs to generate patient-specific KGs, which are then used to train our proposed Bi-attention AugmenTed BAT graph neural network GNN for healthcare predictions. We evaluate GraphCare on two public datasets: MIMIC-III and MIMIC-IV. Our method outperforms baseline models in four vital healthcare prediction tasks: mortality, readmission, length-of-stay, and drug recommendation, improving AUROC on MIMIC-III by average margins of 10.4%, 3.8%, 2.0%, and 1.5%, respectively. Notably, GraphCare demonstrates a substantial edge in scenarios with limited data availability. Our findings highlight the potential of using external KGs in healthcare prediction tasks and demonstrate the promise of GraphCare in generating personalized KGs for promoting personalized medicine.

翻译：摘要：临床预测模型通常依赖于患者的电子健康记录（EHR），但整合医学知识以增强预测和决策制定颇具挑战性。这是因为个性化预测需要个性化知识图谱（KG），而此类知识图谱难以从患者EHR数据中生成。为解决这一问题，我们提出GraphCare——一种利用外部KG改进基于EHR预测的开放世界框架。该方法从大型语言模型（LLM）和外部生物医学KG中提取知识，生成患者特异的KG，进而用于训练我们提出的双注意力增强蝙蝠图神经网络（Bi-attention AugmenTed BAT graph neural network，简称BAT GNN）以实现医疗预测。我们在两个公开数据集（MIMIC-III和MIMIC-IV）上评估GraphCare。在死亡率预测、再入院预测、住院时长预测及药物推荐四项关键医疗预测任务中，我们的方法优于基线模型，在MIMIC-III数据集上平均AUROC分别提升10.4%、3.8%、2.0%和1.5%。值得注意的是，GraphCare在数据可用性有限的场景中表现出显著优势。我们的研究结果凸显了将外部KG应用于医疗预测任务的潜力，并展示了GraphCare在生成个性化KG以推动个性化医疗方面的前景。