Trajectory-guided discharge stratification for heart failure using short-context electronic health record sequence modeling
利用序列建模预测心力衰竭患者一年临床不稳定性和死亡率
机构 * Sahlgrenska University Hospital(斯德哥尔摩大学医院) ; Department of Computer Science and Engineering, Chalmers University of Technology and University of Gothenburg(计算机科学与工程系,查尔姆斯理工大学和乌普萨拉大学) ; Department of Molecular and Clinical Medicine, Sahlgrenska Academy, University of Gothenburg(分子与临床医学系,斯德哥尔摩学院,乌普萨拉大学) ; Department of Medicine, Geriatrics and Emergency Medicine, Sahlgrenska University Hospital(医学、老年医学和急诊医学系,斯德哥尔摩大学医院) ; Department of Food and Nutrition, and Sport Science, Faculty of Education, University of Gothenburg(营养学与体育科学系,教育学院,乌普萨拉大学) ; School of Public Health and Community Medicine, Institute of Medicine, University of Gothenburg(公共卫生与社区医学系,医学院,乌普萨拉大学)
专题命中 临床大模型 :diagnosis(abstract);分类 cs.LG
AI总结 研究通过序列模型预测心力衰竭患者一年内的临床不稳定性和死亡风险,发现Llama模型在有限数据下表现优异,为出院后风险分层提供依据。