知识卡片
纵向记录需要对齐预测时间点
病史与连续测量可为未来预测提供信息,但输入必须限定为预测时已知记录。就诊过程与疾病随时间变化,静态汇总可能丢失时序信息;预测风险也不同于证明某次干预能避免结局。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
- · 现实应用
原书出处 · 1 处
- · 查看原书依据
In addition, several studies have applied deep learning to predict diseases from a patient’s clinical status shown in their EHRs. For example, CNNs have been used to predict congestive heart failure and chronic obstructive pulmonary disease, demonstrating significant advantages over baseline methods. Recurrent neural networks (RNNs) have been used on word embeddings to infer current illness states and predict future outcomes, incorporating medical interventions for
参考来源
- 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第10章《现实应用》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/25_chapter10.xhtml)
- 结论依据:摘录中关于纵向记录需要对齐预测时间点的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。
- 原始内容:In addition, several studies have applied deep learning to predict diseases from a patient’s clinical status shown in their EHRs. For example, CNNs have been used to predict congestive heart failure and chronic obstructive pulmonary disease, demonstrating significant advantages over baseline methods. Recurrent neural networks (RNNs) have been used on word embeddings to infer current illness states and predict future outcomes, incorporating medical interventions for