提出 XTSFormer 模型,通过周期感知时间位置编码(FCPE)和分层多尺度时间注意力机制,有效解决临床 EHR 中不规则时间事件的多尺度交互与预测问题,在真实数据集上优于多种基线方法。
摘要:
与不安全护理相关的不良临床事件是美国十大死因之一。基于电子健康记录(EHRs)的临床事件建模与预测在增强患者安全方面发挥着关键作用。一个例子是对事实护理路径进行建模,以表征治疗或护理的常见分步计划。然而,临床事件数据带来了一些独特的挑战,包括连续事件间时间间隔的不规则性、循环的存在、周期性、多尺度事件交互,以及与长事件序列相关的高计算成本。现有的神经时间点过程(TPPs)方法未能有效捕捉事件交互的多尺度特性,而这在许多现实临床应用中很常见。为解决这些问题,我们提出了跨时间尺度 Transformer(XTSFormer),专为不规则时间事件数据设计。该模型包含两个关键组件:一种新颖的基于特征的周期感知时间位置编码(FCPE),能巧妙捕捉时间的周期性;以及一种分层多尺度时间注意力机制,其不同时间尺度由自底向上的聚类方法确定。在多个真实 EHR 数据集上的大量实验表明,我们的 XTSFormer 优于多种基线方法。代码可在 https://github.com/spatialdatasciencegroup/XTSFormer 获取。
英文摘要:
Adverse clinical events related to unsafe care are among the top ten causes of death in the U.S. Accurate modeling and prediction of clinical events from electronic health records (EHRs) play a crucial role in patient safety enhancement. An example is modeling de facto care pathways that characterize common step-by-step plans for treatment or care. However, clinical event data pose several unique challenges, including the irregularity of time intervals between consecutive events, the existence of cycles, periodicity, multi-scale event interactions, and the high computational costs associated with long event sequences. Existing neural temporal point processes (TPPs) methods do not effectively capture the multi-scale nature of event interactions, which is common in many real-world clinical applications. To address these issues, we propose the cross-temporal-scale transformer (XTSFormer), specifically designed for irregularly timed event data. Our model consists of two vital components: a novel Feature-based Cycle-aware Time Positional Encoding (FCPE) that adeptly captures the cyclical nature of time, and a hierarchical multi-scale temporal attention mechanism, where different temporal scales are determined by a bottom-up clustering approach. Extensive experiments on several real-world EHR datasets show that our XTSFormer outperforms multiple baseline methods. The code is available at https://github.com/spatialdatasciencegroup/XTSFormer.
TA-RNN-Medical-Hybrid: A Time-Aware and Interpretable Framework for Mortality Risk Prediction
Department of Computer Engineering, SRC.,Islamic Azad university(计算机工程系,伊斯兰教Azad大学);School of Computer Engineering, Iran university of Science and technology(计算机工程学院,伊朗科学技术大学)
Temporal Fusion Nexus: A task-agnostic multi-modal embedding model for clinical narratives and irregular time series in post-kidney transplant care
Hahn-Schickard(哈恩-施克特德研究所);University of Freiburg(弗赖堡大学);Charité University Medical Center(查理医院大学医学中心);Hasso Plattner Institute for Digital Engineering, University of Potsdam(哈索·platner数字工程研究所,波茨坦大学);DFKI(德意志联邦防务研究院);University Hospital Essen(埃森大学医院);University Hospital Erlangen(埃尔兰根大学医院)
Time-Aware Attention for Enhanced Electronic Health Records Modeling
Department of Statistics and Data Science, National University of Singapore(统计与数据科学系,新加坡国立大学);School of Computing, National University of Singapore(计算学院,新加坡国立大学);Department of Biostatistics, Harvard T.H. Chan School of Public Health, USA(生物统计学系,哈佛T.H. Chan公共卫生学院,美国)
Mind the Missing: Variable-Aware Representation Learning for Irregular EHR Time Series using Large Language Models
School of Industrial Management Engineering, Korea University(韩国大学工业管理工程学院);Division of Pulmonary, Allergy, Critical Care, and Sleep Medicine, Department of Medicine, University of Pittsburgh(匹兹堡大学医学部呼吸科、过敏科、重症医学科及睡眠医学科)
Self-Supervised Transformer for Sparse and Irregularly Sampled Multivariate Clinical Time-Series
Virginia Tech(弗吉尼亚理工大学)
在重症监护环境中经常观察到多变量时间序列数据,其典型特征是稀疏性(缺失信息)和不规则的时间间隔。在这个领域中用于学习表示的现有方法通过聚集或估算值来处理这些挑战,这反过来又抑制了细粒度信息并将不期望的噪声/开销添加到机器学习模型中。为了解决这个问题,我们提出了一个自我监督的Transformer for Time-Series(STraTS)模型,它克服了这些缺陷,将时间序列视为一组观察三元组,而不是使用标准的稠密矩阵表示。它采用了一种新的连续值嵌入技术来编码连续时间和变量值,而无需离散化。它由一个带有多头注意力层的Transformer组件组成,这些层使它能够学习上下文三重嵌入,同时避免循环架构中发生的循环和梯度消失问题。此外,为了解决标记数据可用性有限的问题(这通常在许多医疗保健应用中观察到),STraTS通过利用未标记数据来利用自我监督,以通过使用时间序列预测作为辅助代理任务来学习更好的表示。在真实世界多变量临床时间序列基准数据集上的实验表明,STraTS具有比最先进的死亡率预测方法更好的预测性能,特别是当标记数据有限时。最后,我们还提出了一个可解释的STraTS版本,它可以识别时间序列数据中的重要测量。我们的数据预处理和模型实现代码可以在https://github.com/sindhura97/STraTS上找到。
Generative Modeling of Clinical Time Series via Latent Stochastic Differential Equations
Department of Machine Learning and Neural Computing, Donders Institute for Brain, Cognition and Behaviour, Radboud University(机器学习与神经计算系,多纳德斯脑认知行为研究所,拉德堡德大学)
DBGL: Decay-aware Bipartite Graph Learning for Irregular Medical Time Series Classification
Department of Electrical and Electronic Engineering, The University of Hong Kong, Hong Kong, China(香港大学电子与电气工程系);Shenzhen MSU-BIT University, China(深圳MSU-BIT大学);Department of Data Science, City University of Hong Kong, China(香港城市大学数据科学系);Engineering Research Centre of Applied Technology on Machine Translation and Artificial Intelligence, Macao Polytechnic University(澳门理工大学机器翻译与人工智能应用技术研究中心)
TrajSurv: Learning Continuous Latent Trajectories from Electronic Health Records for Trustworthy Survival Prediction
Department of Biomedical Informatics and Medical Education University of Washington(生物医学信息学与医学教育系,华盛顿大学);Fred Hutch Cancer Center(Fred Hutch癌症中心);Department of Biomedical Informatics Harvard University(生物医学信息学系,哈佛大学)
Explainable Artificial Intelligence Techniques for Irregular Temporal Classification of Multidrug Resistance Acquisition in Intensive Care Unit Patients
Rey Juan Carlos University(雷伊·胡安·卡洛斯大学);University Hospital of Fuenlabrada(芬拉布拉达大学医院)
Causal-Transformer with Adaptive Mutation-Locking for Early Prediction of Acute Kidney Injury
School of Electrical and Information Engineering, Tianjin University(天津大学电气与信息工程学院);Department of Nephrology, TEDA Hospital, Tianjin University(天津大学TEDA医院肾内科);Tianjin Medical University General Hospital(天津医科大学总医院);Department of Nephrology, First Teaching Hospital of Tianjin University of Traditional Chinese Medicine(天津中医药大学第一教学医院肾内科)
Does Deep Learning REALLY Outperform Non-deep Machine Learning for Clinical Prediction on Physiological Time Series?
Ricoh Software Research Center (Beijing) Co., Ltd.(理光软件研究所(北京)有限公司);Japan Advanced Institute of Science and Technology (JAIST)(日本北陆先端科学技术大学院大学(JAIST));Yale University School of Medicine(耶鲁大学医学院);The Second Xiangya Hospital, Central South University(中南大学湘雅二医院)
Transition-Based Digital Twin Modelling for Alzheimer's Disease under Sparse Longitudinal Data
University of Southampton(南安普顿大学);University Hospital Southampton NHS Foundation Trust(南安普顿大学医院NHS基金会信托);Faculty of Medicine, University of Southampton(南安普顿大学医学院)
Temporal Gaussian Copula For Clinical Multivariate Time Series Data Imputation
Chongqing Institute of Green and Intelligent Technology, CAS(中国科学院重庆绿色智能技术研究院);University of Chinese Academy of Sciences(中国科学院大学);Singapore Management University(新加坡管理大学);Southwest University(西南大学)
SurvLatent ODE : A Neural ODE based time-to-event model with competing risks for longitudinal data improves cancer-associated Venous Thromboembolism (VTE) prediction
Massachusetts Institute of Technology(麻省理工学院);Dana-Farber Cancer Institute(达纳-法伯癌症研究所)
Trajectory Flow Matching with Applications to Clinical Time Series Modeling
McGill University(麦吉尔大学);Mila - Quebec AI Institute(米拉-魁北克人工智能研究所);Yale School of Medicine(耶鲁医学院);School of Clinical Medicine, University of Cambridge(剑桥大学临床医学院);Université de Montréal(蒙特利尔大学);CIFAR(加拿大高级研究院)
Geometry adaptive waveformer for cardio-vascular modeling
Indian Institute of Technology (IIT) Delhi(印度理工学院德里分校);Yardi School of Artificial Intelligence(亚迪人工智能学院);University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)