Patient representation learning based on electronic health records (EHR) is a
critical task for disease prediction. This task aims to effectively extract
useful information on dynamic features. Although various existing works have
achieved remarkable progress, the model performance can be further improved by
fully extracting the trends, variations, and the correlation between the trends
and variations in dynamic features. In addition, sparse visit records limit the
performance of deep learning models. To address these issues, we propose the
Multi-perspective Patient Representation Extractor (MPRE) for disease
prediction. Specifically, we propose Frequency Transformation Module (FTM) to
extract the trend and variation information of dynamic features in the
time-frequency domain, which can enhance the feature representation. In the 2D
Multi-Extraction Network (2D MEN), we form the 2D temporal tensor based on
trend and variation. Then, the correlations between trend and variation are
captured by the proposed dilated operation. Moreover, we propose the
First-Order Difference Attention Mechanism (FODAM) to calculate the
contributions of differences in adjacent variations to the disease diagnosis
adaptively. To evaluate the performance of MPRE and baseline methods, we
conduct extensive experiments on two real-world public datasets. The experiment
results show that MPRE outperforms state-of-the-art baseline methods in terms
of AUROC and AUPRC.

基于电子健康记录（EHR）的患者表征学习是一项关键任务，旨在有效提取动态特征的有用信息。我们提出了多角度患者表征提取器（MPRE），通过频率转换模块（FTM）提取动态特征的趋势和变异信息，在 2D 多提取网络（2D MEN）中捕获趋势和变异之间的相关性，并通过一阶差异注意机制（FODAM）自适应计算邻近变异的对疾病诊断的贡献。实验证明，MPRE 在 AUROC 和 AUPRC 等方面优于最先进的基准方法。

MPRE：疾病预测的多角度患者表示提取器

MPRE: Multi-perspective Patient Representation Extractor for Disease  Prediction

Electronic health records (EHRs) are usually highly dimensional,
heterogeneous, and multimodal. Besides, the random recording of clinical
variables results in high missing rates and uneven time intervals between
adjacent records in the multivariate clinical time-series data extracted from
EHRs. Current works using clinical time-series data for patient representation
regard the patients' physiological status as a discrete process described by
sporadically collected records. However, changes in the patient's physiological
condition are continuous and dynamic processes. The perception of time and
velocity of change is crucial for patient representation learning. In this
study, we propose a time- and velocity-aware gated recurrent unit model
(GRU-TV) for patient representation learning of clinical multivariate
time-series data in a time-continuous manner. The neural ordinary differential
equations (ODEs) and velocity perception mechanism are applied to perceive the
time interval between adjacent records and changing rate of the patient's
physiological status, respectively. Our experiments on two real clinical EHR
datasets (PhysioNet2012, MIMIC-III) establish that GRU-TV is a robust model on
computer-aided diagnosis (CAD) tasks, especially on sequences with
high-variance time intervals.

本研究提出了一种基于时间和速度敏感的门控循环单元模型 (GRU-TV)，用于连续时间学习临床多变量时间序列数据的患者表征。通过神经常微分方程和速度感知机制，实现了感知相邻记录之间的时间间隔和患者生理状态的变化率。实验结果表明，GRU-TV 模型在计算机辅助诊断任务中表现稳健，尤其是在具有高方差时间间隔的序列上表现良好。