TY - JOUR

T1 - CARRNN

T2 - A Continuous Autoregressive Recurrent Neural Network for Deep Representation Learning From Sporadic Temporal Data

AU - Ghazi, Mostafa Mehdipour

AU - Sørensen, Lauge

AU - Ourselin, Sébastien

AU - Nielsen, Mads

N1 - Publisher Copyright: IEEE

PY - 2024

Y1 - 2024

N2 - Learning temporal patterns from multivariate longitudinal data is challenging especially in cases when data is sporadic, as often seen in, e.g., healthcare applications where the data can suffer from irregularity and asynchronicity as the time between consecutive data points can vary across features and samples, hindering the application of existing deep learning models that are constructed for complete, evenly spaced data with fixed sequence lengths. In this article, a novel deep learning-based model is developed for modeling multiple temporal features in sporadic data using an integrated deep learning architecture based on a recurrent neural network (RNN) unit and a continuous-time autoregressive (CAR) model. The proposed model, called CARRNN, uses a generalized discrete-time autoregressive (AR) model that is trainable end-to-end using neural networks modulated by time lags to describe the changes caused by the irregularity and asynchronicity. It is applied to time-series regression and classification tasks for Alzheimer's disease progression modeling, intensive care unit (ICU) mortality rate prediction, human activity recognition, and event-based digit recognition, where the proposed model based on a gated recurrent unit (GRU) in all cases achieves significantly better predictive performance than the state-of-the-art methods using RNNs, GRUs, and long short-term memory (LSTM) networks.

AB - Learning temporal patterns from multivariate longitudinal data is challenging especially in cases when data is sporadic, as often seen in, e.g., healthcare applications where the data can suffer from irregularity and asynchronicity as the time between consecutive data points can vary across features and samples, hindering the application of existing deep learning models that are constructed for complete, evenly spaced data with fixed sequence lengths. In this article, a novel deep learning-based model is developed for modeling multiple temporal features in sporadic data using an integrated deep learning architecture based on a recurrent neural network (RNN) unit and a continuous-time autoregressive (CAR) model. The proposed model, called CARRNN, uses a generalized discrete-time autoregressive (AR) model that is trainable end-to-end using neural networks modulated by time lags to describe the changes caused by the irregularity and asynchronicity. It is applied to time-series regression and classification tasks for Alzheimer's disease progression modeling, intensive care unit (ICU) mortality rate prediction, human activity recognition, and event-based digit recognition, where the proposed model based on a gated recurrent unit (GRU) in all cases achieves significantly better predictive performance than the state-of-the-art methods using RNNs, GRUs, and long short-term memory (LSTM) networks.

KW - Automobiles

KW - Autoregressive (AR) model

KW - Data models

KW - Deep learning

KW - deep learning

KW - gated recurrent unit (GRU)

KW - Logic gates

KW - long short-term memory (LSTM) network

KW - Mathematical models

KW - multivariate time-series regression

KW - Predictive models

KW - recurrent neural network (RNN)

KW - Recurrent neural networks

KW - sporadic time series

U2 - 10.1109/TNNLS.2022.3177366

DO - 10.1109/TNNLS.2022.3177366

M3 - Journal article

C2 - 35666790

AN - SCOPUS:85131797267

VL - 35

SP - 792

EP - 802

JO - IEEE Transactions on Neural Networks and Learning Systems

JF - IEEE Transactions on Neural Networks and Learning Systems

SN - 2162-237X

IS - 1

ER -