Uncertainty-Aware Variational-Recurrent Imputation Network for Clinical Time Series

  title={Uncertainty-Aware Variational-Recurrent Imputation Network for Clinical Time Series},
  author={Ahmad Wisnu Mulyadi and Eunji Jun and Heung-Il Suk},
  journal={IEEE transactions on cybernetics},
Electronic health records (EHR) consist of longitudinal clinical observations portrayed with sparsity, irregularity, and high dimensionality, which become major obstacles in drawing reliable downstream clinical outcomes. Although there exist great numbers of imputation methods to tackle these issues, most of them ignore correlated features, temporal dynamics, and entirely set aside the uncertainty. Since the missing value estimates involve the risk of being inaccurate, it is appropriate for the… 
Time-series Imputation and Prediction with Bi-Directional Generative Adversarial Networks
This paper presents a model for the combined task of imputing and predicting values for the irregularly observed and varying length time-series data with missing entries, and uses a bidirectional recurrent network in a generative adversarial setting.
CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation
Conditional Score-based Diffusion models for Imputation (CSDI), a novel time series imputation method that utilizes score-based diffusion models conditioned on observed data, which improves by 40-65% over existing probabilistic imputation methods on popular performance metrics.
Concurrent imputation and prediction on EHR data using bi-directional GANs: Bi-GANs for EHR imputation and prediction
The proposed model (dubbed as Bi-GAN) uses a bidirectional recurrent network in a generative adversarial setting that receives the EHR data and imputes the existing missing values and shows its superior performance in both settings.
Deep recurrent model for individualized prediction of Alzheimer’s disease progression
A deep recurrent network is proposed that jointly tackles the four problems of missing value imputation, phenotypic measurements forecasting, trajectory estimation of a cognitive score, and clinical status prediction of a subject based on his/her longitudinal imaging biomarkers.
Self-Supervised Graph Learning with Hyperbolic Embedding for Temporal Health Event Prediction
A hyperbolic embedding method with information flow to pre-train medical code representations in a hierarchical structure is proposed and incorporated into a graph neural network to detect disease complications, and a multi-level attention method to compute the contributions of particular diseases and admissions, thus enhancing personalized interpretability.
Unifying Domain Adaptation and Domain Generalization for Robust Prediction Across Minority Racial Groups
A multi-source adversarial domain separation (MS-ADS) framework which unifies domain adaptation and domain generalization is presented which shows robust out-performance by addressing covariate shift across different racial groups and systematic bias across medical systems simultaneously.
An Adversarial Domain Separation Framework for Septic Shock Early Prediction Across EHR Systems
By separating globally-shared from domain-specific representations, this framework significantly improves septic shock early prediction performance in both EHRs and outperforms the current state-of-the-art DA models.


Stochastic Imputation and Uncertainty-Aware Attention to EHR for Mortality Prediction
This paper utilizes the uncertainty obtained from the generative network of the VAE and employ uncertainty-aware attention in imputing the missing values and shows higher performance than other competing methods in mortality prediction task.
GP-VAE: Deep Probabilistic Time Series Imputation
This work proposes a new deep sequential latent variable model for dimensionality reduction and data imputation of multivariate time series from the domains of computer vision and healthcare, and demonstrates that this approach outperforms several classical and deep learning-based data imputations methods on high-dimensional data.
Missing Data Imputation in the Electronic Health Record Using Deeply Learned Autoencoders
It is shown that despite clinical heterogeneity, ALS disease progression appears homogenous with time from onset being the most important predictor.
Recurrent Neural Networks for Multivariate Time Series with Missing Values
Novel deep learning models are developed based on Gated Recurrent Unit, a state-of-the-art recurrent neural network that takes two representations of missing patterns, i.e., masking and time interval, and effectively incorporates them into a deep model architecture so that it not only captures the long-term temporal dependencies in time series, but also utilizes the missing patterns to achieve better prediction results.
Multivariate Time Series Imputation with Generative Adversarial Networks
Experiments show that the proposed model outperformed the baselines in terms of accuracy of imputation, and a simple model on the imputed data can achieve state-of-the-art results on the prediction tasks, demonstrating the benefits of the model in downstream applications.
BRITS: Bidirectional Recurrent Imputation for Time Series
BRITS is a novel method based on recurrent neural networks for missing value imputation in time series data that directly learns the missing values in a bidirectional recurrent dynamical system, without any specific assumption.
Risk Prediction with Electronic Health Records: A Deep Learning Approach
A deep learning approach for phenotyping from patient EHRs by building a fourlayer convolutional neural network model for extracting phenotypes and perform prediction and the proposed model is validated on a real world EHR data warehouse under the specific scenario of predictive modeling of chronic diseases.
Directly Modeling Missing Data in Sequences with RNNs: Improved Classification of Clinical Time Series
This work shows the remarkable ability of RNNs to make effective use of binary indicators to directly model missing data, improving AUC and F1 significantly and evaluating LSTMs, MLPs, and linear models trained on missingness patterns only.
Multi-directional Recurrent Neural Networks : A Novel Method for Estimating Missing Data
Most time-series datasets with multiple data streams have (many) missing measurements that need to be estimated. Most existing methods address this estimation problem either by interpolating within
Handling Incomplete Heterogeneous Data using VAEs
A general framework to design VAEs suitable for fitting incomplete heterogenous data, which includes likelihood models for real-valued, positive real valued, interval, categorical, ordinal and count data, and allows accurate estimation of missing data is proposed.