Stochastic Imputation and Uncertainty-Aware Attention to EHR for Mortality Prediction

Jun, Eunji; Mulyadi, Ahmad Wisnu; Suk, Heung-Il

doi:10.1109/ijcnn.2019.8852132

Cited by 18 publications

(12 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We use the temporal decay factor to consider the effect of irregular time intervals between the EHR time-series data. Jun et al [20] use variational autoencoders to incorporate the variance in the latent distribution of the data. This model was later enhanced by Mulyadi et al [26] by adding recurrent layers to consider the temporal dynamics and the correlations between the input features during the training.…”

Section: Related Workmentioning

confidence: 99%

Concurrent imputation and prediction on EHR data using bi-directional GANs

Gupta

Phan

Bunnell

et al. 2021

Proceedings of the 12th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics

View full text Add to dashboard Cite

Working with electronic health records (EHRs) is known to be challenging due to several reasons. These reasons include not having: 1) similar lengths (per visit), 2) the same number of observations (per patient), and 3) complete entries in the available records. These issues hinder the performance of the predictive models created using EHRs. In this paper, we approach these issues by presenting a model for the combined task of imputing and predicting values for the irregularly observed and varying length EHR data with missing entries. Our proposed model (dubbed as Bi-GAN) uses a bidirectional recurrent network in a generative adversarial setting. In this architecture, the generator is a bidirectional recurrent network that receives the EHR data and imputes the existing missing values. The discriminator attempts to discriminate between the actual and the imputed values generated by the generator. Using the input data in its entirety, Bi-GAN learns how to impute missing elements in-between (imputation) or outside of the input time steps (prediction). Our method has three advantages to the state-of-the-art methods in the field: (a) one single model performs both the imputation and prediction tasks; (b) the model can perform predictions using time-series of varying length with missing data; (c) it does not require to know the observation and prediction time window during training and can be used for the predictions with different observation and prediction window lengths, for short-and long-term predictions. We evaluate our model on two large EHR datasets to impute and predict body mass index (BMI) values and show its superior performance in both settings.

show abstract

Section: Related Workmentioning

confidence: 99%

Concurrent imputation and prediction on EHR data using bi-directional GANs

Gupta

Phan

Bunnell

et al. 2021

Proceedings of the 12th ACM International Conference on Bioinformatics, Computational Biology, and Health Informatics

View full text Add to dashboard Cite

show abstract

“…This mainly includes human faces [e.g., 42,77], but also landscapes, buildings, or other scenes [47,74]. Next to visual data, for instance, VAEs can infer missing values within electronic health records of patients to predict their in-hospital mortality [36] and transform noisy speech (e.g., due to background sounds) to clean speech [e.g., 2, 58].…”

Section: Reconstructingmentioning

confidence: 99%

Innovating with Artificial Intelligence: Capturing the Constructive Functional Capabilities of Deep Generative Learning

Hofmann¹,

Rückel²,

Urbach³

2021

Proceedings of the Annual Hawaii International Conference on System Sciences

View full text Add to dashboard Cite

As an emerging species of artificial intelligence, deep generative learning models can generate an unprecedented variety of new outputs. Examples include the creation of music, text-to-image translation, or the imputation of missing data. Similar to other AI models that already evoke significant changes in society and economy, there is a need for structuring the constructive functional capabilities of DGL. To derive and discuss them, we conducted an extensive and structured literature review. Our results reveal a substantial scope of six constructive functional capabilities demonstrating that DGL is not exclusively used to generate unseen outputs. Our paper further guides companies in capturing and evaluating DGL's potential for innovation. Besides, our paper fosters an understanding of DGL and provides a conceptual basis for further research.

show abstract

“…The training could be regarded as a generative process where a set of data points is drawn from the distribution to approximate the true underlying distribution. [21] proposed to use VAE to impute missing values for electronic health data with uncertaintyaware attention. Experiments on real world datasets show that VAE is able to capture the complexity of EHR distribution.…”

Section: Variational Autoencodermentioning

confidence: 99%

Temporal Clustering with External Memory Network for Disease Progression Modeling

Zhang¹,

Yin²,

Zhang³

2021

Preprint

View full text Add to dashboard Cite

Disease progression modeling (DPM) involves using mathematical frameworks to quantitatively measure the severity of how certain disease progresses. DPM is useful in many ways such as predicting health state, categorizing disease stages, and assessing patients disease trajectory etc. Recently, with wider availability of electronic health records (EHR) and the broad application of data-driven machine learning method, DPM has attracted much attention yet remains two major challenges: (i) Due to the existence of irregularity, heterogeneity and longterm dependency in EHRs, most existing DPM methods might not be able to provide comprehensive patient representations. (ii) Lots of records in EHRs might be irrelevant to the target disease. Most existing models learn to automatically focus on the relevant information instead of explicitly capture the targetrelevant events, which might make the learned model suboptimal.To address these two issues, we propose Temporal Clustering with External Memory Network (TC-EMNet) for DPM that groups patients with similar trajectories to form disease clusters/stages. TC-EMNet uses a variational autoencoder (VAE) to capture internal complexity from the input data and utilizes an external memory work to capture long term distance information, both of which are helpful for producing comprehensive patient states. Last but not least, k-means algorithm is adopted to cluster the extracted comprehensive patient states to capture disease progression. Experiments on two real-world datasets show that our model demonstrates competitive clustering performance against state-of-the-art methods and is able to identify clinically meaningful clusters. The visualization of the extracted patient states shows that the proposed model can generate better patient states than the baselines.

show abstract

Stochastic Imputation and Uncertainty-Aware Attention to EHR for Mortality Prediction

Cited by 18 publications

References 7 publications

Concurrent imputation and prediction on EHR data using bi-directional GANs

Concurrent imputation and prediction on EHR data using bi-directional GANs

Innovating with Artificial Intelligence: Capturing the Constructive Functional Capabilities of Deep Generative Learning

Temporal Clustering with External Memory Network for Disease Progression Modeling

Contact Info

Product

Resources

About