Neural Clinical Event Sequence Prediction through Personalized Online Adaptive Learning

@article{Lee2021NeuralCE,
  title={Neural Clinical Event Sequence Prediction through Personalized Online Adaptive Learning},
  author={Jeong Min Lee and Milos Hauskrecht},
  journal={Artificial intelligence in medicine. Conference on Artificial Intelligence in Medicine},
  year={2021},
  volume={12721},
  pages={
          175-186
        }
}
  • J. Lee, M. Hauskrecht
  • Published 5 April 2021
  • Biology, Psychology, Computer Science, Medicine
  • Artificial intelligence in medicine. Conference on Artificial Intelligence in Medicine
Clinical event sequences consist of thousands of clinical events that represent records of patient care in time. Developing accurate prediction models for such sequences is of a great importance for defining representations of a patient state and for improving patient care. One important challenge of learning a good predictive model of clinical sequences is patient-specific variability. Based on underlying clinical complications, each patient's sequence may consist of different sets of clinical… 
Learning to Adapt Clinical Sequences with Residual Mixture of Experts
TLDR
This work refining a one-fits-all model using a Mixture-of-Experts (MoE) architecture that augment MoE based on the prediction signal from pretrained base GRU model to provide flexible adaptation to the (limited) predictive power of the single base RNN model.
Learning to Adapt Dynamic Clinical Event Sequences with Residual Mixture of Experts
TLDR
This work refining a one-fits-all model using a Mixture-of-Experts (MoE) architecture that augment MoE based on the prediction signal from pretrained base GRU model to provide flexible adaptation to the (limited) predictive power of the single base RNN model.

References

SHOWING 1-10 OF 30 REFERENCES
Modeling multivariate clinical event time-series with recurrent temporal mechanisms
Clinical Event Time-Series Modeling with Periodic Events
TLDR
This work proposes multiple simple periodic mechanisms that help to drive the expression of individual events in time that can be effectively combined with more complex neural architectures capable of modeling the dependencies among different types of events.
Monitoring ICU Mortality Risk with A Long Short-Term Memory Recurrent Neural Network
TLDR
A new framework for dynamic monitoring of patients’ mortality risk uses the bag-of-words representation for all relevant medical events based on most recent history as inputs and shows that the deep learning based framework performs better than the existing severity scoring system, SAPS-II.
Multi-scale Temporal Memory for Clinical Event Time-Series Prediction
TLDR
This work proposes a new neural memory module called Multi-scale Temporal Memory (MTM) linking events in a distant past with the current prediction time and demonstrates the efficacy of MTM by combining it with different patient state summarization methods that cover different temporal aspects of patient states.
CAMP: Co-Attention Memory Networks for Diagnosis Prediction in Healthcare
TLDR
A model called Co-Attention Memory networks for diagnosis Prediction (CAMP) is proposed, which tightly integrates historical records, fine-grained patient conditions, and demographics with a three-way interaction architecture built on co-attention.
Hierarchical Adaptive Multi-task Learning Framework for Patient Diagnoses and Diagnostic Category Classification
TLDR
This work designs a new hierarchical classification learning framework in which multiple diagnostic classification targets are explicitly related via diagnostic hierarchy relations and demonstrates that this framework leads to improved classification performance on individual diagnostic tasks when compared to independently learned diagnostic models.
Nonparametric Regressive Point Processes Based on Conditional Gaussian Processes
TLDR
This work proposes and develops a new nonparametric regressive point process model based on Gaussian processes that can represent better many commonly observed real-world event sequences and capture the dependencies between events that are difficult to model using existingnonparametric Hawkes process variants.
Predicting Patient's Diagnoses and Diagnostic Categories from Clinical-Events in EHR Data
In this paper we develop and study machine learning based models based on latent semantic indexing capable of automatically assigning diagnoses and diagnostic categories to patients based on
Recent context-aware LSTM-based clinical time-series prediction
  • In Intl Conf on AI in Medicine (AIME),
  • 2019
Patient2Vec: A Personalized Interpretable Deep Representation of the Longitudinal Electronic Health Record
TLDR
A computational framework, Patient2Vec, is proposed to learn an interpretable deep representation of longitudinal EHR data, which is personalized for each patient, and it achieves an area under curve around 0.799, outperforming baseline methods.
...
1
2
3
...