• Corpus ID: 19162791

Predicting ICU Mortality Risk by Grouping Temporal Trends from a Multivariate Panel of Physiologic Measurements

@inproceedings{Luo2016PredictingIM,
  title={Predicting ICU Mortality Risk by Grouping Temporal Trends from a Multivariate Panel of Physiologic Measurements},
  author={Yuan Luo and Yu Xin and Rohit Joshi and Leo Anthony Celi and Peter Szolovits},
  booktitle={AAAI},
  year={2016}
}
ICU mortality risk prediction may help clinicians take effective interventions to improve patient outcome. [] Key Method SANMF converts time series into a graph representation and applies frequent subgraph mining to automatically extract temporal trends. We then apply non-negative matrix factorization to group trends in a way that approximates patient pathophysiologic states.

Figures and Tables from this paper

Explainable machine learning prediction of ICU mortality
Supervised Nonnegative Matrix Factorization to Predict ICU Mortality Risk
TLDR
This work proposed a supervised SANMF algorithm by integrating the logistic regression loss function into the NMF framework and solved it with an alternating optimization procedure and applied it to ICU mortality risk prediction and demonstrated its superiority over other conventional supervised NMF methods.
Predicting ICU mortality by supervised bidirectional LSTM networks
TLDR
This paper proposes a novel ICU mortality prediction algorithm combining bidirectional LSTM (Long Short-Term Memory) model with supervised learning and shows that the proposed method can significantly outperform many baseline methods.
Unsupervised phenotyping of sepsis using nonnegative matrix factorization of temporal trends from a multivariate panel of physiological measurements
TLDR
These phenotypes were validated to be prognostic factors of mortality for sepsis patients and suggest that these phenotypes can be used to develop targeted therapies based on phenotypic heterogeneity and algorithms designed for monitoring, validating and intervening clinical decisions for septic patients.
Dynamic Prediction of ICU Mortality Risk Using Domain Adaptation
TLDR
This paper explores domain adaptation strategies in order to learn mortality prediction models that extract and transfer complex temporal features from multivariate time-series ICU data, and shows that the model outperforms all considered baselines.
Reproducibility in critical care: a mortality prediction case study
TLDR
This study reviews publications that have reported performance of mortality prediction models based on the Medical Information Mart for Intensive Care (MIMIC) database and attempts to reproduce the cohorts used in their studies, and compares the performance reported in the studies against gradient boosting and logistic regression models.
Early Prediction of Mortality in Critical Care Setting in Sepsis Patients Using Structured Features and Unstructured Clinical Notes
TLDR
The results showed that the approaches integrating both structured and unstructured clinical features can be effectively applied to assist clinicians in identifying the risk of mortality in sepsis patients upon admission to the ICU.
Feature Engineering for ICU Mortality Prediction Based on Hourly to Bi-Hourly Measurements
TLDR
The obtained results show that the proposed feature engineering approach and the extracted features are valid to be considered and further enhanced for the mortality prediction purpose, and moved the modelling methodology from black- box modelling to grey-box modelling in combination with the powerful classifier of SVMs.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 47 REFERENCES
ICU mortality prediction using time series motifs
TLDR
This paper presents an approach that identifies and integrates information in motifs that are statistically over- or under-represented in ICU time series of patients experiencing in-hospital mortality, and outperformed existing clinical scoring systems such as SAPSII, APACHEII and SOFA.
An imputation-enhanced algorithm for ICU mortality prediction
TLDR
An improved model for predicting in- hospital mortality using data collected from the first 48 hours of a patient's ICU stay is presented, providing better estimates of in-hospital mortality risk than existing methods such as SAPS-I.
Prognostic Physiology: Modeling Patient Severity in Intensive Care Units Using Radial Domain Folding
TLDR
This work describes a new unsupervised learning approach, radial domain folding, to scale and summarize the enormous amount of data collected and to visualize the degradations or improvements in multiple organ systems in real time.
Mortality risk assessment for ICU patients using logistic regression
TLDR
Using Physionet/CinC Challenge 2012 data set, maximum, mean and minimum are identified as potential features extracted from the parameters measured during patients stay of 48hrs at ICU to accurately predict in-hospital mortality risk.
Prediction of mortality in an intensive care unit using logistic regression and a hidden Markov model
TLDR
This work proposes an algorithm based on logistic regression and Hidden-Markov model using vital signs, laboratory values and fluid measurements that are commonly available in ICUs that could be used as a continuous, realtime patient specific indicator of mortality risk.
PhysioNet 2012 Challenge: Predicting mortality of ICU patients using a cascaded SVM-GLM paradigm
TLDR
An algorithm is developed that uses both general descriptors and time-series measurements to predict the in-hospital death (IHD) of ICU patients in Event 1, and to provide a probability estimate of IHD in Event 2.
Towards the prediction of mortality in Intensive Care Units patients: A Simple Correspondence Analysis approach
TLDR
Results show that variables such as creatinine, urine output, bilirubin and mechanical ventilation support were capable to discriminate between patients who survive or do not survive their ICU stays.
Combining machine learning and clinical rules to build an algorithm for predicting ICU mortality risk
TLDR
A decision support application for predicting ICU mortality risk that starts with a clinical analysis of the problem that also leverages machine learning to help create an algorithm with good performance characteristics is developed.
Predicting in-hospital mortality of ICU patients: The PhysioNet/Computing in cardiology challenge 2012
TLDR
The focus of the PhysioNet/CinC Challenge 2012 is to develop methods for patient-specific prediction of in-hospital mortality by measuring performance of a binary classifier and a risk estimator.
A neural network model for mortality prediction in ICU
TLDR
A neural network model was developed to predict the risk of inhospital mortality using various physiological measurements from the ICU and yielded an event 1 score of 0.5088 and an event 2 score of 82.211 on the test data set.
...
1
2
3
4
5
...