Modeling time-series data and temporal features
The key challenge for analysis of clinical data is that EHRs consist of complex multivariate time series of clinical variables collected for a specific patient, such as laboratory test results, medication orders, physiological parameters, past patient's diagnoses, surgical interventions and their outcomes. A fundamental challenge the researchers and data analysts face is how to summarize and represent this complex time-series data in order to make them amenable to statistical analysis and modeling. Our work has focused primarily on predictive modeling problems and solutions where our goal is to identify or construct temporal features important for prediction of adverse events, patient outcomes, and future patient management decisions.
Our work on temporal features covers the following directions:
- Fixed feature template maps
- Temporal predictive patterns features
- State-space models of dynamics
Fixed feature template maps
The approach relies on a fixed set of feature template mappings defined by domain experts that allows the conversion of complex time series of labs, medications or physiological values to different temporal features. For example, the time series of values for a numerical lab (such as platelet count) were converted to 28 different features reflecting the last lab value observed, the time elapsed since the last value, last trend, apex and horizon values, and their differences from last value, etc. We have developed similar feature sets for medications orders (e.g. aspirin or heparin), physiological parameters, input/output volumes, etc. We have successfully used these feature sets in modeling the patient state for adverse event detection, and in the outlier based alerting projects.
Selected publications:
- Milos Hauskrecht, I. Batal, C. Hong, Q. Nguyen, G. Cooper, S. Visweswaran, G. Clermont.
Outlier-based detection of unusual patient-management actions: An ICU study.
Journal of Biomedical Informatics, vol. 64, December 2016, pp 211-221.
- M. Hauskrecht, I. Batal, M. Valko, S. Visweswaran, G. Cooper, G. Clermont.
Outlier-detection for patient monitoring and alerting.
Journal of Biomedical Informatics, 46:1, pages 47 -- 55, February 2013.
- M. Hauskrecht, M. Valko, I.Batal, G. Clermont, S. Visweswaran, G. Cooper.
Conditional Outlier Detection for Clinical Alerting,
Annual American Medical Informatics Association (AMIA) Symposium , November 2010 [Homer Warner Award]
- M. Valko, and M. Hauskrecht.
Feature importance analysis for patient management
decisions.
13th International Congress on Medical Informatics , Cape Town, South Africa,
September 2010.
Features based on temporal predictive patterns
The approach constructs temporal features by extracting predictive patterns characterizing patient subgroups in EHRs that are important for prediction. Such patterns, when explicitly represented, can be used as features one needs to include when defining classification models. Our work focuses on temporal patterns based on temporal abstraction and temporal logic to get a high-level qualitative description of clinical time series. This representation is more flexible and allows us to express various temporal patterns observed in clinical time-series data.
Selected publications:
- I. Batal, G. Cooper, D. Fradkin, J. Harrison, F. Moerchen, and M. Hauskrecht.
An Efficient Pattern Mining Approach for Event
Detection in Multivariate Temporal Data (review version)
Knowledge and Information Science Journal , 46(1): 115-150, 2016 (online since 2014)
- I. Batal, H. Valizadegan, G. Cooper and M. Hauskrecht.
A Temporal
Pattern Mining Approach for Classifying Electronic Health Record Data.
Transactions on Intelligent Systems and Technology,
Special Issue on Health Informatics, 4: 4, 2013.
- I. Batal, D. Fradkin, J. Harrison, F. Moerchen, and M. Hauskrecht.
Mining Recent Temporal Patterns for Event Detection in Multivariate Time Series Data.
The 18th ACMSIGKDD Conference on Knowledge Discovery and Data Mining (KDD), Beijing, China, August 2012.
- I. Batal, H. Valizadegan, GF. Cooper, and M. Hauskrecht.
A Pattern Mining Approach for Classifying Multivariate Temporal Data,
IEEE International Conference on Bioinformatics and Biomedicine , Atlanta, Georgia, November 2011.
State-space models of dynamics
The approach we are currently exploring attempts to build models of multivariate time series data and their behaviors by exploring lower-dimensional representations of the patient state with the help of Markov process models. Briefly, our goal is to find a lower dimensional patient state representation that summarizes and compactly encodes all information about past patient's observations that is needed to predict well the behavior of the time series in the future. Such a state representation can yield a compact feature space for various prediction problems in clinical time series and would be an alternative to features based on fixed feature mappings or predictive patterns discussed earlier. We have studied and developed models based on Linear dynamical system and Gaussian processes frameworks, and their hierarchical combination to model time-series of different laboratory tests. Our most recent work focuses on the development of a sparse LDS framework that aims to recover lower dimensional representation of the hidden LDS state.
Selected publications:
- Zitao Liu and Milos Hauskrecht.
Learning Adaptive Forecasting Models from Irregularly Sampled Multivariate Clinical Data
The 30th AAAI Conference on Artificial Intelligence(AAAI), Phoenix, AZ, 2016. (revised version with additional results)
- Zitao Liu and Milos Hauskrecht.
Learning Linear Dynamical Systems from Multivariate Time Series: A Matrix Factorization Based Framework
SIAM International Conference on Data Mining (SDM), Miami, FL, 2016.
- Z. Liu, and M. Hauskrecht.
A Regularized Linear Dynamical System
Framework for Multivariate Time Series Analysis.
The Twenty-Ninth AAAI Conference on
Artificial Intelligence (AAAI-15), Austin, TX, January 2015.
- Z. Liu, and M. Hauskrecht.
Clinical time series prediction: Towards a hierarchical dynamical system framework
Journal of Artificial Intelligence in Medicine, 2014, accepted
- Z. Liu, and M. Hauskrecht.
Sparse Linear Dynamical System with Its Application in Multivariate Clinical Time Series.
NIPS 2013 Workshop on Machine Learning for Clinical Data Analysis and Healthcare, December 2013.
- Z. Liu, and M. Hauskrecht.
Clinical Time Series Prediction with a
Hierarchical Dynamical System.
The 14th Conference on Artificial Intelligence in Medicine, Murcia, Spain, May 2013.
- Z. Liu, L. Wu, and M. Hauskrecht.
Modeling Clinical Time Series Using Gaussian Process
Sequences.
SIAM Data Mining Conference , Austin, TX, April 2013.
Funding:
- NIH. 1R01LM010019. Using medical records repositories to improve the alert system design. PI: Hauskrecht, September 2009- September 2013.
- NIH. 1R01GM088224 Detecting deviations in clinical care in ICU data stream. PIs: Hauskrecht and Clermont , August 2009-June 2018.
- NIH. 1R21LM009102-01A1 Evidence-based Anomaly Detection in Clinical Databases.
PI: Hauskrecht , April 2007-September 2009.
The web page is updated by milos.