TY - GEN
T1 - Learning outcome-discriminative dynamics in multivariate physiological cohort time series
AU - Nemati, Shamim
AU - Lehman, Li Wei H.
AU - Adams, Ryan P.
PY - 2013
Y1 - 2013
N2 - Model identification for physiological systems is complicated by changes between operating regimes and measurement artifacts. We present a solution to these problems by assuming that a cohort of physiological time series is generated by switching among a finite collection of physiologically-constrained dynamical models and artifactual segments. We model the resulting time series using the switching linear dynamical systems (SLDS) framework, and present a novel learning algorithm for the class of SLDS, with the objective of identifying time series dynamics that are predictive of physiological regimes or outcomes of interest. We present exploratory results based on a simulation study and a physiological classification example of decoding postural changes from heart rate and blood pressure. We demonstrate a significant improvement in classification over methods based on feature learning via expectation maximization. The proposed learning algorithm is general, and can be extended to other applications involving state-space formulations.
AB - Model identification for physiological systems is complicated by changes between operating regimes and measurement artifacts. We present a solution to these problems by assuming that a cohort of physiological time series is generated by switching among a finite collection of physiologically-constrained dynamical models and artifactual segments. We model the resulting time series using the switching linear dynamical systems (SLDS) framework, and present a novel learning algorithm for the class of SLDS, with the objective of identifying time series dynamics that are predictive of physiological regimes or outcomes of interest. We present exploratory results based on a simulation study and a physiological classification example of decoding postural changes from heart rate and blood pressure. We demonstrate a significant improvement in classification over methods based on feature learning via expectation maximization. The proposed learning algorithm is general, and can be extended to other applications involving state-space formulations.
UR - http://www.scopus.com/inward/record.url?scp=84886457948&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84886457948&partnerID=8YFLogxK
U2 - 10.1109/EMBC.2013.6611195
DO - 10.1109/EMBC.2013.6611195
M3 - Conference contribution
C2 - 24111382
AN - SCOPUS:84886457948
SN - 9781457702167
T3 - Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBS
SP - 7104
EP - 7107
BT - 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013
T2 - 2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2013
Y2 - 3 July 2013 through 7 July 2013
ER -