New insights into the mechanisms underlying cardiac arrhythmias may be possible from a greater understanding of variables unobserved in experimental settings. When experimental observations are sparse, data assimilation can be effective at recovering a more complete system state estimate. We have recently shown using synthetic observations of one- and three-dimensional cardiac tissue that data assimilation can be used to recover state variables observed sparsely in space and time or unobserved. Here we extend these findings to show that data assimilation remains effective in reconstructing cardiac states with underlying system spatial heterogeneity and in the presence of significant model error.
BT - Proceedings of the 5th {International} {Conference} on {Computational} and {Mathematical} {Biomedical} {Engineering} CY - Pittsburgh, PA LA - eng N2 -New insights into the mechanisms underlying cardiac arrhythmias may be possible from a greater understanding of variables unobserved in experimental settings. When experimental observations are sparse, data assimilation can be effective at recovering a more complete system state estimate. We have recently shown using synthetic observations of one- and three-dimensional cardiac tissue that data assimilation can be used to recover state variables observed sparsely in space and time or unobserved. Here we extend these findings to show that data assimilation remains effective in reconstructing cardiac states with underlying system spatial heterogeneity and in the presence of significant model error.
PP - Pittsburgh, PA PY - 2017 T2 - Proceedings of the 5th {International} {Conference} on {Computational} and {Mathematical} {Biomedical} {Engineering} TI - Data assimilation for cardiac electrical dynamics ER -