TY - JOUR
T1 - Gaussian process manifold interpolation for probabilistic atrial activation maps and uncertain conduction velocity
T2 - Gaussian Process Manifold Interpolation
AU - Coveney, Sam
AU - Corrado, Cesare
AU - Roney, Caroline H.
AU - O'Hare, Daniel
AU - Williams, Steven E.
AU - O'Neill, Mark D.
AU - Niederer, Steven A.
AU - Clayton, Richard H.
AU - Oakley, Jeremy E.
AU - Wilkinson, Richard D.
PY - 2020/6/12
Y1 - 2020/6/12
N2 - In patients with atrial fibrillation, local activation time (LAT) maps are routinely used for characterizing patient pathophysiology. The gradient of LAT maps can be used to calculate conduction velocity (CV), which directly relates to material conductivity and may provide an important measure of atrial substrate properties. Including uncertainty in CV calculations would help with interpreting the reliability of these measurements. Here, we build upon a recent insight into reduced-rank Gaussian processes (GPs) to perform probabilistic interpolation of uncertain LAT directly on human atrial manifolds. Our Gaussian process manifold interpolation (GPMI) method accounts for the topology of the atrium, and allows for calculation of statistics for predicted CV. We demonstrate our method on two clinical cases, and perform validation against a simulated ground truth. CV uncertainty depends on data density, wave propagation direction and CV magnitude. GPMI is suitable for probabilistic interpolation of other uncertain quantities on non-Euclidean manifolds. This article is part of the theme issue 'Uncertainty quantification in cardiac and cardiovascular modelling and simulation'.
AB - In patients with atrial fibrillation, local activation time (LAT) maps are routinely used for characterizing patient pathophysiology. The gradient of LAT maps can be used to calculate conduction velocity (CV), which directly relates to material conductivity and may provide an important measure of atrial substrate properties. Including uncertainty in CV calculations would help with interpreting the reliability of these measurements. Here, we build upon a recent insight into reduced-rank Gaussian processes (GPs) to perform probabilistic interpolation of uncertain LAT directly on human atrial manifolds. Our Gaussian process manifold interpolation (GPMI) method accounts for the topology of the atrium, and allows for calculation of statistics for predicted CV. We demonstrate our method on two clinical cases, and perform validation against a simulated ground truth. CV uncertainty depends on data density, wave propagation direction and CV magnitude. GPMI is suitable for probabilistic interpolation of other uncertain quantities on non-Euclidean manifolds. This article is part of the theme issue 'Uncertainty quantification in cardiac and cardiovascular modelling and simulation'.
KW - atrial fibrillation
KW - cardiac conduction velocity
KW - Gaussian process
KW - local activation time
KW - manifold
KW - probabilistic interpolation
UR - http://www.scopus.com/inward/record.url?scp=85085449607&partnerID=8YFLogxK
U2 - 10.1098/rsta.2019.0345
DO - 10.1098/rsta.2019.0345
M3 - Article
C2 - 32448072
AN - SCOPUS:85085449607
SN - 1471-2962
VL - 378
JO - Philosophical transactions. Series A, Mathematical, physical, and engineering sciences
JF - Philosophical transactions. Series A, Mathematical, physical, and engineering sciences
IS - 2173
M1 - 20190345
ER -