Date of Award

Spring 5-15-2015

Author's School

Graduate School of Arts and Sciences

Author's Department


Degree Name

Doctor of Philosophy (PhD)

Degree Type



Multiple modalities are routinely used in clinical cardiology to determine cardiovascular function, and many of the indexes derived from these modalities are causally interconnected. A correlative approach to cardiovascular function however, where indexes are correlated to disease presence and progression, fails to fully capitalize on the information content of the indexes.

Causal quantitative modeling of cardiovascular physiology on the other hand offers a predictive rather than accommodative approach to cardiovascular function determination. In this work we apply a kinematic modeling approach to understanding diastolic function. We discuss novel insights related to the physiological determinants of diastolic function, and define novel causal indexes of diastolic function that go beyond the limitations of current established clinical indexes. Diastolic function is typically characterized by physiologists and cardiologists as being determined by the interplay between chamber stiffness, chamber relaxation/viscoelasticity, and chamber filling volume or load. In this work we provide kinematic modeling based analysis of each of these clinical diastolic function determinants.

Considering the kinematic elastic (stiffness) components of filling, we argue for the universality of diastolic suction and define a novel in-vivo equilibrium volume. Application of this novel equilibrium volume in the clinical setting results in a novel approach to determination of global chamber stiffness.

Considering the viscoelastic components of filling, we demonstrate the limitations associated with ignoring viscoelastic effects, an assumption often made in the clinical setting. We extend the viscoelastic component of filling into the invasive hemodynamic domain, and demonstrate the causal link between invasively recorded LV pressure and noninvasively recorded transmitral flow by describing a method for extracting flow contours from pressure signals alone.

Finally, in considering load, we solve the problem of load dependence in diastolic function analysis. Indeed all traditional clinical indexes of diastolic function are load dependent, and therefore are imperfect indexes of intrinsic diastolic function. Applying kinematic modeling, we derive a load independent index of diastolic function. Validation involves showing that the index is indeed load-independent and can differentiate between control and diastolic dysfunction states. We apply this novel analysis to derive surrogates for filling pressure, and generalize the kinematic modeling approach to the analysis of isovolumic relaxation.

To aid widespread adoption of the load independent index, we derive and validate simplified expressions for model-based physiological parameters of diastolic function.

Our goal is to provide a causal approach to cardiovascular function analysis based on how things move, to explain prior phenomenological observations of others under a single causal paradigm, to discover `new physiology', facilitate the discovery of more robust indexes of cardiovascular function, and provide a means for widespread adoption of the kinematic modeling approach suitable for the general clinical setting.


English (en)

Chair and Committee

Sándor J Kovács

Committee Members

Anders E Carlsson, Guy M Genin, Mark R Holland, James G Miller, Samuel A Wickline


Permanent URL:

Included in

Physics Commons