Model-based image processing of transmitral Doppler velocity profiles: automated determination of E-wave data set start point

The authors have previously reported a model-based image processing method for quantification of transmitral Doppler velocity profiles (DVP), for the quantitative assessment of left ventricular diastolic function. For a given image, the process of fitting the model to the data has been shown to be sensitive to the choice of the data set to be modeled. The authors analyzed Doppler E-waves (early, rapid filling phase of diastole) having high acceleration and deceleration rates which are particularly sensitive to the E-wave data set start position. Examination of the acceleration portion of the time-averaged maximum velocity envelope (MVE) of the clinical DVP showed a change-of-sign in the second derivative of the MVE (an inflection point) not present in the parameterized diastolic filling (PDF) formalism used to model the E-wave. The inflection point defines a consistent choice for the E-wave data set start position. The existence of a local minimum in the estimated standard deviation of the model parameters as a function of the E-wave data set start position validates the use of the inflection point as the data set start position.