Left Ventricular Shape Analysis and Cardiac Wall Motion Estimation with Cardiac Imaging

• Main Authors: Jeng-Ren Duann, 段正仁
• Other Authors: Jenn-Lung Su
• Format: Others
• Language: en_US
• Published: 1999
• Online Access: http://ndltd.ncl.edu.tw/handle/83489540635520187137

博士 === 中原大學 === 物理學系 === 87 === The purpose of this dissertation is mainly to derive a new set of parameters and methods to evaluate the functionality and geometry of left ventricle (LV). They include 2D/3D cardiac shape analysis, MR color kinesis, and measurement of conventionally used cardiac indexes (e.g., capacity of left ventricular chamber at both end-diastolic and end-systolic phases, EDV, ESV; stroke volume, SV; cardiac output, CO; and left ventricular mass). In the study of 3D cardiac shape analysis, 3-dimensional wire-frame LV model is constructed and used to estimate several shape descriptors. Using these shape descriptors, the geometry of LV endocardial surface can be well described. For the reason of being hands-on method, 2D cardiac shape analysis is also included in this dissertation to evaluate the cardiac shape in 2-dimension. A new analytic method based on 2D-curvature calculation is developed to serve as an index of smoothness of a specified surface line obtained from 2D echocardiogram. Such a method transforms the calculated 2D curvature of the selected surface line to the spectrogram with short-term Fourier transformation (STFT). The harmonics and power of lower frequency band contained in the spectrogram constitute the analytical factors of 2D cardiac shape analysis. The method of MR color kinesis (MRCK) uses the stack of contour obtained from multiphase MR cardiac images through ejection. Once the contour of each time frame is achieved, they are color encoded according to the time of occurrence and overlaid together. This color encoded overlaid picture is called the MRCK image. It can be used to evaluate the excursion of endocardial wall and thus assess the cardiac wall motion abnormalities. These parameters or methods developed in this dissertation are all applied to the measurement of various cardiac diseases. They are proved feasible to applied to the daily clinics.