The Analysis and Applications of Optical Flow in Cardiac Functions

• Main Authors: Chou, Hsiao-Hsuan, 周曉璇
• Other Authors: Chen, Jyh-Horng
• Format: Others
• Language: zh-TW
• Published: 1996
• Online Access: http://ndltd.ncl.edu.tw/handle/57482256576176396762

碩士 === 國立臺灣大學 === 電機工程研究所 === 84 === The motion of the heart is very complicated. Understanding the wall motion of the left ventricle is especially important for analysizing the cardiac functions. Due to the development of imaging technology, it beomes easier to get the dynamic cardiac image sequence. If we can get the correct velocity information from the dynamic images, it will be helpful for the cardiac diagnosis and the heart function evalution. Optical flow is the velocity distribution of the patterns in image. With the optical flow method, we can acquire the magnitude and the direction of the velocity of each pixel in an image.We apply this method on dynamic Cardiac Magnetic Resonance Images (MRI) and Radionuclide Angiography images (RNA) , focusing on the wall motion of the left ventricle. From the optical flow we study, the complex motion pattern during the diastole and systole of a heart cycle can be observed and analyzed. In order to increase the precision, we use anisotropic diffusion and gaussian filters in space and time domain for smoothing the gray-levels in advance. A median filter is than applied on the velocity field . The velocity fields calculated from the optical flow method show many details of the heart motion. By tracing the locations of the points we interested, we can get useful physiological parameters, eg. ejection fraction(EF), peak filling rate(PFR), etc. Moreover, by calaulating the strain rate from the velocity field, one can tell the differences between the healthy and abnormal cardiac functions. Other mechanical parameters of the heart could be also derived displayed using this method noninvasively, which is not possible with traditional imaging techniques.