Detection Of Cardiac Phase In Intracoronary Ultrasonic Image Sequence

Intracoronary ultrasound (ICUS) is one of interventional imaging modalities widelyused for diagnosing coronary artery diseases in clinic. Coronary arteries cover theepicardial surface and follow the cardiac motion. During the acquisition of ICUS imagesequences through continuous pulling back the ultrasounic catheter, the vascularmorphology changes and the ultrasound catheter moves relative to the lumenaccompanied by periodic cardiac motion. This leads to cyclical changing of the grayscalecharacteristics of ICUS images. Consequently, ICUS image sequences imply phaseinformation of the cardiac cycle.Electrocardiosignal, one of important biological electrical signals, reflects theworking condition and reliable internal characteristics of the heart. It plays importantroles in vascular disease study. Modeling and simulation of electrocardiosignal canprovide basis for medical research, management and teaching.In this dissertation, the problem of detecting cardiac phase information fromcontinuous pullback ICUS image sequence based on the grayscale features of images isaddressed. First, the approximation of average heart rate is estimated from the imagesequence. Second, the cyclical change of grayscale characteristic in each frame isanalyzed to extract a signal reflecting cardiac cycles. Third, the signal is filtered with abandpass filter to remove the components caused by any factors except cardiac dynamics.Finally, the local extrem of the filtered signal is detected to extract the cardiac phase andthe image sequence is gated. The proposed method was validated with clinically acquiredimage data. The performance of various kinds of filters in cardiac phase detection wasdiscussed according to experimental results. Moreover, the accuracy of the method wasquantitatively assessed. Based on the characteristics of the electrocardiosignal wave, thecurrent models and simulation methods of electrocardiosignals were analyzed. Also,the modeling and simulation of electrocardiosignal was preliminarily studyied as afoundation for the successive research.