Coronary heart disease is the fatal cardiovascular disease, as well as the leading killer threatening people’s health. Coronary atherosclerosis is the main cause of coronary heart disease, the accurate identification of the coronary atherosclerotic plaques is of great importance for the judgement of the pathological conditions of the vascular and the guidance of subsequent treatment. Due to its extremely high imaging resolution, intravascular optical coherence tomography (IVOCT) has been widely used in the clinical diagnosis and treatment of coronary heart disease, of which one of the important functions is to judge the type of plaque in the diseased vessels. The purpose of this paper is to study an algorithm of the IVOCT plague image automatic recognition, which assists the doctors to analyze the images, so as to improve the accuracy of the plaque recognition.
Optical coherence tomography (OCT) is a new medical imaging technology that developed at the end of the 20th century after X-ray, computed tomography (CT), magnetic resonance imaging (MRI), and intravascular ultrasound (IVUS). It is called “optical biopsy” technology with the advantages of no radiation, simple structure and high resolution that can reach ten times that of IVUS. However, OCT also has the disadvantage of insufficient depth of detection that only a few millimeters and imaging speed. Even so, OCT can be used in combination with microscopes, medical catheters and endoscopes; therefore, it has broad application prospects in the field of biomedicine. The OCT system is simple in structure, mainly Michelson interferometer. Using the principle of optical coherence imaging, it detects the back-reflecting or scattering signals of incident light at different depths of biological tissue to obtain the surface and subsurface imaging of transparent or opaque substances. The combination of OCT and endoscopy extends the use of OCT to the diagnosis of cardiovascular diseases, which is called intravascular optical coherence tomography (IVOCT). It enables rapid visualization of microscopic images of vascular cross sections and is a powerful tool for clinical detection of coronary atherosclerosis, in which coronary artery calcification is a common problem in the clinic and is closely related to cardiovascular diseases. This review will briefly introduce the principle of OCT technology and its application in cardiovascular diseases, and focus on the research progress of detection of coronary artery calcification based on OCT technology.
The segmentation of the intracoronary optical coherence tomography (IVOCT) images is the basis of the plaque assessment. Calcified plaque is one of the main thrombus plaques. Accurate segmentation of calcified plaque is important to the plaque feature analysis, vulnerable plaque recognition and further coronary disease diagnosis. Based on the knowledge of imaging processing, the inner boundary of calcified plaques is clear, but the outer boundary is hard to identify because of the weak edge. This paper proposed an algorithm about calcified plaque segmentation for IVOCT. Taking the segmented vessel wall by using dynamic threshold as the region of interest, the location of calcified plaque was determined by K-means clustering to obtain the inner edge. The Local Binary Fitting (LBF) active contour model is used to solve the problem of weak edge to clarify the outer edge. Then the distribution of superficial calcification can be evaluated. Ten coronary images with typical plaques from 3 patients in our experiment were used to taking the segmentation. The processing results were compared with the clinician manual segmentation. It is indicated that the proposed algorithm could segment the plaque regions accurately. This work hopefully can be used for automatic processing the serials of IVOCT images to reduce subjectivity and divergence between different clinician and contribute to the diagnosis and treatment of coronary artery disease from IVOCT.