Johan H. C. Reiber

Comparison of accuracy and precision of quantitative coronary arterial analysis between cinefilm and digital systems

In this chapter we have compared the accuracy and precision of two state-of-the-art analytical software packages for quantitative coronary arteriography (QCA) developed in our laboratory. The two packages are the QCA package (Versions 1.11 and 1.2) on the film-based Cardiovascular Measurement System (CMS), and the ACA package (Version 1.0) on the Philips Digital Cardiac Imaging System (DCI). In these studies the accuracy is defined as the systematic error described in terms of the mean signed difference between actual and measured values of phantoms or catheters, or between the values from repeated measurements; ideally the systematic errors should be as small as possible. The precision or random error is described in terms of the standard deviation of these signed differences; ideally these random errors should be as small as possible as well. In this chapter four comparative studies have been described.

State of the Art in Quantitative Coronary Arteriography as of 1996

In this chapter the important developments which have led to the third generation in quantitative coronary arteriographic (QCA) analytical software are presented, as well as current developments in the fields of image compression and storage. The conventional QCA approaches with automated contour detection techniques based on Minimum Cost contour detection Algorithms (MCA) have been well established and validated. However, further improvements in the calculations of the diameter and reference diameter functions were needed, especially for complex morphology and for stent applications. The development of the Gradient Field Transform (GFTR) approach for the quantitation of complex lesions represents a major step forward in QCA. With the advent of the cineless catheterization laboratory, the issue of image compression has become of major relevance. Phantom studies with lossy JPEG image compression at 5122 matrix size demonstrate that the compression factor (CF) should not exceed the level of 10. On the other hand, if JPEG and LOT lossy compression schemes (CF’s of 5,8 and 12) are applied to routinely acquired coronary angiographic image results, QCA measurements demonstrate that all three compression factors lead to significantly increased random differences in the measurements. These results suggest that even the JPEG and LOT compression ratio of 5 is not acceptable for QCA. Finally, an extensive QCA study has demonstrated that S-VHS video tape is unacceptable for QCA and should be excluded from quantitative angiographic clinical trials.

Quantitative coronary ultrasound: state of the art

Intravascular Ultrasound (IVUS) provides real-time high resolution images of the arterial wall. By performing a three-dimensional reconstruction, it permits an advanced assessment of the vessel, lumen and wall morphology. Recently, the straight stacking of the IVUS images has been extended by a geometrically correct orientation of the images in 3D space, using biplane angiographic images. Quantification of IVUS images, both in 2D and 3D, requires segmentation of the images. Automated segmentation of IVUS images for quantitative analysis reduces the required time and the subjectivity of boundary tracing. Different segmentation approaches for 2D and 3D IVUS are discussed, including the commercially available packages for analysis of IVUS images. Furthermore different approaches for the 3D reconstruction including the use of biplane angiographic images are discussed. This chapters finishes with a discussion about the future directions of IVUS including the developments in the area of RF-data analysis and the developments of new devices.

Overview of automated quantitation techniques in 2D echocardiography

Many methods for automated quantitation in two-dimensional echocardiography have been published, but few have gained practical importance. This chapter describes the problems and pitfalls of border detection in cardiac ultrasound, gives an overview of methods described in the literature and categorizes the applied techniques into a hierarchy of abstraction levels. Furthermore, a practical system for automated border detection (ECHO-CMS) and its evaluation will be discussed, and the chapter will be concluded with an overview of the general developments anticipated for the near future.

Lipid-lowering therapy and progression of coronary atherosclerosis

An overview of all major angiographic lipid intervention trials. Basic mechanisms and methodological aspects, including biochemical as well as angiographic aspects, are discussed. A comparison of all available data permits an analysis to be made of what may currently be considered proved, which aspects merit further investigation, and which hypotheses should be rejected.

Issues in the performance of quantitative coronary angiography in clinical research trials

Quantitative coronary analysis remains the classical and most commonly used tool to assess the results of coronary pharmacological or mechanical intervention. The authors review the methodology of this type of analysis, highlighting the pros and cons of previous recommendations. Special interest has been devoted to catheter calibration and the choice of angiographic views for optimal measurement and reliability. Significant additional information in terms of acute gain, late loss or restenosis rate is not gained by the use of averaged orthogonal measurements as compared to the more simple single view approach. Also calibration procedures can be simplified by using tables of mean measured values for various types of catheters instead of measuring each catheter with a precision micrometer and by doing calibration measurements on contrast filled instead of flushed catheters. Moreover, specific in vitro and in vivo criteria have been proposed and tested for acceptance of catheter type and size in QCA analyses. Based on these criteria only catheters of sufficient size (6F or greater) and approved for QCA are recommended.

Postmortem validation of the automated coronary analyis (ACA) software package

Goal of this study was to compare the quantitative coronary arteriographic (QCA) results obtained with the Philips DCI/ACA analytical software package with those from postmortem casts in an animal experimental setting. Standard digital coronary arteriograms were obtained from 6 mongrel dogs. After the imaging procedure, the dogs were sacrificed and casts were made of the coronary trees by filling the vessels with a mixture of radio-opaque barium and silicone gel at a fixed pressure of 100 mmHg. Vessel diameters were measured from the digital arteriograms at a total of 118 selected locations with the ACA package. Thin slices were cut from the casts at these same measurement locations and the areas of the cross sections were obtained by manual tracing of the outline of each slice in an approximately 40 × magnified image. From these cross-sectional areas, cast diameters were derived using the formula for circular cross-sections. Cast diameters ranged in size from 0.69 to 3.30 mm. The systematic error between the measurements was found to be 0.058 mm; (p < 0.015) and the standard deviation of the signed difference 0.255 mm; the correlation coefficient was r = 0.91. The largest error sources are supposed to be the slight differences in the selection of identical positions in the X-ray images and on the casts, and the ‘out-of-plane’ magnification for a number of vessel locations. This postmortem study demonstrates that the diameters of coronary vessels can be measured from digital arteriograms with the DCI/ACA package with a high degree of accuracy and precision.

Current and future developments in QCA and image fusion with IVUS

Although quantitative coronary arteriography (QCA) has been around now for quite some time, research and development continue to take place along several directions. First of all, the imaging medium has changed from the traditional 35 mm analog cinefilm to the digital world with the CD-R as the preferred carrier. This required adaptations of the basic contour detection algorithms. Stimulated by these same changes, digital review stations or DICOM-Viewers have been developed. In addition, third-generation QCA algorithms have been designed and implemented, and applied to quantitate complex morphology and radiopaque stents.

Automated labeling of coronary arterial tree segments in angiographic projection data

The fully automated reporting of the extent of disease from coronary arteriograms is likely to be a four-step procedure: (1) segmentation of the center lines of the coronary tree from one or more angiographic projection(s); (2) detection of the arterial boundaries based on the center lines; (3) automated detection of possible narrowings in the coronary segments, and (4) identification of the coronary arteries. In this paper we will concentrate on the developments of techniques for the last step, the automated identification of coronary arterial segments. Our approach is based on the representation of the projection of the coronary tree by a graph so that graph matching techniques can be used for labeling of the coronary arterial segments. The coronary tree, which is a branching structure with possible crossings and overlaps as visualized in an angiographic projection, is assumed to be segmented more or less successfully from an angiogram with known projection geometry. A model graph is composed from the projection of a three-dimensional representation of the normal coronary tree, and matched with the data graph by inexact graph matching. Two types of graph representations will be discussed. In the first type arterial branching points are represented by nodes and arterial segments by arcs between nodes. In the second type the arterial segments are represented by nodes and relationships between the arteries by arcs. Nodes and arcs in both types of representation are attributed with a semantic vector of object (node) features or relational (arc) features, which is a mixture in the first type of representation. Both types have been implemented and we are currently in the process of determining optimal parameter values for the associated cost functions.

Lessons learned from angiographic coronary atherosclerosis trials

From observational studies, we may conclude that progression and regression of coronary atherosclerosis is still a highly unpredictable process. Medical intervention studies have demonstrated that lipid lowering in general, and administration of HMG-CoA reductase inhibitors in particular, retards progression and promotes regression of coronary atherosclerosis and diminishes subsequent clinical events, even when cholesterol levels are not strongly elevated. Risk factor modification-changes in lifestyle, ileal bypass surgery and low-density-cholesterol apheresis also have their merits in reducing progression, whereas the definite place of calcium channel blockers in retarding established coronary atherosclerosis yet has to be determined.