Title :
An on-line system for the quantitative analysis of coronary arterial segments
Author :
van der Zwet, P.M.J. ; von Land, C.D. ; Loois, G. ; Gerbrands, J.J. ; Reiber, J.H.C.
Author_Institution :
Lab. for Clinical & Exp. Image Process., Erasmus Univ., Rotterdam, Netherlands
Abstract :
The online quantification of coronary arteries is important for the objective and accurate assessment of the effects of coronary intervention procedures. The basic principles of a quantitative coronary arteriography (QCA) software package, to be implemented on the Philips DCI system, is described. First, a centerline of the selected arterial segment is defined. After the beginning and endpoints are manually defined, a modified beam algorithm (MBA) is used to automatically find a path line of this segment. With this path line as a first model, the arterial contours are detected automatically on the basis of the minimum-cost contour detection technique. For accurate results, the image is magnified by a factor of 2 by linear interpolation, and the minimum cost technique is used again, now employing the initially found contours as a model. On the basis of these final contours, a number of relevant clinical parameters are calculated. The QCA software package will be undergoing extensive evaluation studies; the MBA technique has been shown to be very reliable (success score 89.3%)
Keywords :
cardiology; computerised picture processing; medical diagnostic computing; arterial contours; clinical parameters; coronary arterial segments; coronary intervention procedures; linear interpolation; minimum cost technique; modified beam algorithm; online quantification; path line; quantitative coronary arteriography; software package; Angiography; Application software; Costs; Hospitals; Image processing; Image segmentation; Iterative algorithms; Laboratories; Quantum cellular automata; Software packages;
Conference_Titel :
Computers in Cardiology 1989, Proceedings.
Conference_Location :
Jerusalem
Print_ISBN :
0-8186-2114-1
DOI :
10.1109/CIC.1989.130509