An expert system for the labeling and 3D reconstruction of the coronary arteries from two projections

  • Carl Smets
  • F. van de Werf
  • P. Suetens
  • A. Oosterlinck


In this paper we present a rule-based expert system for the automatic delineation and 3D reconstruction of the left coronary artery on standard RAO and LAO angiographic projections. The approach is based on the application of a general blood vessel model and on anatomical models which take into account the normal variations of the coronary artery structure. In a first step, the arteries are delineated by detecting the maximum intensity on the centerline of the vessels. Then, we label the blood vessel segments according to an anatomical model of the left coronary artery. In general, only 1–2 labels remain for each blood vessel segment. Finally, these results are used for an automatic 3D reconstruction of the left coronary artery from two projections. Results from clinical RAO and LAO angiograms will be presented.

Key words

quantitative coronary angiography knowledge-based systems expert systems three-dimensional reconstruction 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Reiber JHC. Morphologic and densitometric quantification of coronary stenoses; an overview of existing quantitation techniques. In: JHC Reiber, PW Serruys, eds. New developments in Quantitative Coronary Arteriography. Kluwer Academic Publishers, Dordrecht, 1988: 34–88.Google Scholar
  2. 2.
    Tsuji S, Nakano H. Knowledge-based identification of artery branches in cine-angiograms: an image understanding system which utilizes production-type knowledge. In: T Kitagawa, ed. Computer Science and Technologies. North-Holland, 1982: 311–321.Google Scholar
  3. 3.
    Ritchings RT, Colchester ACF, Wang HQ. Knowledge-based Analysis of Carotid Angiograms. Image and Vision Computing 1985; 3: 217–222.CrossRefGoogle Scholar
  4. 4.
    Wang HQ, Ritchings RT, Colchester ACF. Image Understanding System for Carotid Angiograms. Image and Vision Computing 1987; 5: 79–83.CrossRefGoogle Scholar
  5. 5.
    Elion JL, Nissen SE. A knowledge-based image processing system for the interpretation of coronary arteriograms. Proceedings of the SPIE 1987; 767: 428–432.Google Scholar
  6. 6.
    Catros JY, Mischler D. An artificial intelligence approach for medical picture analysis. Pattern Recognition Letters 1988; 8: 123–130.CrossRefGoogle Scholar
  7. 7.
    Stansfield SA. ANGY: A rule-based Expert System for Automatic Segmentation of Coronary Vessels from Digital Subtracted Angiograms. IEEE Trans Pattern Analysis and Machine Intelligence 1986; PAMI-8: 188–199.Google Scholar
  8. 8.
    Smets C, Verbeeck G, Suetens P, Oosterlinck A. Knowledge-based delineation of coronary angiograms. In: Lacoume JL, Chehikian A, Martin N, Malbos J, eds. Signal Processing IV: Theories and Applications, Elsevier Science Publishers BV: North-Holland, 1988: 1259–1262.Google Scholar
  9. 9.
    Brown BG, Bolson EG, Dodge HT. Dynamic mechanisms in human coronary stenosis. Circ 1984; 70: 917–922.Google Scholar
  10. 10.
    Fischler M, Wolf H. Linear Delineation. Proc IEEE Computer Vision and Pattern Recognition 1983: 351–356.Google Scholar
  11. 11.
    Smets C, Verbeeck G, Suetens P, Oosterlinck A. A knowledge-based system for the delineation of blood vessels on subtraction angiograms. Pattern Recognition Letters 1988; 8: 113–121.CrossRefGoogle Scholar
  12. 12.
    Gensini GG. Coronary Arteriography, Mount Kisco, New York, Futura Publishing, 1975.Google Scholar
  13. 13.
    Mackworth AK. Constraint Satisfaction. In: Shapiro SC, ed. Encyclopedia of Artificial Intelligence. 1987; 1: 205–211.Google Scholar
  14. 14.
    Gerbrands JJ, Backer E, Van Der Hoeven WAG. Quantitative Evaluation of Edge Detection by Dynamic Programming. In: Gelsema ES, Kanal LN, eds. Pattern Recognition in Practice II. Elsevier Science Publishing Company, 1986: 91–100.Google Scholar
  15. 15.
    Parker DL, Pope DL, Van Bree R, Marshall HW. Three-dimensional Reconstruction of Moving Arterial Beds from Digital Subtraction Angiography. Comp Biomed Res 1987; 20: 166–185.CrossRefGoogle Scholar
  16. 16.
    Marr D. Vision: A computational investigation into the human representation and processing of visual information. WH Freeman and Company, San Francisco. 1982.Google Scholar

Copyright information

© Kluwer Academic Publishers 1990

Authors and Affiliations

  • Carl Smets
    • 1
  • F. van de Werf
    • 1
  • P. Suetens
    • 1
  • A. Oosterlinck
    • 1
  1. 1.Department of Cardiology, Interdisciplinary Research Unit for Radiological Imaging (ESAT + Radiology)Katholieke Universiteit LeuvenHeverleeBelgium

Personalised recommendations