Skip to main content Skip to local navigation
John K. Tsotsos
Home » The Alven System

The Alven System

ALVEN takes as input a sequence of X-ray images depicting a human heart beat.  In each case, the subject has undergone coronary bypass surgery, and during surgery, has had tiny tantalum markers implanted into the left ventricular wall so that they may be imaged afterwards. The goal is to track and assess quantitatively the performance of the left ventricle post-operatively.

The ALVEN takes measurements along each of the  dimensions shown in the figure at the right, for  each of the images of a sequence.   An example of an actual X-ray image from one of the test sequences, along with marker finding is on the left.

Image 1 Image 2 Image 3

The only parts of this figure that  correspond to visual landmarks that ALVEN finds in each image  are the small black rectangles seen along the outline of the left ventricle. The final paths of the markers are seen in the middle figure above for (a) the inwards (contraction) phase and (b)  the outwards (expansion) phase of the left ventricular cycle.

ALVEN’s knowledge base contain definitions for a normal plus 10 different anomalous left ventricular cycles (examples can be found in the papers cited below). The heart motions are defined in terms of general motion concepts (such as contract, translate, move inwards, and so forth) whose definitions also reside in the knowledge base. Each concept is represented using a Minsky-like frame complete with similarity links to related concepts. The overall knowledge base contains over 250 frames.

The final output of ALVEN consists of two components: the first is a detailed textual description  of all of the motions, anomalies, and parameter values that the system has detected. The second is a graphical  summary, shown below, that gives the ejection fraction (EF), the % shortening of each of the radial dimensions measured,  a timeline in comparison to the ‘textbook’ time scale, lines showing the relative paths of each marker, and shading depicting the direction and quality of segmental motion.

The key to the shading is: 

  • blank region – no motion;
  • black region – outwards motion;
  • vertical lines – inwards motion;
  • light dots – severe hypokinesis;
  • medium dots – hypokinetic outwards motion;
  • heavy dots – hypokinetic inwards motion.

One way to think of this graph is to imagine the left ventricular outline, opened up, and laid out along the vertical axis. Then, during the course of a left ventricular cycle, the motion of the markers is tracked horizontally in time.

References

  1. Tsotsos, J.K., Shibahara, T., “Knowledge Organization and its Role in Temporal and Causal Signal Understanding: The ALVEN and CAA Projects”, in Expert System Applications, ed. by L. Bolc and M. Coombs, Springer-Verlag, pp. 429 – 468, 1988.
  2. Tsotsos, J.K., “Artificial Intelligence Techniques in the Analysis of Time-Varying Cardiac Events”, PACE (Pacing and Cardiac     Electrophysiology) , November 1988.
  3. Tsotsos, J.K., Shibahara, T., “The Role of Knowledge Organization for Temporal and Causal Reasoning: The ALVEN and CAA Projects”, in The Knowledge Frontier: Essays in the Representation of Knowledge, pp. 221 – 261, ed. by N. Cercone and G. McCalla, Springer-Verlag, 1987.
  4. Tsotsos, J.K., “The Computer Assessment of Left Ventricular Wall Motion: The ALVEN Expert System”, Computers and Biomedical Research 18, p254 – 277, 1985.
  5. Tsotsos, J.K., “The Role of Knowledge Organization in Representation and Interpretation  of Time-Varying Data: The ALVEN System”, Computational Intelligence, Vol. 1, No. 1., Feb. 1985, p16 – 32.
  6. Tsotsos, J.K., Covvey, H.D., Mylopoulos, J., McLaughlin, P., “The Role of Symbolic Processing in the Computer Evaluation of Left Ventricular Wall Motion: The ALVEN System”, in Ventricular Wall Motion, pp. 315 – 325, ed. by U. Sigwart and P. Heintzen, Georg Thieme Verlag, 1984.
  7. Mylopoulos, J., Shibahara, T., Tsotsos, J., “Towards a Technology for Building Knowledge Based Systems: The PSN Experience”, IEEE Computer, Vol. 16, No. 10, October, 1983, p83 – 89.
  8. Tsotsos, J.K., “Medical Knowledge and its Representation: Problems and Perspectives”, Proc. IEEE MEDCOMP’83, Glouster, Ohio, Sept. 1983.
  9. Tsotsos,J., Mylopoulos,J., Covvey,H.D., Zucker,S.W., “A Framework for Visual Motion Understanding”, IEEE Pattern Analysis and Machine Intelligence, “Special Issue on Computer Analysis of Time-Varying Imagery”, Nov. 1980, p563 – 573.
  10. Tsotsos, J.K., “Can the Computer Ever take over the Practice of Medicine?”, Canadian Medical Association Journal 122(9), p993, 1980.
  11. Tsotsos, J.K., Covvey, H., Mylopoulos, J., McLaughlin, P., “The Role of Symbolic Processing in the Computer Evaluation of Left Ventricular Wall Motion: The ALVEN System”, Proc. International Symposium on Left Ventricular Function, May 1982, Laussane, Switzerland.
  12. Tsotsos, J.K., “Temporal Event Recognition: An Application to Left Ventricular Performance”, Proc. International Joint Conference on Artificial Intelligence, Vancouver, B.C., Aug. 1981, p900-907
  13. Tsotsos, J.K., “On Classifying Time-Varying Events”, Proceedings on Pattern Recognition and Image Processing, Dallas, Aug. 1981, p193 – 199
  14. Tsotsos, J., Druck, M., Burns, R., Covvey, D., Mylopoulos, J., Weisel, R., Pym, J., Bar-Shlomo, B., McLaughlin, P., “Evaluation of Left Ventricular Performance Using Myocardial Tantalum Marker Implants”, Annals of the Royal College of Physicians and Surgeons of Canada, Vol. 14, No. 3, June 1981, p. 167.
  15. Tsotsos, J.K., Covvey, H.D., Mylopoulos, J., McLaughlin, P., “ALVEN: A System for the Evaluation of LV Performance from Myocardial Implants”, Proceedings World Association on Medical Informatics, Strasbourg, April 1981, p36-41.
  16. Covvey, H., McLaughlin, P., Tsotsos, J., Ridsdale, G., Wigle, E.D., “An Image Processing System for Cardiovascular Wall Motion Studies”, Proc. 4th Symposium on Computer Applications in Medical Care, Washington, Nov. 1980, p1154-1157.
  17. Tsotsos, J.K., Covvey, H.D., Mylopoulos, J., “A Representational Formalism for Left Ventricular Wall Motion Studies”, Proc. IEEE Computers in Cardiology, Williamsburg, VA, Oct. 1980, p451-454.
  18. Tsotsos, J., Mylopoulos, J., Covvey, H.D., Zucker, S.W., “ALVEN: A Study on Motion Understanding by Computer”, Proc. International Joint Conference on Artificial Intelligence, Tokyo, Aug. 1979, p890-892.
  19. Tsotsos,J.K., Covvey,H.D., Mylopoulos,J., Zucker, S.W., Wigle, E.D., “ALVEN: A System for LV Segmental Wall Motion Analysis”, Proc. of IEEE Computers in Cardiology, Geneva, Sept. 1979, p445-446.
  20. Tsotsos,J., Mylopoulos,J., Covvey,H.D., Zucker,S.W., “A Framework for Visual Motion Understanding”, Proc. Workshop on Computer Analysis of Time-Varying Imagery”, Philadelphia, 1979.
  21. Tsotsos,J., Covvey,H.D., Mylopoulos,J., Wigle,E.D., “Gross and Segmental Wall Motion Analysis in Cardiac Imagery”, Proc. IEEE Computers in Cardiology, Stanford University, Sept. 1978.
  22. Tsotsos,J., Covvey,H.D., Mylopoulos,J., Wigle,E.D., “Gross and Segmental Motion Analysis in Dynamic Cardiac Imagery”, Proc. Clinical Research Society of Toronto, 1978.
  23. Tsotsos,J., Covvey,H.D., Mylopoulos,J., Wigle,E.D., “Gross and Segmental Wall Motion Analysis in Cardiac Imagery”, Proc. 2nd Symposium on Computer Applications in Medical Care, Washington, DC, Nov. 1978, p45-48.
  24. Tsotsos,J., Baecker,R., Covvey,H.D., Reeves,W., Mylopoulos,J., Wigle,E.D., “An Interactive Knowledge-Based Systems Approach to Cardiac Image Description and Analysis”, Proc. IEEE Computers in Cardiology, Erasmus University, Rotterdam, Oct. 1977, p377-384.
  25. Tsotsos,J., “Knowledge-Base Driven Analysis of Cinecardioangiograms”, Proc. Fifth International Joint Conference on Artificial Intelligence, MIT, Cambridge, Mass., Aug. 1977, p699.
  26. Tsotsos,J., “Some Notes on Motion Understanding”, Proc. Fifth International Joint Conference on Artificial Intelligence, MIT, Cambridge, Mass.,  Aug. 1977, p611.