Abstract
We are presenting a method to obtain diagnosis and prognosis information by searching similar images into a large database of Myocardial Perfusion Scintigraphy (MPS) cases for which diagnosis is known. We are applying similarity measures to cardiac images pre-registered with a template. Our database is composed of 1430 patient cases with associated clinical information. For each new case, we sort all the patients of the database from most to less similar ones and compute a severity criterion, based on a statistical analysis of normal and diseased most similar patients. By varying a threshold on the severity criterion and testing the classification of controlled cases, we have measured the operational characteristic of this test (ROC curves), and shown increased performance in sensitivity and specificity for disease detection with respect to clinicians and to experts in consensus. Through the extension of database to patients’ outcome information, we expect to extend this method to prognosis.
Chapter PDF
Similar content being viewed by others
References
Declerck, J., Feldmar, J., Goris, M.L., Betting, F.: Automatic Registration and Alignment on a Template of Cardiac Stress & Rest SPECT Images. Mathematical Methods in Biomedical Image Analysis (June1996)
Goris, M.L., Pace, W., Petersen, M., Kwan, A.: The Effect of Lesion Modulation on the Operating Characteristics of Myocardial Perfusion Studies. In: EANM 1998, Berlin, Germany (August 1998)
Goris, M.L., Pace, W.M., Thirion, J.-P., Hotz, B., Similon, P.: Case specific reference images for patient and acquisition variables in Myocardial Perfusion Analysis. In: EANM 1999, Barcelona, Spain, September 1999, vol. 26(9) (1999)
Guimond, A., Subsol, G., Thirion, J.-P.: Automatic MRI Database Exploration and Applications. Int. J. on Pattern Recognition and Artificial Intelligence (IJPRAI) 11(8), 1345–1365 (1997)
Diamond, G.A., Forrester, J.S.: Analysis of Probability as an Aid in the Clinical Diagnosis of Coronary-Artery Disease. The New England Journal of Medicine 300(24), 1350–1358 (1979)
Venot, A., Lebruchec, J.F., Golmard, J.L., Roucayrol, J.C.: An Automated Method for the Normalization of Scintigraphic Images. The Journal of Nuclear Medicine 24, 529–531 (1983)
Nastar, C., Mitschke, M., Meilhac, C.: Efficient query refinement for image retrieval. In: Computer Vision and Pattern Recognition (CVPR 1998), Santa Barbara (June 1998)
Aksoy, S., Haralick, R.: Textural features for image database retrieval. In: Proceedings of CVPR, Workshop on Content-Based Access of Image and Video Libraries (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Hotz, B., Thirion, JP. (2000). HeartPerfect: Data Mining in a Large Database of Myocardial Perfusion Scintigraphy. In: Delp, S.L., DiGoia, A.M., Jaramaz, B. (eds) Medical Image Computing and Computer-Assisted Intervention – MICCAI 2000. MICCAI 2000. Lecture Notes in Computer Science, vol 1935. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-40899-4_37
Download citation
DOI: https://doi.org/10.1007/978-3-540-40899-4_37
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41189-5
Online ISBN: 978-3-540-40899-4
eBook Packages: Springer Book Archive