Toward Unsupervised Classification of Calcified Arterial Lesions
There is growing evidence that calcified arterial deposits play a crucial role in the pathogenesis of cardiovascular disease. This paper investigates the challenging problem of unsupervised calcified lesion classification. We propose an algorithm, US-CALC (UnSupervised Calcified Arterial Lesion Classification), that discriminates arterial lesions from non-arterial lesions. The proposed method first mines the characteristics of calcified lesions using a novel optimization criterion and then identifies a subset of lesion features which is optimal for classification. Second, a two stage clustering is deployed to discriminate between arterial and non-arterial lesions. A histogram intersection distance measure is incorporated to determine cluster proximity. The clustering hierarchies are carefully validated and the final clusters are determined by a new intra-cluster compactness measure. Experimental results indicate an average accuracy of approximately 80% on a database of electron beam CT heart scans.
KeywordsFeature Space Lesion Feature Arterial Lesion Cophenetic Correlation Cluster Step
- 1.Rumberger, J., Kaufman, L.: A rosetta stone for coronary calcium risk stratification: agatston, volume, and mass scores in 11,490 individuals. AJR Am. J. Roentgenol. 181(3), 743–748 (2003)Google Scholar
- 3.Tiwari, P., Madabhushi, A., Rosen, M.: A hierarchical unsupervised spectral clustering scheme for detection of prostate cancer from magnetic resonance spectroscopy MRS. In: Ayache, N., Ourselin, S., Maeder, A. (eds.) MICCAI 2007, Part II. LNCS, vol. 4792, pp. 278–286. Springer, Heidelberg (2007)CrossRefGoogle Scholar
- 7.Laws, K.: Rapid texture identification. In: Proc. SPIE Conference on Missile Guidance, vol. 238, pp. 376–380 (1980)Google Scholar
- 8.Hopkins, B.: A new method for determining the type of distribution of plan-individuals. Annals of Botany 18, 213–226 (1954)Google Scholar
- 11.Lewis, D., Gale, W.: A sequential algorithm for training text classifiers. In: SIGIR 1994: Proc. 17th ACM SIGIR Conf. on Research and Development in Information Retrieval, pp. 3–12. Springer, New York (1994)Google Scholar