Using functional PCA for cardiac motion exploration Denis Clot   View Paper (PDF)   Return to Main-Track Regular Papers Abstract Principal component analysis (PCA) (14, 6) is a main tool in multivariate data analysis. Its paradigms are also used in the Karhunen-Loeve decomposition (5), a standard tool in image processing. Extensions of PCA to the framework of functional data have been proposed. The analy-sis provided by the functional PCA seems to be a powerful tool to find principal sources of variability in curves or images, but it fails in providing us with easy interpretations in the case of multifunctional data. Guide lines aiming at spot information from the outputs of PCA applied to functionals with values in space of continuous functions upon a bounded domain are proposed. An application to cardiac motion analysis illustrates the complexity of the multi-functional framework and the results provided by functional PCA. DiSC'03 © 2003 Association for Computing Machinery