High-throughput data dimension reduction via seeded canonical correlation analysis

Canonical correlation analysis (CCA) is one of popular statistical methodologies in multivariate analysis, especially, in studying relation of two sets of variables. However, if sample sizes are smaller than the maximum of the dimensions of two sets of variables, it is not plausible to construct canonical coefficient matrices due to failure of inverting sample covariance matrices. In this article, we develop a two step procedure of CCA implemented in such situation. For this, seeded dimension reduction is adapted into CCA. Numerical studies confirm the approach, and two real data analyses are presented. Copyright (c) 2014 John Wiley & Sons, Ltd. This article focuses on the results of finite element method simulations, aimed at assessing the mechanical behavior of a composite metal/polymer Nuss bar to correct chest wall deformities, such as Pectus Excavatum. Two different geometries (sheets ad rods) for the metal elements to be embedded in the polymeric matrix were tested, and different metal and polymeric materials were considered. The study allowed to identify the optimal configuration for a composite Nuss bar, showing good promises for future clinical applications.