Dynamic contrast-enhanced MRI diagnostics in oncology via principal component analysis

摘要

This paper proposes the use of latent variable models based on principal component analysis (PCA) as a robust alternative to pharmacokinetic modeling for the analysis of the dynamic contrast-enhanced magnetic resonance images (DCE-MRI) often obtained during oncological studies. Theoretical and practical justifications are provided for the advantages of the PCA approach. A pilot study using clinical DCE-MRI data on prostate patients is presented to demonstrate the method.