Identifying Relevant Eigenimages - a Random Matrix Approach

2008/12/25 by Yu Ding, Yiu-Cho Chung, Ding, Yu +5
Mathematics · Physics and Astronomy · #Data Analysis #Disordered Systems and Neural Networks (cond-mat.dis-nn) #FOS: Computer and information sciences #FOS: Physical sciences #Machine Learning (stat.ML) #Medical Physics (physics.med-ph) #Statistics and Probability (physics.data-an) #cond-mat.dis-nn #physics.data-an #physics.med-ph #stat.ML

paper · pdf · doi:10.48550/arxiv.0812.4618

7 pages, 5 figures

arxiv created 2008/12/25 · arxiv updated 2009/12/01

Abstract

Dimensional reduction of high dimensional data can be achieved by keeping only the relevant eigenmodes after principal component analysis. However, differentiating relevant eigenmodes from the random noise eigenmodes is problematic. A new method based on the random matrix theory and a statistical goodness-of-fit test is proposed in this paper. It is validated by numerical simulations and applied to real-time magnetic resonance cardiac cine images.

Related

• Assessment of Spatial-variant Noise Level in Real-time Dynamic MR Images - a Random Matrix Approach
• Applications of Random Matrix Theory in Machine Learning and Brain Mapping
• Multi-Sensor Fusion via Reduction of Dimensionality
• Random matrix route to image denoising
• A Comparative Analysis of Principal Component Analysis (PCA) and Singular Value Decomposition (SVD) as Dimensionality Reduction Techniques