Sparse representation of complex MRI images. Academic Article

abstract

• Sparse representation of images acquired from Magnet Resonance Imaging (MRI) has several potential applications. MRI is unique in that the raw images are complex. Complex wavelet transforms (CWT) can be used to produce flexible signal representations when compared to Discrete Wavelet Transform (DWT). In this work, five different schemes using CWT or DWT are tested for sparse representation of MRI images which are in the form of complex values, separate real/imaginary, or separate magnitude/phase. The experimental results on real in-vivo MRI images show that appropriate CWT, e.g., dual-tree CWT (DTCWT), can achieve sparsity better than DWT with similar Mean Square Error.

authors

• Ji, Jim

published proceedings

• Annu Int Conf IEEE Eng Med Biol Soc

author list (cited authors)

• Nandakumar, H. P., & Ji, J.

citation count

• 0

complete list of authors

• Nandakumar, Hari Prasad||Ji, Jim

publication date

• August 2008

publisher

• Institute of Electrical and Electronics Engineers (IEEE)  Publisher

published in

• n2375-7477ISSN  Journal

keywords

• Algorithms
• Data Compression
• Image Enhancement
• Image Interpretation, Computer-Assisted
• Magnetic Resonance Imaging
• Pattern Recognition, Automated
• Reproducibility Of Results
• Sensitivity And Specificity

PubMed Central ID

• 19162677

Digital Object Identifier (DOI)

• 10.1109/IEMBS.2008.4649174

start page

• 398

end page

• 401

volume

• 2008

URL

• http://dx.doi.org/10.1109/iembs.2008.4649174