Adapted waveform libraries in magnetic resonance imaging

The authors consider several implementation issues associated with wavelet encoding, an application of wavelet techniques to signal acquisition in magnetic resonance imaging (MRI). A quantitative analysis of the signal-to-noise limitations of this method is presented. It is found that the localized support features of wavelet bases which provide desirable properties of imaging speed and artifact reduction also degrade the signal-to-noise performance. A partial solution, utilizing adapted waveform libraries, is suggested. In many cases, a priori information may be used in the creation of tailored bases. These provide resistance to artifacts only where needed, while encoding other portions of the image with more globally supported waveforms in order to enhance the signal-to-noise performance in these regions.<<ETX>>