Quantitative analysis of NMR spectra by linear prediction and total least squares

Tirendi, Charles F.; Martin, Joel F

doi:10.1016/0022-2364(89)90331-4

Cited by 25 publications

(12 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These MC studies were carried out on a simulation signal that was designed more important during the first iterations of the enhancement procedure. Therefore, it is advised, after fixing the number in order to represent an in vivo MRS signal as closely as (3), LPSVD preceded by the Cadspectrum displays specific problems such as the closely overzow EP (9) with one iteration (EPLPSVD), and the linearlapping multiplets from adenosine triphosphate (ATP) and prediction total-least-squares (LPTLS) method (6,7). the narrow inorganic phosphate (P i ) peak wedged between These comparative studies were not performed with respect the broad phosphomonoester (PME) and the very broad to rank estimation but in terms of parameter accuracy.…”

Section: Methodsmentioning

confidence: 99%

“…In the first set of MC studies, we investigate the space methods such as Hankel Lanczos singular-value deperformance of LPSVD(CR) as a stand-alone method to composition (HLSVD) (4), methods using total least determine the correct number of components and find estisquares (TLS) such as Hankel TLS (HTLS) (5), the combimates for the spectral parameters. The automatic quantitation nation of TLS with LP (LPTLS) (6,7), and others.…”

Section: Introductionmentioning

confidence: 99%

“…[5] tion of TLS with LP ( LPTLS ) ( 6,7 ) . The essential difference between LPSVD and LPTLS is that, in the latter algorithm, the SVD of the matrix [ X , x ] instead of X is and truncate to rank K õ M j : truncated to rank K and the minimum-norm total-leastsquares solution is calculated.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

The Use of Continuous Regularization in the Automated Analysis of MRS Time-Domain Data

Totz

Boogaart

Huffel

et al. 1997

Journal of Magnetic Resonance

View full text Add to dashboard Cite

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

The Use of Continuous Regularization in the Automated Analysis of MRS Time-Domain Data

Totz

Boogaart

Huffel

et al. 1997

Journal of Magnetic Resonance

View full text Add to dashboard Cite

“…A lot of variants of the above outlined algorithm have been developed to solve eqn (10) (or its backward equivalent) in such a way that the computational complexity is reduced and that a large number of data points and peaks can be handled. [37][38][39][40] In the LPTLS method 39 the SVD of [H lp h lp ] is truncated to rank K and the total least squares (TLS) solution is computed, resulting in improved parameter accuracy. Instead of truncating the SVD to rank K, a continuous regularization technique, called LPSVD(CR) 41 can be applied, an approach that leads to increased frequency resolution and the ability to automatically detect the number of significant peaks in the signal.…”

Section: Quantitation Algorithmsmentioning

confidence: 99%

MR spectroscopy quantitation: a review of time‐domain methods

et al. 2001

View full text Add to dashboard Cite

In this article an overview of time-domain quantitation methods is given. Advantages of processing the data in the measurement domain are discussed. The basic underlying principles of the methods are outlined and from them the situations under which these algorithms perform well are derived. Also an overview of methods to preprocess the data is given. In that respect, signal-to-noise and/or resolution enhancement, the removal of unwanted components and corrections for model imperfections are discussed.

show abstract

“…There are a variety of algorithms for calculating the coefficients of the AR and MA model components through minimization of prediction error. Similar methods of data extrapolation have been applied to MR spectroscopy to deal with omitted data points at the start of a free induction decay (10).…”

mentioning

confidence: 99%

Padé methods for reconstruction and feature extraction in magnetic resonance imaging

Callaghan

Larkman

Hajnal

2005

Magnetic Resonance in Med

View full text Add to dashboard Cite

Methods utilizing Padé approximants are investigated for implementation with magnetic resonance imaging data and are presented both for direct image reconstruction and for feature extraction. Padé approximants are a numerical tool that can be used to accelerate the convergence of a slowly converging sequence by estimating the fully converged sequence values from early data points. Padé approximants can be calculated directly from k-space data by solving a set of linear matrix equations to produce signal values for any desired location in the image domain. This gives an estimate of the fully converged signal intensity at each pixel location in the image, raising the possibility of reconstructing a better estimate of the object from a reduced data set. These methods have been tested on phantom and human data both for image reconstruction and for feature extraction. In image reconstruction, considerable con- Image resolution is a critical parameter in many magnetic resonance imaging (MRI) studies. High-resolution data are often required in order to achieve accurate anatomic assessment and/or quantitative measurements. In conventional phase-encoded MRI, acquiring higher resolution data increases the scan time due to the serial nature of the phase encoding process. This reduces the achievable temporal resolution and also increases the risk of motion artifact corrupting the data. Lowering the resolution decreases the scan time and also increases the level of truncation artifact, which manifests as unrepresentatively high and low signal bands running parallel to any interface where there is an abrupt change in the NMR signal (1). This artifact arises when the collected k-space data do not contain all the high-frequency information present in the slice that has been imaged and is a fundamental property of the Fourier encoding scheme used in MRI. Truncation artifact can be mistaken for, or occlude, structure.For a given rate of data acquisition with a fixed field of view (FoV) (i.e., number of k-space points per unit time), a number of methods have been implemented to achieve faster image update rate, while maintaining high spatial resolution (2-6). These methods increase apparent temporal resolution by partially updating the data in each new image. In keyhole based methods (2,3) new low-resolution data sets are inserted into a high-resolution reference to give a dynamic series. In sliding window techniques a number of k-space segments are acquired and used to dynamically update a high-resolution data set by replacing the oldest segment with the newest (4,5). The k-t BLAST (6) approach allows for reduced data acquisitions by tailoring the sampling strategy to exploit correlations in both k-space and time.Another approach is to extrapolate from incomplete data. This approach has been applied to MR imaging to reduce truncation artifacts and improve resolution (7-9). In the TERA algorithm, the MR data are modeled as a subset of the transient response of an infinite impulse filter and the data are approximated by a determini...

show abstract

Quantitative analysis of NMR spectra by linear prediction and total least squares

Cited by 25 publications

References 5 publications

The Use of Continuous Regularization in the Automated Analysis of MRS Time-Domain Data

The Use of Continuous Regularization in the Automated Analysis of MRS Time-Domain Data

MR spectroscopy quantitation: a review of time‐domain methods

Padé methods for reconstruction and feature extraction in magnetic resonance imaging

Contact Info

Product

Resources

About