Fast quantification of proton magnetic resonance spectroscopic imaging with artificial neural networks

Bhat, Himanshu; Sajja, Balasrinivasa R.; Narayana, Ponnada A.

doi:10.1016/j.jmr.2006.08.004

Cited by 20 publications

(25 citation statements)

References 45 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To suppress the noise at the ends of the echo, we applied a modified Gaussian-Wiener filtering window to the signal, yielding the following deconvolution:

s_{V}^{0} (t_{d}) = s_{V} (t_{d}) \cdot w (t_{d}) / L_{V} (t_{d})

where w ( t d ) is the time domain-modified Gaussian-Wiener window function. The term w ( t d ) is given as:

w (t_{d}) = G (t_{d}) \frac{{| L_{V} (t_{d}) |}^{2}}{{| L_{V} (t_{d}) |}^{2} + α K_{V} (t_{d})},

where G is the Gaussian function, α is a scaling constant, K V is defined as (29),

K_{V} (t_{d}) = \frac{σ^{2}}{{| s_{V} (t_{d}) |}^{2}},

and σ 2 is the noise power calculated by

σ^{2} = \overset{true¯}{\sum_{δ} {| s_{V} (t_{δ}) |}^{2}}, t_{δ} \in last 1 / 8 th points of the FID .

…”

Section: Methodsmentioning

confidence: 99%

Spectral resolution amelioration by deconvolution (SPREAD) in MR spectroscopic imaging

Dong

Peterson

2009

Magnetic Resonance Imaging

View full text Add to dashboard Cite

Purpose: To develop, implement, and evaluate a novel postprocessing method for enhancing the spectral resolution of in vivo MR spectroscopic imaging (MRSI) data. Materials and Methods:Magnetic field inhomogeneity across the imaging volume was determined by acquiring MRI datasets with two differing echo times. The lineshapes of the MRSI spectra were derived from these field maps by simulating an MRSI scan of a virtual sample whose resonance frequencies varied according to the observed variations in the magnetic field. By deconvolving the lineshapes from the measured MRSI spectra, the linebroadening effects of the field inhomogeneities were reduced significantly.Results: Both phantom and in vivo proton MRSI spectra exhibited significantly enhanced spectral resolutions and improved spectral lineshapes following application of our method. Quantitative studies on a phantom show that, on average, the full width at half maximum of water peaks was reduced 42%, the full width at tenth maximum was reduced 38%, and the asymmetries of the peaks were reduced 86%. SPECTRAL RESOLUTION DEFINES the ability to distinguish two closely spaced peaks in a spectrum. It is one of the most important determinants of the quality of MR spectroscopy (MRS) data (1-3). Low spectral resolution can obscure the information available from a spectrum of tissue metabolites, hindering the detection and quantification of some or all of those metabolites. This is especially true for in vivo proton MRS ( 1 H MRS) of the brain, for several reasons. First, the differences in the chemical shifts of brain metabolites are relatively small (i.e., the spectral peaks are close to one another), and the line splitting caused by J-coupling further reduces the separation of the peaks in the spectrum. Consequently, some spectral lines overlap intrinsically and cannot be easily distinguished. Second, inhomogeneities of the magnetic field caused by the spatial variation of the external field, B 0 , and by the local differences in magnetic susceptibility of different tissues, produce line broadening and distortion of the lineshape, thereby reducing spectral resolution. Consequently, some lines that are intrinsically separated may overlap. Third, motion of the tissue being imaged may produce line broadening and reduce spectral resolution. ConclusionMany techniques have been developed to improve the spectral resolution of in vivo MRS. These techniques generally fall into one of two classes, depending on whether they are used during the acquisition or the processing of spectral data. The most commonly used strategy for improving spectral resolution during data acquisition is to improve the homogeneity of the main magnetic field, B 0 . Fast, high-order shimming techniques (4) have been implemented on modern scanners, yet these methods cannot eliminate variations in the local magnetic field across the imaging volume that are caused by differing magnetic susceptibilities of the various tissues that are interposed within the body. Another strategy to increase spectral resolu...

show abstract

“…To suppress the noise at the ends of the echo, we applied a modified Gaussian-Wiener filtering window to the signal, yielding the following deconvolution:

s_{V}^{0} (t_{d}) = s_{V} (t_{d}) \cdot w (t_{d}) / L_{V} (t_{d})

where w ( t d ) is the time domain-modified Gaussian-Wiener window function. The term w ( t d ) is given as:

w (t_{d}) = G (t_{d}) \frac{{| L_{V} (t_{d}) |}^{2}}{{| L_{V} (t_{d}) |}^{2} + α K_{V} (t_{d})},

where G is the Gaussian function, α is a scaling constant, K V is defined as (29),

K_{V} (t_{d}) = \frac{σ^{2}}{{| s_{V} (t_{d}) |}^{2}},

and σ 2 is the noise power calculated by

σ^{2} = \overset{true¯}{\sum_{δ} {| s_{V} (t_{δ}) |}^{2}}, t_{δ} \in last 1 / 8 th points of the FID .

…”

Section: Methodsmentioning

confidence: 99%

Spectral resolution amelioration by deconvolution (SPREAD) in MR spectroscopic imaging

Dong

Peterson

2009

Magnetic Resonance Imaging

View full text Add to dashboard Cite

show abstract

“…Taking into account that each peak can be approximated by the Voigt line shape [6], consisting of a Lorentzian and a Gaussian parts, the process of peak detection in such spectrums, constitutes a typical optimization problem. This is exactly our view compared to that of the state of the art.…”

Section: Proposed Methodologymentioning

confidence: 99%

“…Significant contributions based on Artificial Intelligence (AI) tools, such as Neural Networks (NNs), with accurate results have been presented lately [1][2][3][4][5][6]. However, the usage of NNs to approximate the metabolites of a retrieved spectrum has the drawback of constructing many different neural models to address the possible characteristics of the spectrum in process.…”

Section: Introductionmentioning

confidence: 99%

Dealing with peaks overlapping issue in quantifying metabolites in MRSI

Papakostas

Karras

Mertzios

2009

2009 IEEE International Workshop on Imaging Systems and Techniques

View full text Add to dashboard Cite

A novel methodology deals with the peaks overlapping issue in quantifying metabolites in MRSI is proposed in this paper. The introduced method encounters the metabolites quantification procedure as a typical optimization problem able to be solved by using optimization methods of the computational intelligence. A simple genetic algorithm is applied in order to find the metabolites peaks parameters that best match the spectrum in process. This novel approach comes to overcome the disadvantages of the neural networks, used for the same purpose, where the overlapping issue is handled difficultly. The experimental results are very promising showing that highly overlapped metabolites peaks can quantified accurately, by using the proposed method and place the basis for further investigation.

show abstract

“…(1) The water signal can be used as a reference to correct lineshape distortion caused both by eddy currents induced in the metal surface of the magnet due to the varying magnetic gradient field [44][45][46] and by inhomogeneity of the static magnetic field [47][48][49]. (2) The water signal can be used as an internal reference for quantification of metabolites [18,[49][50][51][52][53][54][55][56][57][58]. (3) In MRSI, the water signal can be used to correct voxel-to-voxel frequency shifts caused by the inhomogeneity of the main magnetic field B 0 [49,59].…”

Section: Introductionmentioning

confidence: 99%

“…(2) The water signal can be used as an internal reference for quantification of metabolites [18,[49][50][51][52][53][54][55][56][57][58]. (3) In MRSI, the water signal can be used to correct voxel-to-voxel frequency shifts caused by the inhomogeneity of the main magnetic field B 0 [49,59]. (4) In single voxel MRS with a large number (usually of signal averaging to increase SNR, water signal can be used to correct lineshape distortions and to eliminate artifacts caused by subject motion [60].…”

Section: Introductionmentioning

confidence: 99%

Proton MRS and MRSI of the brain without water suppression

Dong

2015

Progress in Nuclear Magnetic Resonance Spectroscopy

View full text Add to dashboard Cite

Fast quantification of proton magnetic resonance spectroscopic imaging with artificial neural networks

Cited by 20 publications

References 45 publications

Spectral resolution amelioration by deconvolution (SPREAD) in MR spectroscopic imaging

Spectral resolution amelioration by deconvolution (SPREAD) in MR spectroscopic imaging

Dealing with peaks overlapping issue in quantifying metabolites in MRSI

Proton MRS and MRSI of the brain without water suppression

Contact Info

Product

Resources

About