Method for correcting spectral and imaging data and for using such corrected data in magnet shimming

Claim 3

A method for phase correcting spectral and imaging data according to claim 1, wherein the mathematical model includes phase parameters.

Claim 4

A method for phase correcting spectral and imaging data according to claim 2, wherein the data includes more than one signal, wherein preselected parameters are estimated for a plurality of signals, wherein steps c and d are repeated for each of the plurality of signals so that the initial estimates of preselected parameters are adjusted to obtain adjusted estimates of said parameters separately for each signal through regression analysis for each of the plurality of signals for which initial estimates of parameters have been made, and wherein the adjusted estimates of selected parameters determined for each of the plurality of signals are used in determining the phase corrected data.

Claim 5

A method for phase correcting spectral and imaging data according to claim 3, wherein the phase corrected data are determined by using frequency parameters and the adjusted estimates of phase parameters to obtain a correction function, and applying the correction function to the data in need of correction.

Claim 6

A method for phase correcting spectral and imaging data according to claim 3, wherein the phase corrected data are simulated by setting the phase parameters to zero, and calculating the simulated corrected data from the predetermined mathematical model using the phase parameters set to zero and other selected adjusted estimates of parameters.

Claim 7

A method for phase correcting spectral and imaging data according to claim 3, wherein the phase corrected data are determined by: f. using the adjusted estimates of parameters and the predetermined mathematical model, calculating a simulation of the data to be corrected; subtracting the calculated simulation of the data to be corrected from the data to be corrected to obtain a residual spectrum without the signals for which preselected parameters were estimated; h. using frequency parameters and the adjusted estimates of phase parameters to obtain a correction function, and applying the correction function to the residual spectrum to thereby phase correct the residual spectrum; i. simulating the phase corrected data by setting the phase parameters to zero, and calculating the simulated corrected data from the predetermined mathematical model using the phase parameters set to zero and the other adjusted estimates of parameters; and j. adding the phase corrected residual spectrum obtained in step h to the simulated phase corrected data of step i to produce phase corrected data.

Claim 8

A method for phase correcting spectral and imaging data according to claim 6, including the additional step of determining from the residual spectrum a baseline correction function, and using the baseline correction function to correct the data for baseline distortion.

Claim 9

A method for phase correcting spectral and imaging data according to claim 7, wherein the baseline correction function is applied to the residual spectrum to correct the residual spectrum for baseline distortion before addition of the residual spectrum to the simulated phase corrected data.

Claim 10

A method for phase correcting spectral and imaging data according to claim 3, wherein the predetermined mathematical model takes into account baseline distortion to provide adjusted baseline parameters, and the adjusted baseline parameters are used to correct the data for baseline distortion.

Claim 11

A method for phase correcting spectral and imaging data according to claim 3, wherein the predetermined mathematical model of the expected spectral data defines signals in the expected spectral data to have a shape defined substantially as a convolution of a Lorentzian and a sinc function.

Claim 12

A method for phase correcting spectral and imaging data according to claim 3, including, after adjusting the parameters for each of the plurality of signals, the additional steps of: k. identifying sets of overlapping signals in th plurality of signals; l. inserting the adjusted estimates of parameters for each of the signals in one set of overlapping signals into a predetermined overlap mathematical model of expected spectral data that takes into account the overlap of the signals; m. correcting the adjusted estimates of parameters for each of the signals in the set of overlapping signals for the effect of overlap through regression analysis with the overlap mathematical model; n. repeating steps l and m for each additional identified set of overlapping signals to obtain corrected adjusted estimates of parameters for each of the determined overlapping signals; and o. using the corrected adjusted estimates of parameters in subsequent steps in place of the adjusted estimates of parameters for each signal for which corrected adjusted estimates have been obtained.

Claim 13

A method for phase correcting spectral and imaging data according to claim 11, wherein sets of overlapping signals are identified as those signals for which a predetermined measure of fit between the predetermined mathematical model and the signal is not obtained through the process of claim 3.

Claim 14

A method for phase correcting spectral and imaging data according to claim 11, wherein sets of overlapping signals are identified by determining signal peaks and the separation of such peaks and comparing such separation to a predetermined function of signal widths.

Claim 15

A method for phase correcting spectral and imaging data according to claim 11, including, after obtaining corrected adjusted estimates of parameters for signals in overlapping sets of signals, the additional steps of: p. identifying sets of overlapping signals which form heterogeneously broadened signals not satisfactorily defined by the predetermined overlap mathematical model; q. inserting relevant estimated preselected parameters for an identified heterogeneously broadened signal into a predetermined decomposition mathematical model which represents a predefined decomposition of heterogeneous signals into a superposition of homogeneous signals; r. revising the initial estimates of preselected parameters through regression analysis with the decomposition mathematical model to obtain revised estimated parameters for the signals included in the heterogeneously broadened signal; s. repeating steps q and r for each additional heterogeneously broadened signal identified; and t. using the revised estimates of parameters in subsequent steps in place of the adjusted estimates of parameters for each signal for which revised estimates have been obtained.

Claim 16

A method for phase correcting spectral and imaging data according to claim 14, wherein the decomposition mathematical model includes a set of equations, each equation defining the difference between the expected value at a location along the heterogeneous signal due to the superposition of homogeneous signals at that location and the measured value of the heterogeneous signal at that location.

Claim 17

A method for phase correcting spectral and imaging data according to claim 15, wherein the regression analysis using the decomposition mathematical model changes the parameters a plurality of times in obtaining revised parameters providing the best-fit of the model to the data and includes the step of solving the set of equations each time the parameters are changed.

Claim 18

A method for phase correcting spectral and imaging data according to claim 16, wherein it is assumed that the phase angle over the spectral width of the data to be analyzed is constant, wherein as part of the regression analysis the phase of the data is changed as well as the non-phase model parameters, and the best-fit of the model is to the phase changed data, the phase change of the data at best-fit representing the revised phase parameters.

Claim 19

A method for phase correcting spectral and imaging data according to claim 14, wherein the decomposition mathematical model takes into account baseline distortion whereby baseline distortion corrections are applied to the data in revising the parameters during regression analysis.

Claim 20

A method for phase correcting spectral and imaging data according to claim 1, wherein the phase corrected spectrum is also corrected for baseline distortion.

Claim 21

A method for phase correcting spectral and imaging data according to claim 19, wherein the predetermined mathematical model takes into account baseline distortion to provide adjusted base line parameters, and the adjusted baseline parameters are used to correct the data for baseline distortion.

Claim 22

A method according to claim 20, additionally including the steps of: u. estimating preselected parameters for a chosen portion of the spectral data for which parameters have not been estimated; v. inserting the estimated preselected parameters for the chosen portion of the spectral data not containing signals into the predetermined mathematical model of expected spectral data; w. adjusting the initial estimates of preselected parameters through regression analysis with the mathematical model to obtain adjusted estimates of the preselected parameters representing the best-fit of the mathematical model to the spectral data for the chosen portion of the spectral data; x. repeating steps u,v, and w for additional chosen portions of the spectral data; and y. using the adjusted baseline parameters determining a baseline distortion corrected spectrum.

Claim 23

A method for phase correcting spectral and imaging data according to the claim 1, additionally including the steps of: z. using the adjusted estimates of parameters and the predetermined mathematical model, calculating a simulation of the data to be corrected; aa. subtracting the calculated simulation of the data to be corrected from the data to be corrected to obtain a residual spectrum without the signals for which preselected parameters were estimated; bb. using frequency parameters and the adjusted estimates of phase parameters to obtain a correction function, and applying the correction function to the residual spectrum to thereby phase correct the residual spectrum; cc. simulating the phase corrected data by setting the phase parameters to zero, and calculating the simulated corrected data from the predetermined mathematical model using the phase parameters set to zero and the other adjusted estimates of parameters; and dd. adding the phase corrected residual spectrum obtained in step bb to the simulated phase corrected data of step cc to produce phase corrected data.

Claim 24

A method for phase correcting spectral and imaging data according to claim 22, additionally including determining from the residual spectrum a baseline correction function, and using the baseline correction function to correct the data for baseline distortion.

Claim 25

A method for phase correcting spectral and imaging data according to claim 23, wherein the baseline correction function is applied to the residual spectrum to correct the residual spectrum for baseline distortion before addition of the residual spectrum to the simulated phase corrected data.

Claim 26

A method for phase correcting spectral and imaging data according to claim 22, including the additional step between steps bb and cc of changing selected parameters to modify the calculated data in a desired manner.

Claim 27

A method for phase correcting spectral and imaging data according to claim 25, wherein selected parameters are changed to suppress unwanted signals in the corrected data.

Claim 28

A method for phase correcting spectral and imaging data according to claim 22, additionally including after step aa, the steps of: ee. determining if any signals are present in the residual spectrum; ff. if signals are determined as present in the residual spectrum, adding random noise to the spectrum to disguise such signals; and gg. performing steps bb on the residual spectrum with random noise added to obtain a spectrum corrected for baseline distortion.

Claim 29

A method for phase correcting spectral and imaging data according to claim 1, wherein the mathematical model represents a predefined decomposition of heterogeneous signals into a superposition of homogeneous signals.

Claim 30

A method for phase correcting spectral and imaging data according to claim 28, wherein the mathematical model includes a set of equations, each equation defining the difference between the expected value at a location along the heterogeneous signal due to the superposition of homogeneous signals at that location and the measured value of the heterogeneous signal at that location.

Claim 31

A method for phase correcting spectral and imaging data according to claim 29, wherein the regression analysis using the decomposition mathematical model changes the parameters a plurality of times in obtaining revised parameters providing the best-fit of the model to the data and includes the step of solving the set of equations each time the parameters are changed.

Claim 32

A method for phase correcting spectral and imaging data according to claim 30, wherein it is assumed that the phase angle over the spectral width of the data to be analyzed is constant, wherein as part of the regression analysis the phase of the data is changed as well as the non-phase model parameters, and the best-fit of the model is to the phase changed data, and wherein the phase change of the data at best-fit represents the revised phase parameters.

Claim 33

A method for phase correcting spectral and imaging data according to claim 1, wherein a plurality of phase parameters are determined, wherein the phase parameters are used to determine a phase function, wherein there are ambiguities in the determined phase parameters, and wherein the ambiguities are removed before determination of a phase function by the additional steps of: determining from the method used for collection of the data the phase ambiguity which is equal to the minimum change of phase angles that will not change the appearance of the signal; determining a desired range equal to the phase ambiguity shifting determined phase parameters outside the desired range by the number of phase ambiguities necessary to bring such phase parameters inside the desired range; choosing a maximum allowed phase shift over the spectral frequency range of the data; determining from the phase ambiguity and the maximum allowed phase shift, the maximum number of break points over the frequency range of the data; dividing the determined phase parameters into a number of sets of adjacent phase parameters equal to one more than the number of break points, each phase parameter being in a maximum of two sets; for each set of phase parameters, interpolating between the phase parameters in the set to obtain an interpolation function associated with the particular set; for each interpolation function, positioning the phase parameters outside the associated set by allowing such phase parameters to be shifted by the phase ambiguity if such shift increases the closeness; interpolating all phase parameters as positioned as close as possible to the interpolation function to obtain a second interpolation function associated with each set; determining the goodness-of-fit of the second interpolation function to all phase parameters included in the interpolation comparing the goodness-of-fit for each of the second interpolation functions selecting the second interpolation function with the best goodness-of-fit which does not extend beyond the maximum phase shift over the range of resonance frequencies as the phase function to be used for phase correcting the data.

Claim 34

A method for phase correcting spectral and imaging data according to claim 1, wherein the data is collected from a sample in a magnetic field, wherein the predetermined mathematical model takes into account shim distortions to provide adjusted shim parameters, and the adjusted shim parameters are used to correct distortions in th magnetic field.

Claim 35

A method for phase correcting spectral and imaging data according to claim 1, and useful for adjusting the shim currents in a magnet having shim coils and in need of shim adjustments, wherein the data to be corrected is spectral data obtained with a sample producing at least one isolated signal and using a magnet in need of shim adjustments, additionally including: obtaining data using a magnet in need of shim adjustments with a sample producing at least one isolated signal of known shape; phase correcting in accordance with claim 1 the spectral data obtained; using the phase corrected data, calculating an adjustment criterion for such data, said criterion chosen to provide a single minimum corresponding to the best overall shim adjustment; adjusting in increments the current in various shim coils; obtaining additional spectral data for the sample with each adjustment of shim coil currents; phase correcting in accordance with claim 1 each additional spectral data obtained; calculating the adjustment criterion for each additional spectral data until a minimum in the adjustment criterion is obtained, the adjustment of shim coil currents when the criteria minimum is obtained being the best overall adjustment obtainable for such magnet.