Automated determination of the resonance frequencies of protons for magnetic resonance examinations

1. A method for automated determination of a resonance frequency of an atomic nucleus for magnetic resonance examinations, comprising:

with a computer, operating a magnetic resonance data acquisition apparatus to acquire at least one magnetic resonance signal from a subject;

in said computer, Fourier-transforming said at least one signal into a measured spectrum comprised of elements that can be represented as a vector;

in said computer, generating a spectrum matrix comprising at least two vectors determined from said measured spectrum, each of said at least two vectors being formed using at least a portion of said measured spectrum;

in said computer, automatically cross-correlating said spectrum matrix, with a model spectrum represented as a vector, said model spectrum comprising at least one spectral peak at a modeled position in said model spectrum for a known atomic nucleus, in order to identify a peak in said measured spectrum that has a highest correlation coefficient in said cross-correlation;

in said computer, identifying a frequency in said measured spectrum of the identified peak in said measured spectrum;

with said computer, operating said magnetic resonance data acquisition apparatus at a resonance frequency corresponding to the identified frequency in said measured spectrum, in order to acquire diagnostic magnetic resonance data from the subject; and

in said computer, transforming said diagnostic magnetic resonance data into image data of the subject, and making the image data available from the computer in electronic form, as a data file.

2. A method as claimed in claim 1 comprising, in said computer, shifting a position of said measured spectrum, or portions of the measured spectrum, in said at least two vectors of the spectrum matrix with respect to each other in order to produce said spectrum matrix.

3. A method as claimed in claim 2 comprising, in said computer, determining a number of the elements in each of said at least two vectors of said spectrum matrix as a function of a number of elements of said measured spectrum and as a function of displacement steps in said shifting of said spectrum.

4. A method as claimed in claim 1 comprising always using an entirety of said measured spectrum, or always using a same part of said measured spectrum, for forming said at least two vectors of said spectrum matrix.

5. A method as claimed in claim 1 comprising forming said at least two vectors of said spectrum matrix by adding elements having a predetermined numerical value to said measured spectrum or a portion thereof.

6. A method as claimed in claim 5 wherein said numerical value is zero.

7. A method as claimed in claim 5 comprising forming a first of said at least two vectors by first elements occupied by the added elements and by remaining elements that are occupied by elements of the measured spectrum, and forming subsequent vectors of said at least two vectors after said first vectors by shifting said measured spectrum or a portion of said measured spectrum by one element with respect to a first element in said first vector.

8. A method as claimed in claim 1 comprising using different or only partially identical portions of said measured spectrum for forming said at least two vectors of said spectrum matrix.

9. A method as claimed in claim 1 comprising forming spectrum matrix with said at least two vectors as rows or columns of said spectrum matrix.

10. A method as claimed in claim 1 comprising generating a model spectrum matrix in said computer by combining model spectra so as to each be a vector of said model spectrum matrix.

11. A method as claimed in claim 10 comprising forming each vector of said model spectrum matrix respectively from each model spectrum of said model spectra.

12. A method as claimed in claim 10 comprising partitioning said spectrum matrix and said model spectrum matrix into sub-matrices, and performing said cross-correlation with pairs of said spectrum matrix, sub-matrices and said model spectrum sub-matrices, one pair at a time, in said computer.

13. A method as claimed in claim 12 comprising forming each of said sub-matrices as a function of a size of at least one cache of said computer.

14. A method as claimed in claim 1 comprising operating said magnetic resonance data acquisition apparatus to acquire an FID signal as said at least one magnetic resonance signal.

15. A magnetic resonance apparatus comprising:

a magnetic resonance data acquisition unit;

a computer configured to operate said magnetic resonance data acquisition apparatus to acquire at least one magnetic resonance signal from a subject;

said computer being configured to Fourier-transform said at least one signal into a measured spectrum comprised of elements that can be represented as a vector;

said computer being configured to generate a spectrum matrix comprising at least two vectors determined from said measured spectrum, each of said at least two vectors being formed using at least a portion of said measured spectrum;

said computer being configured to automatically cross-correlate said spectrum matrix, with a model spectrum represented as a vector, said model spectrum comprising at least one spectral peak at a modeled position in said model spectrum for a known atomic nucleus, in order to identify a peak in said measured spectrum that has a highest correlation coefficient in said cross-correlation;

said computer being configured to identify a frequency in said measured spectrum of the identified peak in said measured spectrum;

said computer being configured to operate said magnetic resonance data acquisition apparatus at a resonance frequency corresponding to the identified frequency in said measured spectrum, in order to acquire diagnostic magnetic resonance data from the subject; and

said computer being configured to transform said diagnostic magnetic resonance data into image data of the subject, and to make the image data available from the computer in electronic form, as a data file.