Method of quantifying magnetic resonance spectrum

1. A method of quantifying a magnetic resonance spectrum, comprising the steps of:

(1) observing and collecting n one-dimensional (1D) magnetic resonance spectra, each consisting of real and imaginary parts, by quadrature detection;

(2) finding integrated values, x i and y i (where i=1, 2, . . . , n), of the real and imaginary parts, respectively, over given ranges of frequencies of each of the obtained n 1D magnetic resonance spectra;

(3) fitting the integrated values x i and y i of the real and imaginary parts obtained in step (2) to a straight line y=a x+b, where b is an offset value, on a complex plane by a linear regression technique;

(4) projecting actually measured data points (x i , y i ) onto the fitting straight line y=a x+b and taking the projected points as novel data points (x i , y i );

(5) taking reference data items as coordinate values (x 1 , y 1 ) of a first point, finding a sign (sign i ) of (x i −x 1 ) of an ith data item (x i , y i ), and taking this sign as the sign of a spectral peak (magnetization);

(6) finding a value of sign i ×√{square root over ((x i −x 1 ) 2 +(y i −y 1 ) 2 )}{square root over ((x i −x 1 ) 2 +(y i −y 1 ) 2 )} based on the sign (sign i ) obtained in step (5) and handing the value over to a least squares fitting program for calculating a pulse width; and

(7) curve-fitting dependence of the strength of the spectral peak (magnetization) on the pulse width to a given model function based on the least squares fitting program for calculation of a pulse width and determining a pulse width optimal for observation of the spectrum from parameters of the model function to which the curve-fitting has been done.

2. A method of quantifying a magnetic resonance spectrum, comprising the steps of:

(1) observing and collecting n one-dimensional (1D) magnetic resonance spectra, each consisting of real and imaginary parts, by quadrature detection;

(2) finding integrated values, x i and y i (where i=1, 2, . . . , n), of real and imaginary parts, respectively, over given ranges of frequencies of each of the obtained n 1D magnetic resonance spectra;

(3) fitting the integrated values x i and y i of the real and imaginary parts obtained in step (2) to a straight line y=a x+b, where b is an offset value, on a complex plane by a linear regression technique;

(4) projecting actually measured data points (x i , y i ) onto the fitting straight line y=a x+b and taking the projected points as novel data points (x i , y i );

(5) finding a value of distance √{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )}{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )} from a projected data point (x 0 , y 0 ) under conditions where the peak intensity is zero to each projected data point (x i , y i ) and taking the value as an absolute value of an integral intensity of a spectral peak (magnetization) in the spectrum;

(6) taking reference data items as coordinate values (x 1 , y 1 ) of a first point, finding a sign (sign i ) of (x i −x 0 )/(x 1 −x 0 ) of an ith data item (x i , y i ), and taking this sign as the sign of a spectral peak (magnetization); and

(7) finding a value of sign i ×√{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )}{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )} based on absolute value obtained in step (5) and the sign obtained in step (6) and taking this value as an integral intensity of a spectral peak (magnetization).

3. A method of quantifying a magnetic resonance spectrum according to claim 2 , wherein an optimum pulse width is automatically determined during measurement of the magnetic resonance spectrum using the integral intensity of the spectral peak (magnetization).

4. A method of quantifying a magnetic resonance spectrum, comprising the steps of:

(1) observing and collecting n one-dimensional (1D) magnetic resonance spectra, each consisting of real and imaginary parts, by quadrature detection;

(2) finding integrated values, x i and y i (where i=1, 2, . . . , n), of real and imaginary parts, respectively, over given ranges of frequencies of each of the obtained n 1D magnetic resonance spectra;

(3) fitting the integrated values x i and y i obtained in step (2) to a straight line y=a x+b, where b is an offset value, on a complex plane by a linear regression technique;

(4) projecting actually measured data points (x i , y i ) onto the fitting straight line y=a x+b and taking the projected points as novel data points (x i , y i ); and

(5) finding distance √{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )}{square root over ((x i −x 0 ) 2 +(y i −y 0 ) 2 )} from a projected data point (x 0 , y 0 ) under conditions where zero peak intensity is produced to each projected data point (x i , y i ) and taking the values of the distance as an integral intensity of a spectral peak (magnetization) in the spectrum.

5. A method of quantifying a magnetic resonance spectrum, comprising the steps of:

(1) finding spectra of real and imaginary parts of the spectrum by quadrature detection;

(2) finding integrated values, a and b, over given regions of the obtained spectra of the real and imaginary parts, respectively; and

(3) calculating ±√{square root over ((a 2 +b 2 ))} as an integral intensity of a spectral peak in the magnetic resonance spectrum.

6. A method of quantifying a magnetic resonance spectrum according to claim 5 , wherein an optimum pulse width is automatically determined during measurement of the magnetic resonance spectrum using the integral intensity of the magnetic resonance peak (magnetization), the integral intensity.

7. A method of quantifying a magnetic resonance spectrum as set forth in any claim 5 , wherein said magnetic resonance spectrum is an NMR spectrum.

8. A method of quantifying a magnetic resonance spectrum as set forth in claim 5 , wherein said magnetic resonance spectrum is an ESR spectrum.