Method for generating an image using nuclear magnetic resonance signals

Claim 1

A method for generating an image of an examination subject using nuclear magnetic resonance signals, said method comprising the steps of: disposing said examination subject in a fundamental magnetic field; subjecting the examination subject to RF pulses to induce nuclear magnetic resonance signals in said examination subject; generating first and second gradient magnetic fields, at least one of which is substantially periodic and a selected portion of said substantially periodic gradient magnetic field defining a scan; reading out said nuclear magnetic resonance signals in said patient in the presence of said first and second gradient magnetic fields; entering the samples obtained during a scan in one row of a measurement matrix, said measurement matrix having a number of columns corresponding to the number of scans; dividing said measurement matrix into a first sub-matrix and a second sub-matrix, said first sub-matrix containing only even-numbered rows of the measurement matrix with intervening rows set to zero and the second sub-matrix containing only the odd-numbered rows of the measurement matrix with intervening rows set to zero; subjecting said first and second sub-matrices to a Fourier transformation at least in the column direction to obtain a first intermediate image matrix U having entries U(i,j) and a second intermediate image matrix G having entries G(i,j), wherein i is a row number between 1 and N and j is a column number; multiplying one of said intermediate image matrices by a factor F(j) calculated from edge rows i=1 through i 1 and i=i 2 through N of the intermediate image matrices according to the following equation: ##EQU2## wherein G(i,j)* is the complex conjugate of the intermediate image matrix entries G(i,j); and adding the intermediate image matrix multiplied by the factor F(j) with the other intermediate image matrix.

Claim 2

A method for generating an image of an examination subject using nuclear magnetic resonance signals, said method comprising the steps of: disposing said examination subject in a fundamental magnetic field; subjecting the examination subject to RF pulses to induce nuclear magnetic resonance signals in said examination subject; generating first and second gradient magnetic fields, at least one of which is substantially periodic and a selected portion of said substantially periodic gradient magnetic field defining a scan; reading out said nuclear magnetic resonance signals in said patient in the presence of said first and second gradient magnetic fields; entering the samples obtained during a scan in one row of a measurement matrix, said measurement matrix having a number of columns corresponding to the number of scans; dividing said measurement matrix into a first sub-matrix and a second sub-matrix, said first sub-matrix containing only even-numbered rows of the measurement matrix with intervening rows set to zero and the second sub-matrix containing only the odd-numbered rows of the measurement matrix with intervening rows set to zero; subjecting said first and second sub-matrices to a Fourier transformation at least in the column direction to obtain a first intermediate image matrix U having entries U(i,j) and a second intermediate image matrix G having entries G(i,j), wherein i is a row number between 1 and N and j is a column number; multiplying one of said intermediate image matrices by a factor F(j) calculated from edge rows i=1 through i 1 and i=i 2 through N of the intermediate image matrices according to the following equation: ##EQU2## wherein G(i,j)* is the complex conjugate of the intermediate image matrix entries G(i,j); and adding the intermediate image matrix multiplied by the factor F(j) with the other intermediate image matrix.

Claim 5

A method as claimed in claim wherein the step of subjecting said first and second sub-matrices to Fourier transformation is further defined by subjecting said first and second sub-matrices to a Fourier transformation in the column direction only, and comprising the additional step of subjecting a matrix, obtained in the step of adding, to a Fourier transformation in the row direction only.