Method and apparatus for enhanced multiple coil imaging

1. A method of processing magnetic resonance imaging signals from a plurality of magnetic resonance imaging coils, comprising:

determining a noise covariance matrix, N, of a plurality of magnetic resonance imaging coils, wherein the noise covariance matrix, N, of the plurality of magnetic resonance imaging coils is a Hermitian symmetric matrix;

receiving a corresponding plurality of signals, s 1 ,s 2 , . . . ,s n , from the plurality of coils to produce signal vector S=[s 1 ,s 2 , . . . ,s n ], wherein the plurality of signals represent a corresponding plurality of pixel values for a location;

converting signal vector S to signal vector Ŝ, where, Ŝ=(K * )·S and N=K * K, where K is a matrix for which the conjugate row-column transpose of K(ie K * ), multiplied by the matrix K is the noise covariance matrix, N; and

calculating a composite pixel value for the location, √Ŝ • ·Ŝ, where Ŝ 500 =[(K * ) −1 ·S] * .

2. The method according to claim 1 , wherein converting signal vector S to signal vector Ŝ comprises converting signal vector S to signal vector Ŝ in the time domain.

3. The method according to claim 1 , wherein Ŝ=[ŝ 1 ,ŝ 2 , . . . ,ŝ n ] is produced by inputting S=[s 1 ,s 2 , . . . ,s n ] into a circuit, wherein the output of the circuit is

ŝ 1 =a 1.1 s 1 +a 1.2 s 2 + . . . +a 1.n s n

ŝ 2 =a 2.1 s 1 +a 2.2 s 2 + . . . +a 2.n s n

ŝ n =a n.1 s 1 +a n.2 s 2 + . . . +a n.n s n

wherein a 1.1 ,a 1.2 , . . . ,a 1.n ,a 2.1 ,a 2.2 , . . . ,a 2.n , . . . ,a n.1 ,a n.2 , . . . ,a n.n are constants.

4. The method according to claim 3 , wherein a 1.1 ,a 1.2 , . . . ,a 1.n ,a 2.1 ,a 2.2 , . . . ,a 2.n , . . . ,a n.1 ,a n.2 , . . . ,a n.n are values of (K * ) −1 such that a 1.1 ,a 1.2 , . . . ,a 1.n ,a 2.1 ,a 2.2 , . . . ,a 2.n , . . . ,a n.1 ,a n.2 , . . . ,a n.n are equal to

(K * ) −1 1.1 ,(K * ) −1 1.2 , . . . ,(K * ) −1 2.1 ,(K * ) −1 2.2 , . . . ,(K * ) −1 2.n , . . . ,(K * ) −1 N.1 , (K * ) −1 n.2 , . . . ,(K * ) −1 n.n respectively.

5. The method according to claim 4 , wherein K is constructed via eigenvalue/vector decomposition of the noise covariance matrix N.

6. The method according to claim 5 , wherein when the eigenvalue/vector decomposition of N yields one or more eigenvalues with substantially degenerate eigenvectors, the values of (K * ) −1 are adjusted such that the output of the circuit is Ŝ=[ŝ 1 ,ŝ 2 , . . . ,ŝ m ]where m<n.

7. The method according to claim 6 , wherein substantially all of the image information in received signals S=[s 1 ,s 2 , . . . ,s n ] is in output signals Ŝ=[ŝ 1 ,ŝ 2 , . . . ,ŝ m ].

8. The method according to claim 6 , wherein the adjustment of the values of (K * ) −1 results in the received signals with substantially the same eigenvalues being added together with a phase.

9. The method according to claim 8 , wherein two of the received signals are received from quadrature volume coils having fields which are substantially uniform and substantially perpendicular, wherein the values of (K * ) −1 are adjusted to accomplish the circularly polarized addition of the two received signals, such that substantially all of the image information in the two received signals is in one output signal.

10. The method according to claim 4 , further comprising: pre-amplifying the received signals prior to inputting S=[s 1 ,s 2 , . . . ,s n ] into the circuit.

11. The method according to claim 10 , further comprising:

mixing output signal S to lower frequencies;

sampling Ŝ with a lower frequency by A/D converters to produce a digital Ŝ signal;

applying a 2D Fourier Transform to the digital Ŝ signal; and

processing the Ŝ signal after 2D Fourier Transform applied with image domain matrices to produce a plurality of pixel values, Ŝ, for a location,

wherein a composite pixel value for the location, √Ŝ • ·Ŝ, utilizes Ŝ after processing with image domain matrices.