Method and Apparatus to Improve an MRI Image

The invention is a method for improving image quality in MR imaging methods using the SENSE (SENSitivity Encoding) method, which is known to have degraded image quality due to numerical ill-conditioning (so called g-factor loss). The invention improves the numerical conditioning by means of an adaptive regularization (matrix conditioning), thereby improving image quality for a given scan time. This is accomplished by adaptively adjusting the regularization parameter for each pixel position to achieve a target ghost artifact suppression. In this manner, a higher degree of matrix conditioning is used in regions which have less artifact, thus improving the SNR in these regions.