Validation and application of a perceptual difference model for keyhole MR imaging

To optimize the many parameters in keyhole MR imaging, we validated and applied a human visual system (HVS) perceptual difference model (PDM) to simulated keyhole images. A series of full k-space images were acquired during the insertion of a needle into ex vivo bovine liver. Human observers rated the quality of images, and we compared ratings with the PDM output to validate and calibrate the model. In a second experiment, eight keyhole sequences were evaluated by the HVS PDM. A linear relationship with R/sup 2/=99.68% was found by regressing the model output versus human observer ratings. According to the perceptual model, the quality of the entire image is preserved most favorably with a stripe parallel to the direction of insertion. For a region of interest around the needle, a perpendicular stripe resulted in the lowest level of image error. Examination of rotating stripes of k-space show that a step of 45 degrees gives better image quality. Initial experience indicates that the HVS model is an objective, promising tool for the automated evaluation and optimization of keyhole imaging sequences.