METHODS AND DEVICES FOR OPTIMIZATION OF CONTRAST INHOMOGENEITY CORRECTION IN MAGNETIC RESONANCE IMAGING

摘要

MRI techniques are widely and successfully applied in medicine and biophysics because MRI provides good contrast between different soft tissues without ionizing radiation. MRI protocols are optimized in four aspects: imaging parameters, k-space strategies, RF system calibration and contrast inhomogeneity correction. The signal intensities of normal and disease tissues are simulated, for example, using Bloch equations for an imaging sequence with tissue MR parameters. The relationships between imaging parameters and tissue contrasts are calculated using the numerically simulated signal intensities. The optimal imaging parameters and/or k-space strategies are determined to maximize image contrast and minimize artifacts with acceptable spatial-temporal resolution based on characterization of the imaging hardware. The RF system is optionally calibrated to improve the accuracy of the imaging parameters and reduce inter-scanner variability. Additionally, contrast-to-noise inhomogeneity caused by transmit field and receive sensitivity is optionally corrected by optimal flip angle and measured receiver sensitivity.

主权项

1. A method of correcting for inhomogeneous contrast in an MRI image, comprising:
simulating tissue contrast for an MRI acquisition sequence; finding one or more optimal imaging parameters for reducing a sensitivity of a contrast inhomogeneity caused by a non-uniform acquisition condition; acquiring a set of signal intensity images using the one or more optimal imaging parameters; estimating a receive sensitivity; registering the receive sensitivity with the set of signal intensity images to produce a relative correction matrix; normalizing the relative correction matrix to obtain a correction matrix; and correcting the contrast inhomogeneity caused by the receive sensitivity in the set of signal intensity images by calculating a ratio of the set of signal intensity images and the correction matrix.