Gradient weighted smoothing for MRI intensity correction
Feng Huang, Hu Cheng, Srinivasan Vijayakumar · 2005
In magnetic resonance imaging (MRI), surface coils have the advantage of higher signal to noise ratio (SNR) but suffer from image inhomogeneity due to the non-uniform sensitivity profile. To remove bright spots caused by non-uniform sensitivity profiles, a gradient weighted smoothing method is discussed in this work. A partial differential equations (PDE) based model is applied for this locally adaptive smoothing. The filtered gradient of the corrupted image is used as the weight for smoothing. For accuracy evaluation, the energy of the histogram spectrum difference and the correlation of the intensity corrected image with the image acquired using the body coil are used as quantitative criteria. Clinical data collected on various MRI systems are used for evaluation of stability. The experimental results show that the proposed method is accurate and robust for intensity correction, and outperforms non-adaptive methods.