A Robust Parametric Method for Bias Field Estimation and Segmentation of MR Images

Chunming Li¹, Li Wang², J. Chris Gatenby¹, Adam Anderson¹, John C. Gore¹

In this work, we propose a novel parametric method for joint image segmentation and bias field estimation for MR images. The bias field is parameterized as a linear combination of smooth basis functions. Image segmentation and bias field estimation are performed by minimizing a cost function. A desirable advantage of the proposed method is its robustness to initialization, which thereby allows fully automatic applications. Comparisons with other methods show the advantage of our method in terms of accuracy and robustness.