Muqing Lin1, Siwa Chan2,
Jeon-Hor Chen1,3, Daniel H-E. Chang1, Ke Nie1,
Shih-Ting Chen4, Cheng-Ju Lin4, Tzu-Ching Shih4,
Orhan Nalcioglu1, Min-Ying
1Tu & Yuen Center for Functional Onco-Imaging & Department of Radiological Sciences, University of California, Irvine, CA, United States; 2Department of Radiology, Taichung Veterans General Hospital, Taichung, Taiwan; 3Department of Radiology, China Medical University, Taichung, Taiwan; 4Department of Biomedical Imaging & Radiological Science, China Medical University, Taichung, Taiwan
The purpose is to test a new bias-field correction method by combining N3 (nonparametric non-uniformity normalization) and Fuzzy-C-Means clustering based methods on breast MR images. The new algorithm is based on N3 for an initial correction; then FCM-based correction with smoothing using B-spline surface fitting was repeated iteratively for gradually refined improvements. The results indicated that the N3+FCM correction method performs significantly better than N3 or FCM, and comparable to CLIC. This new bias-field correction method can be implemented to improve the segmentation quality of breast density on inhomogeneous breast MRI, or other images acquired using a surface coil.