Application of a fuzzy inference system to the quantification of 3D magnetic resonance imaging of breast tissue

摘要

The objective of this study was to develop a segmentation technique to quantify breast tissue and total breast volume from MRI data. The goal of our research is to quantify breast density using MRI to help better assess breast cancer risk for certain high-risk populations for whom mammography is of limited usefulness due to their high breast density. A semi-automatic segmentation technique was implemented based on a fuzzy inference system to segment 3D breast tissue from fat, and quantify the total volume of the breast in order to obtain an index of MR breast density on 10 healthy volunteers. The algorithm was based on two non-contrast 3D MR sequences. A fuzzy c-means algorithm was used to provide a first estimate of the segmentation of breast tissue from fat on specific slices. Based on the means and standard deviations of the segmented groups (breast tissue and fat) Sugeno-type fuzzy inference systems were built and then used as the main segmentation tools to segment surrounding slices. Results of volumetric measurements and breast density index obtained with the semi-automated method were compared with quantitative results obtained using classical global thresholding segmentation technique.