Meeting Banner
Abstract #5202

Biased Fuzzy C means based intensity inhomogeneity correction for segmentation of abdominal fat in DIXON MR Images

Krishna Kanth Chitta1, Bhanu Prakash KN1, Suresh Sadananthan2, and Sendhil Velan S1

1Laboratory of Molecular Imaging, Singapore Bioimaging Consortium, Singapore, Singapore, 2Singapore Institute for Clinical Sciences, Singapore

Uniform distribution of intensity values for a given tissue type is desirable for accurate segmentation and quantification. Factors like non-uniform static magnetic field, motion artifacts, and inconsistent RF coil sensitivity introduce intensity inhomogeneity during MR image acquisition. Several methods for intensity inhomogeneity correction are proposed in the literature. We explored the suitability of Biased fuzzy C-means (BFCM) correction for quantification of abdominal fat from Dixon images. In our study we formulated a new 2-pass, 2D (intra-slice and inter-slice) BFCM framework for improved segmentation and quantification of abdominal fat.

This abstract and the presentation materials are available to members only; a login is required.

Join Here