CMStatistics 2018: Start Registration
View Submission - CMStatistics
B1374
Title: A Bayesian spatial clustering method and its application in radiology Authors:  Jing Cao - Southern Methodist University (United States) [presenting]
Song Zhang - University of Texas Southwestern Medical Center (United States)
Abstract: Kidney cancer is among the ten most common cancers in human. The dynamic contrast-enhanced MRI (DCE-MRI) takes advantage of the interaction between a contrast agent and adjacent water protons which generates brighter signals in the scan image. A novel Bayesian spatial clustering method based on a mixture of multivariate normal distribution is proposed. A latent conditional regression (CAR) process is employed to account for the spatial correlation among clustering indexes. The proposed method is demonstrated to provide smoother and more accurate clustering of pixels. A simulation study and a real application example are presented.