Fuzzy Tissue Classification for Non-Linear Patient-Specific Biomechanical Models for Whole-Body Image Registration

摘要: Comparison of whole-body medical images acquired for a given patient at different times is important diagnosis, treatment assessment and surgery planning. Prior to comparison, the need be registered (aligned) as changes in patient’s posture other factors associated with skeletal motion deformations organs/tissues lead differences between images. For images, such are large, which poses challenges traditionally used registration methods that rely solely on image processing techniques. Therefore, our previous studies, we successfully applied using patient-specific biomechanical models predicting treated non-linear problem computational mechanics. Constructing tends time-consuming it involves tedious segmentation divides into non-overlapping constituents material properties. To eliminate segmentation, propose Fuzzy C-Means (FCM) classification assign properties integration points finite element mesh. In this study, present an application FCM tissue algorithm analyse sensitivity accuracy parameters. We show accurate (within two voxel size) can achieved.