Extension of the 4D NURBS-based Cardiac-Torso (NCAT) Phantom to X-ray CT


W. Paul Segars, and Benjamin M. W. Tsui


The 4D NCAT phantom was developed to provide a realistic and flexible anatomical and physiological model of the human torso for nuclear medicine imaging research, specifically single photon emission computed tomography (SPECT) and positron emission tomography (PET). One limitation of the 4D NCAT is that the anatomy is based solely on the Visible Male CT dataset from the National Library of Medicine (NLM). Female subjects are simply modeled with the addition of user-defined breast extensions. Another limitation is that the phantom, originally designed for low-resolution imaging research, lacks the anatomical details for application to high-resolution imaging such as x-ray CT. In this work, we greatly enhanced the male and female anatomical detail of the phantom and extended it beyond the region of the torso to include structures in the head and abdomen.



Fig. 1.  Extension of the 4D NCAT anatomy, male (left) and female (right) template anatomies are shown.

The original template or base anatomy for the NCAT male was updated based on the high-resolution Visible Male anatomical dataset from the NLM. In addition, a separate template for the female was created based on the Visible Female anatomical data. The anatomical structures for the male and female were segmented from these datasets using the Analyze software from the Mayo Clinic. Models for the segmented structures were then created using a combination of NURBS and subdivision (SD) surfaces and incorporated into the 4D NCAT phantom. Cubic NURBS surfaces were used to model concave anatomical structures while SD surfaces were used to model structures with an arbitrary topological type, such as the brain, skull, muscle tissue, and vasculature.


The male and female base anatomies of the 4D NCAT phantom were developed to include over 300 structures or tissues in the torso, abdomen, and head, Fig. 1. With the improved anatomical detail and the extension to the head and abdomen, the 4D NCAT is applicable to more low-resolution medical imaging applications using nuclear medicine or high-resolution techniques such as x-ray CT where small anatomical details are more prominent and important.


Unlike other phantoms used in imaging research, the new 4D NCAT provides a realistic and flexible model of the human anatomy and physiology, thereby, providing simulated image data that are far more consistent with that of actual patients. It can be used to model a multitude of different anatomical variations and patient motions from both male and female subjects. Combined with accurate models of the imaging process, it can provide a wealth of simulated patient data for imaging research. We conclude that the new, extended 4D NCAT will provide an important tool in imaging research.


NIH Research Grant RO1EB001838