Author: Phillip Cuculich, Geoffrey D. Hugo, Xiwen Li, Michael T. Prusator, Clifford Robinson, Pamela Samson, Xue Wu 👨🔬
Affiliation: Washington University School of Medicine in St Louis, Washington University School of Medicine in St. Louis, University of Utah, WashU Medicine 🌍
Purpose:
For development of novel motion management methods it is useful to have a digital phantom capable of realistic simulation of respiratory 4DCT acquisition of the thorax, including cycle-to-cycle breathing variation and heartbeat. Here, based on the widely-available XCAT phantom, we develop an open-source python tool for virtual generation of respiratory 4DCT capable of simulating heartbeat- and irregular motion-induced artifacts.
Methods:
We designed a virtual respiratory 4DCT generator with cardiac and respiratory motions based on the XCAT digital phantom. In this generation process, 1) a set of 3D CT volumes were generated with cardiac and respiratory motion over several breathing cycles at high temporal frequency using XCAT. 2) CT projection data were simulated based on the 3D CT volumes at real temporal frequencies of motion. 3) the projection data for each stack of slices were grouped and selected based on the image acquisition time course of a helical respiratory 4DCT with user-selectable scanning parameters. 4) the stacks of slices were reconstructed based on the grouped projection data and combined into a respiratory 4DCT image. Artifacts were visually evaluated relative to static XCAT output.
Results:
The experiments were designed concentrating on the accuracy of capturing heart motion with respiratory period, cardiac period, maximum diaphragm and AP motion. They produced respiratory 4DCT scans with regular breathing, occasionally deep breathing and slight rotation of the body (< 5 degree). The results are consistent with realistic motion artifacts in respiratory 4DCT.
Conclusion:
We developed a virtual respiratory 4DCT generation based on the XCAT digital phantom. It can be used for 4DCT related development when clinical data is not available or when a realistic ground truth is desired. It also can be potentially used for other 4dct projection datasets.