Motion simulation of radio-labeled cells in whole-body positron emission tomography

Nils Marquardt,Tobias Hengsbach,Marco Mauritz,Benedikt Wirth,Klaus Schäfers

2024-07-10

Abstract:Cell tracking is a subject of active research gathering great interest in medicine and biology. Positron emission tomography (PET) is well suited for tracking radio-labeled cells in vivo due to its exceptional sensitivity and whole-body capability. For validation, ground-truth data is desirable that realistically mimics the flow of cells in a clinical situation. This study develops a workflow (CeFloPS) for simulating moving radio-labeled cells in a human phantom. From the XCAT phantom, the blood vessels are reduced to nodal networks along which cells can move and distribute to organs and tissues. The movement is directed by the blood flow which is calculated in each node using the Hagen-Poiseuille equation and Kirchhoffs laws assuming laminar flow. Organs are voxelized and movement of cells from artery entry to vein exit is generated via a biased 3D random walk. The probabilities of whether cells move or stay in tissues are derived from rate constants of physiologically based compartment modeling. PET listmode data is generated using the Monte-Carlo simulation framework GATE based on the definition of a large-body PET scanner with cell paths as moving radioactive sources and the XCAT phantom providing attenuation data. From the flow simulation of 10000 cells, 100 sample cells were further processed by GATE and listmode data was reconstructed into images for comparison. As demonstrated by comparisons of simulated and reconstructed cell distributions, CeFloPS can realistically simulate the cell behavior of whole-body PET providing valuable data for development and validation of cell tracking algorithms.

Medical Physics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to simulate the movement of radiolabeled cells in whole - body positron emission tomography (PET). Specifically, the researchers have developed a workflow (CeFloPS) for simulating the dynamic flow of radiolabeled cells in a human phantom. The dynamic cell - flow data generated by this workflow can serve as the basis for real data and be used to develop and validate new image reconstruction and tracking tools. The following are the specific problems that the paper attempts to solve:

1. **Cell Tracking**: Tracking the movement of radiolabeled cells in the body through PET technology, which is of great significance for medical and biological research. Cell tracking can help researchers better understand the behavior of cells in the body, especially in stem - cell imaging and cell - based therapies.

2. **Generating Real - data**: In order to verify the effectiveness of cell - tracking algorithms, data that can truly reflect cell - flow in clinical situations need to be generated. CeFloPS provides such real data by simulating the movement of cells in the vascular network and their distribution in organs and tissues.

3. **Blood - flow Simulation**: In order to accurately simulate the movement of cells, the researchers used the Hagen - Poiseuille equation and Kirchhoff's laws to calculate the blood - flow volume and vascular resistance at each node. These calculations are based on the laminar - flow assumption and take into account the flow characteristics of blood in blood vessels.

4. **Cell Distribution within Organs**: The researchers simulated the process of cells entering an organ from an artery and leaving it through a vein by means of a biased 3D random - walk method. In addition, a physiology - based compartment model was introduced to describe the transition probabilities of cells between different virtual compartments, thereby more realistically simulating the residence time of cells in tissues.

5. **PET Physics Simulation**: The GATE framework was used to generate PET list - mode data. These data are based on the definition of a large - scale PET scanner, in which cell paths are regarded as moving radiation sources and the XCAT model provides attenuation data. Through these simulations, the researchers can generate and reconstruct PET images for comparison with real data.

In summary, the main objective of this paper is to develop a workflow that can accurately simulate the movement of radiolabeled cells in the human body and provide reliable data support for the development and validation of cell - tracking algorithms.