Realistic 3D printed CT imaging tumor phantoms for validation of image processing algorithms.

Area of Science:

• Medical Imaging
• Additive Manufacturing
• Radiotherapy

Background:

• Medical imaging phantoms are crucial for validating systems and algorithms in surgical guidance and radiation oncology.
• Accurate phantoms require replication of human tissue morphology and radiological properties.
• Additive manufacturing offers a customizable and cost-effective solution for phantom fabrication.

Purpose of the Study:

• To develop a simple, inexpensive protocol for manufacturing realistic tumor phantoms using Fused Deposition Modeling (FDM) 3D printing.
• To evaluate the radiodensity similarity between 3D printed phantoms and actual lung cancer patient CT data.
• To investigate the impact of phantom radiodensity heterogeneity on image registration algorithm validation.

Main Methods:

• Utilized Fused Deposition Modeling (FDM) filament 3D printing technology.
• Fabricated tumor phantoms with both homogenous and heterogeneous radiodensity.
• Evaluated radiodensity similarity using CT images from lung cancer patients.

Main Results:

• Achieved a radiodensity range of -217 to 226 Hounsfield Units (HU) in 3D printed phantoms, matching human lung tumor tissue.
• 3D printed phantoms accurately replicated patient tumor morphology and radiodensity heterogeneity.
• No significant influence of heterogeneity on the accuracy or robustness of image registration algorithms was observed.

Conclusions:

• The proposed FDM protocol provides a cost-effective method for creating realistic CT tumor phantoms.
• The achieved HU range and morphological accuracy support the use of these phantoms for validating medical imaging algorithms.
• Heterogeneity in 3D printed tumor phantoms does not negatively impact image registration algorithm performance.