TomoPhantom

TomoPhantom generates 2D–4D analytical phantoms for benchmarking and testing CT image reconstruction algorithms.

Key Features:

• Modular Analytical Phantoms: Provides 2D and 3D phantoms with temporal extensions by combining geometrical objects such as Gaussians, parabolas, cones, ellipses, rectangles, and their volumetric counterparts.
• Benchmarking and Testing: Enables benchmarking and testing of tomographic image-processing and reconstruction algorithms across 2D and 3D scanning geometries.
• Analytical Tomographic Projections: Computes analytical tomographic projections to support realistic testing and to avoid the "inverse crime".
• Performance and Efficiency: Core modules are implemented in C with OpenMP to enable efficient generation of high–spatial-resolution volumetric phantoms.
• Language Support: Provides Python and MATLAB wrappers for integration into analysis workflows.

Scientific Applications:

• Tomography: Development and testing of tomographic imaging techniques.
• Image Reconstruction: Evaluation and development of iterative and advanced CT reconstruction algorithms.
• Research and Development: Creation of realistic phantom models for algorithm development in academic and industrial research.

Methodology:

Phantoms are constructed by composing parametric geometrical primitives (Gaussians, parabolas, cones, ellipses, rectangles and volumetric counterparts); analytical tomographic projections are computed; core modules are implemented in C with OpenMP; and Python and MATLAB wrappers are provided.