Three-dimensional in silico breast phantoms for multimodal image simulations

Area of Science:

• Medical Imaging
• Computational Biology
• Biomedical Engineering

Background:

• Advances in breast tissue cellular biology are driving new targeted diagnostic and therapeutic strategies, including molecular imaging.
• Realistic anatomic simulators are crucial for developing and evaluating novel medical imaging techniques.

Purpose of the Study:

• To establish a scalable framework for creating realistic 3-D in silico breast phantoms (ISBP).
• To develop a versatile phantom adaptable for integrating future discoveries in breast imaging.

Main Methods:

• ISBP frames were constructed using scalable anatomical shapes and morphologic features from existing literature.
• Frames were populated with essential tissue subtypes, and object contrast functions were assigned.
• Data were resampled for mammography, sonography, and computed tomography; end-to-end simulation included stochastic variability and physical principles.

Main Results:

• Initial comparisons between simulated and clinical breast images showed reasonable agreement.
• The framework demonstrates the capability to integrate diverse imaging modalities and future scientific insights.
• The simulation techniques effectively incorporated stochastic variability and deterministic physical principles.

Conclusions:

• The developed ISBP framework provides a foundation for realistic breast phantom construction.
• This approach supports the evaluation of new imaging methods and the integration of evolving breast biology knowledge.
• Further development is guided by initial comparisons, aiming for enhanced realism in breast imaging simulations.