Statistical limitations in proton imaging.

Area of Science:

• Medical Physics
• Radiotherapy Imaging
• Proton Beam Applications

Background:

• Proton imaging is a valuable tool in proton radiotherapy, enabling direct tissue stopping power measurement, multi-modality reconstruction, calibration, motion tracking, and pre-treatment positioning.
• A comprehensive end-to-end characterization of proton imaging quality, specifically signal-to-noise ratio and spatial resolution relative to dose, has been lacking.
• Understanding the interplay between imaging parameters and dose is crucial for optimizing proton imaging's clinical utility.

Purpose of the Study:

• To establish a quantitative model relating proton imaging quality metrics (noise, spatial resolution, blurring uncertainty) to radiation dose.
• To investigate the influence of proton energy and object size on these image quality characteristics.
• To provide a framework for optimizing proton imaging usage in radiotherapy.

Main Methods:

• Developed a model to analyze image noise originating from Coulomb scattering and energy loss straggling.
• Investigated the impact of traversed object thickness and proton energy on noise levels.
• Assessed the behavior of spatial resolution and noise metrics across different proton energies and object sizes.

Main Results:

• Image noise increases with traversed thickness and decreases with decreasing proton energy.
• Scattering noise dominates at high-gradient edges, while straggling noise is maximal in homogeneous regions.
• Lower proton energy reduces both noise and spatial resolution, necessitating application-specific energy optimization.

Conclusions:

• The developed model provides insights into the trade-offs between proton imaging quality and energy settings.
• This work facilitates the optimal application of proton imaging in radiotherapy and enables fair comparisons with other imaging modalities.
• Understanding these relationships is key to fully realizing the potential of proton imaging technology.