Four-dimensional on-beam computed tomography reconstruction using projection-difference images

Area of Science:

• Medical Physics
• Radiotherapy
• Image Reconstruction

Background:

• Accurate dose verification in radiotherapy necessitates real-time anatomical data during beam delivery.
• Conventional cone-beam computed tomography (CBCT) is acquired before treatment, missing intrafractional tumor motion.
• This limits the precision of delivered dose assessment and adaptive radiotherapy strategies.

Purpose of the Study:

• To develop and evaluate a 4D On-Beam Computed Tomography (4D OBCT) framework.
• To capture dynamic anatomical states of the tumor during irradiation.
• To enable accurate intrafractional motion assessment for improved radiotherapy.

Main Methods:

• A thoracic phantom with simulated tumor motion was used.
• Megavoltage transmission data were acquired during treatment delivery.
• 4D OBCT images were reconstructed using transmission data and Monte Carlo simulations, compared against 4D planning CT (pCT) and 4D CBCT.

Main Results:

• 4D OBCT demonstrated superior accuracy with lower Root Mean Square Error (RMSE) compared to 4D CBCT in lung and chest wall regions.
• Dosimetric evaluation showed significantly higher gamma pass rates for 4D OBCT (99.76% at 2%/2mm) versus 4D CBCT (72.50%).
• The framework successfully reconstructed phase-resolved images reflecting intrafractional motion.

Conclusions:

• The developed 4D OBCT framework accurately reconstructs images of the target during irradiation.
• It provides high-resolution, motion-aware volumetric data crucial for radiotherapy.
• 4D OBCT offers a robust solution for precise post-delivery dosimetric verification and adaptive radiotherapy.