Markerless lung tumor tracking and trajectory reconstruction using rotational cone-beam projections: a feasibility...

Area of Science:

• Medical Physics
• Radiotherapy Technology
• Image-guided Therapy

Background:

• Accurate tumor localization is crucial for effective radiotherapy.
• Current methods may struggle with tumor motion during rotational imaging.
• Need for real-time tumor tracking in dynamic treatment delivery.

Purpose of the Study:

• Develop and validate algorithms for direct tumor tracking in rotational cone-beam projections.
• Reconstruct accurate 3D tumor trajectories from tracking data.
• Assess the clinical feasibility and accuracy of the developed technique.

Main Methods:

• Developed algorithms for direct tumor tracking in rotational cone-beam projections.
• Reconstructed phase-binned 3D tumor trajectories.
• Validated the technique using digital phantoms, physical phantoms, and patient data.
• Compared tracking results against reference templates from 4D CT.

Main Results:

• Tracking errors (e(95)) were within 1.7 mm for phantoms and 3.3 mm for patients.
• Reconstructed 3D trajectories had maximum errors of 1.0 mm (phantoms) and 2.0 mm (patients).
• Demonstrated feasibility of direct tumor tracking and accurate trajectory reconstruction.

Conclusions:

• Direct tumor tracking in rotational images is feasible.
• Accurate 3D tumor trajectories can be reconstructed even from less precise tracking data.
• Potential applications include improved patient setup, quality assurance, and rotational therapy (VMAT).