A fiducial detection algorithm for real-time image guided IMRT based on simultaneous MV and kV imaging

Area of Science:

• Medical Physics
• Radiation Oncology
• Image-Guided Therapy

Background:

• Highly conformal radiation therapy techniques like 3DCRT and IMRT are limited by intrafraction organ motion.
• Real-time 3D positioning of internal fiducial markers is crucial for overcoming these limitations.

Purpose of the Study:

• To develop and validate an algorithm for near real-time 3D tracking of fiducial markers during radiation therapy.
• To address key challenges hindering the clinical application of image-guided radiation therapy.

Main Methods:

• Analysis of simultaneous onboard kV and treatment MV beam images to determine fiducial marker positions.
• Investigated the impact of fiducial motion blur, MV beam scattering, and MLC-modified MV fields on detection.
• Implemented methods to predict marker positions and reduce search regions for real-time tracking.

Main Results:

• Fiducial markers were successfully detected at speeds up to 1.6 cm/s.
• Common MV beams did not hinder fiducial detection in simultaneous kV images.
• The algorithm achieved an average detection time of 0.1 s or less per frame for three markers.
• Demonstrated real-time tracking capability even with degraded images, rapid motion, and blocked fiducials.

Conclusions:

• The developed algorithm enables real-time tracking of moving fiducial markers during 3DCRT and IMRT treatments.
• This technique can provide gating signals or enable intra-fractional tumor tracking.
• The system can alert therapists to patient motion, allowing for treatment suspension and repositioning.