Cardiac SABR: Image matching techniques for accurate treatment delivery

Area of Science:

• Radiation Oncology
• Cardiology
• Medical Imaging

Background:

• Ventricular tachycardia (VT) is a life-threatening arrhythmia often refractory to conventional treatments.
• Cardiac SABR offers a final treatment option for high-mortality VT cases.
• Accurate image guidance is crucial for cardiac SABR due to proximity of organs at risk.

Purpose of the Study:

• To evaluate the inter-observer variability of different CBCT image matching techniques for cardiac SABR.
• To compare 3DoF and 6DoF image matching for cardiac SABR treatment delivery.
• To assess the reliability of automatic and manual image matching for cardiac SABR.

Main Methods:

• Four experienced therapeutic radiographers performed image matching on 40 CBCTs from 10 VT patients.
• Image matching was performed using 3DoF and 6DoF, including manual, automatic (heart structure), and PTV-adjusted matches.
• Inter-observer variability was quantified using modified Bland-Altman analysis.

Main Results:

• Automatic image matching demonstrated the smallest limits of agreement, indicating algorithm reliability.
• Manual adjustment to the PTV after auto-match is clinically viable for optimizing target coverage.
• 6DoF PTV matches (1.06, 1.24, 1.68 mm) showed less variability than 3DoF PTV matches (1.57, 2.06, 2.11 mm).

Conclusions:

• Six degrees of freedom (6DoF) CBCT image matching exhibits reduced variability, supporting its use in cardiac SABR delivery.
• Optimizing CBCT acquisition parameters and radiographer training are critical for successful cardiac SABR implementation.