Calibration correction to improve registration during cone-beam CT guided histotripsy

Area of Science:

• Medical physics
• Biomedical engineering
• Oncology

Background:

• Histotripsy uses focused ultrasound to mechanically destroy tissue via cavitation.
• Current histotripsy is limited by visualization challenges, particularly bone or gas obstructions.
• Cone-beam computed tomography (CBCT) guided histotripsy enhances visualization for broader treatment application.

Purpose of the Study:

• To enhance the precision of CBCT-guided histotripsy.
• To calibrate the location of the therapeutic bubble cloud relative to the histotripsy robot arm.

Main Methods:

• A multi-layered agar phantom with alternating X-ray attenuation was used for calibration.
• An automated algorithm localized bubble clouds by analyzing pre- and post-treatment CBCT intensity differences.
• 3D offsets between theoretical and actual bubble cloud locations were calculated for robot arm calibration.

Main Results:

• The automated algorithm accurately identified bubble cloud locations from CBCT images.
• A multi-bubble cloud calibration approach (four clouds) yielded significantly lower mean absolute deviation in 3D offset measurements compared to single-bubble clouds.
• Statistically significant differences in 3D transducer offsets were measured among four different histotripsy transducers.

Conclusions:

• Analyzing multiple adjacent bubble clouds improves the accuracy and reproducibility of 3D offset measurements in histotripsy.
• The automated calibration method is accurate, efficient, and integrates seamlessly into existing histotripsy workflows.
• This technique expands the potential for accurate CBCT-guided histotripsy in clinical settings.