Automated Robotic Targeting System for In-Vivo Histotripsy Treatment
M Wagner*, S Periyasamy, A Kutlu, J Swietlik, T Ziemlewicz, M Speidel, F Lee, P Laeseke, University of Wisconsin - Madison, Madison, WI
Presentations
TU-GH-BRB-5 (Tuesday, 7/12/2022) 1:45 PM - 3:45 PM [Eastern Time (GMT-4)]
Ballroom B
Purpose: Histotripsy is a noninvasive, nonionizing, and nonthermal ultrasound-based focal therapy undergoing clinical trials for the treatment of liver tumors. Tumor targeting is currently performed using an integrated diagnostic ultrasound probe. However, not all tumors can be visualized on ultrasound and manual targeting results in operator-dependent accuracy. This study investigates a C-arm guided targeting system, where the tumor can be segmented on a pre-treatment cone-beam CT (CBCT) and the histotripsy therapy transducer is then aligned on the target automatically using the integrated robotic arm.
Methods: The robot was calibrated to the C-arm prior to the procedure by moving the transducer to four preprogrammed poses and acquiring 2D fluoroscopic images, which were used to estimate the 3D pose of the therapy transducer for each location. The robot was calibrated to the C-arm using a dual quaternion approach to estimate the transformations between C-arm, robot and transducer. The therapy transducer can then be automatically focused on any point selected in CBCT images. A porcine study (n=3) was conducted, where targets in the liver were selected in pre-treatment contrast-enhanced CBCTs and automatically targeted. Spherical treatments (2.5-cm diameter) were created during continuous ventilation of the animal. A post-treatment contrast-enhanced CBCT was acquired, and the effective treatment zone manually segmented to evaluate the accuracy.
Results: The mean (STD) Euclidean distance between the center of the effective treatment zone and the selected target in the pre-treatment CBCT was 4.40±0.45mm. The mean absolute difference between the predicted and actual treatment diameter was 2.72±0.13mm. The overlap between the predicted and actual treatment zones was 93.87±3.35%.
Conclusion: Despite respiratory motion during treatment, the selected targets were successfully targeted and treated. This would allow for the treatment of tumors not visible under ultrasound imaging, reduce operator dependency, and open up histotripsy to a larger cohort of patients.
Funding Support, Disclosures, and Conflict of Interest: Funding was received through a sponsored research agreement from HistoSonics, Inc.
Keywords
Target Localization, Image-guided Therapy, Cone-beam CT
Taxonomy
IM/TH- Cone Beam CT: Development (New Technology and Techniques)
Contact Email