(Tuesday, 7/31/2018) 1:45 PM - 3:45 PM
Room: Karl Dean Ballroom C
Purpose: X-ray induced acoustic (XA) imaging is potentially a real-time in vivo dosimetry technique. This work examines the feasibility of XA imaging for real-time monitoring using clinical relevant soft-tissue phantom: 1) the relative radiation dose delivered to the target region and 2) the dose delivery position in relation to the target tissue.
Methods: A prototype X-ray induced acoustic computed tomography (XACT) system, which had an immersion ultrasound transducer scanning around target sample along an entire circle, was used to acquire XA signals produced by a medical linear accelerator to form an XA image of irradiated phantom. Five lard cylinders imbedded into porcine gel were imaged explore the dose sensitivity of XACT by using physical wedges and the results were verified by comparison radiochromic film measurements. An integrated XA and Ultrasound dual-modality imaging system was established to locate the delivery position of the dose and track the motion of the target in real-time. The performance of dual-modality imaging system was tested with soft phantom experiments.
Results: An intensity difference as small as 2.9% in delivered dose region was measured from XA images (p=0.02). The intensities of XA images were highly correlated to the film measurements with an R² better than 0.986. The US modality in dual-modality imaging system can track the movement of pseudo tumor, while the XA images clearly distinguished the movement of the X-ray beam with an accuracy of 3mm. The displacement of X-ray beam and target can be real-time visualized in the fusion image of dual-modality.
Conclusion: XA imaging has a clinically acceptable sensitivity for monitoring relative dose. The combination of Ultrasound and X-ray induced acoustic imaging would allow real-time monitoring and alignment of the delivered X-ray beam and the intended target. This work shows potential for real-time in vivo dosimetry tool and image guided radiation therapy.
Funding Support, Disclosures, and Conflict of Interest: This work was supported in part by the University of Michigan Cancer Center fund G017459.
Keywords
Not Applicable / None Entered.
Taxonomy
Not Applicable / None Entered.
Contact Email
