Radiotherapy can be a highly effective treatment for many types of cancers. A major impediment to achieving its full curative promise is the current delivery process, where typically the originally planned tumor area is exposed to a fixed pattern of ionizing radiation over time irrespective of target deformations, organ motion, or function. To avoid misses, geometric uncertainties in this feedforward process are dealt with by increasing the planning margin around the tumor, but of necessity result in unnecessary exposure of uninvolved tissue which can lead to debilitating toxicities. We hypothesize that the unwanted radiation dose to normal tissues could be significantly reduced by using a feedback system that would ?know? the shape and location of the tumor as well as the location and intensity of the irradiated dose during delivery. This framework would require the unique ability to simultaneously image the absorbed dose and the targeted tumor anatomy during radiation delivery, which is not possible with currently existing technologies. A known phenomenon in radiation physics is the generation of acoustic waves due to thermal expansion of a substance following the absorption of penetrating radiation. Detection of this radiation induced acoustic signal from clinical treatment beams has been recently demonstrated but has not been clinically realized. That signal exists ?for free? in real time as a consequence of the treatment beam. The signal can be measured with ultrasound detectors and processed to reveal the location and intensity of the deposited energy/dose. Furthermore, ultrasound technologies have also long been established for medical imaging and monitoring of tumor size, shape and location, without introducing ionizing radiation. Therefore, we propose to combine measurements of radiation acoustics and ultrasound imaging in an integrated system using advanced matrix array probes to determine in real-time the volumetric delivered radiation dose with respect to that day?s tumor shape and location, and ultimately to optimize tumor targeting via online feedback. The system will be optimized in phantoms and preclinical models. Then, its feasibility and versatility will be tested for treatment of tumors in the liver and the pancreas, two aggressive cancer sites where misplaced dose due to deformation and physiological motion not only compromises tumor eradication but also affects vital functions in the patient and subsequent treatment outcomes. Impact statement:
We aim to implement new, safe, simple, cost effective technology and methods for online guidance of radiotherapy delivery that can provide simultaneous tumor tracking and dose compensation capabilities. These technologies will be evaluated in a pilot clinical study of liver and pancreatic cancers to demonstrate feasibility and potentials for translation. If successful, this feedback technology will have a significant impact on personalizing radiotherapy delivery and achieving optimal treatment outcomes.

Public Health Relevance

Uncertainties in the current offline feedforward processes of radiotherapy delivery are limiting its tumor curative effectiveness, or increasing the exposure of uninvolved normal tissue resulting in short- and long-terms side effects, or both. Therefore, this proposal will investigate an online feedback, ultrasound-based system to better guide the delivery of the radiation beam onto the tumor at the time of treatment. This system integrates X-ray acoustics and ultrasound imaging, which uniquely enables simultaneous imaging of the deposited dose and the targeted tumor anatomy during radiation delivery. The system will be evaluated for measurement accuracy and motion compensation using phantoms and animals; and for clinical use for tumors in the liver and pancreas.

National Institute of Health (NIH)
National Cancer Institute (NCI)
Method to Extend Research in Time (MERIT) Award (R37)
Project #
Application #
Study Section
Radiation Therapeutics and Biology Study Section (RTB)
Program Officer
Obcemea, Ceferino H
Project Start
Project End
Budget Start
Budget End
Support Year
Fiscal Year
Total Cost
Indirect Cost
University of Michigan Ann Arbor
Schools of Medicine
Ann Arbor
United States
Zip Code
Tseng, Huan-Hsin; Wei, Lise; Cui, Sunan et al. (2018) Machine Learning and Imaging Informatics in Oncology. Oncology :1-19