Therapeutic Medical Physics Research Lab
Medical Physics Graduate Program


Chihray, Liu, PhD; Faculty Member

Under the direction of Dr Chihray Liu, the Therapeutic Medical Physics research lab in the Department of Radiation Oncology is engaged in a variety of projects that are focused on cancer treatment using radiation.

These treatment techniques include:

  • High energy (MV) external photon beams
  • Low energy (OV) external photon beams
  • Heavy charged particle (Proton) external beams
  • Light charge particle (Electron) external beams
  • Brachy-therapy (Isotope or Micro-X-Ray Tube)

Projects

Clinical Workflow Optimization (Drs Liu, Maloney, Yan)

The main mission of a therapeutic physicist is to provide a safe treatment environment and high quality of care for cancer patients. Workflow optimization includes (1) providing an accurate and efficient system of communication between different teams in the radiation oncology department that will result in the best possible quality of patient care; and (2) streamlining quality assurance procedures for patient treatment devices such as the linear accelerator. Development of these workflow procedures is of tantamount importance in providing a highly sophisticated, streamlined radiation oncology department. We are also involved in additive manufacturing methodologies, including 3D printing, to improve treatment optimization and delivery.

Prototype Detector Development and Imaging Applications (Drs Maloney, Samant)

Imaging is a critical component of image-guided radiation therapy. While much work in IGRT has focused on image reconstruction algorithms and applications in adaptive therapy, much work in the development of suitable detectors remains. In collaboration with UF Nuclear Engineering, current activities include development of proton portal imaging detector based on exit dose imaging. Our prototype system involves a CCD camera system using LiF/ ZnS scintillation screen. New scintillation materials are being investigated for improved imaging signal performance. Time of flight detection to separate neutron and xray signals are also being investigated. A Monte Carlo model is also being developed to optimize detector geometry and performance limits in a high neutron scatter environment in proton therapy. The goal is to provide accurate visualization of proton collimation for QA and in vivo beam delivery. An ancillary project is the development and testing of He4 and Cs2LiYCI6 (CLYC) scintillation detectors capable of neutron dose and spectra measurement. These detectors can be used for measurements at proton therapy and nuclear facilities for more accurate radiation protection calculations.

AI Strategies for IGRT (Drs Samant, Yan, Sun, Wu)

GPU based AI has enabled computational clinical medicine. We are using the UF HiPerGator cluster for investigating radiation gating strategies for use in LinacMR systems based on high dimension predictive imaging, and auto-segmentation of moving targets in treatment planning for conventional linacs. Our work builds on traditional AI models such as LSTM and UNET, including NVIDIA developed MONAI package, to include hybrid models utilizing optical flow and diffusion models. Predictive imaging seeks to overcome the latency between target position during image acquisition and actual target position when image is displayed to ensure more accurate radiation delivery for moving targets, especially complex motion/ distortion in abdomen. Auto-segmentation allows one to track target motion and deformation during irradiation, as well as generate probability-based target volumes that ensure full irradiation of moving targets during a treatment session.

Application of Machine Learning in Radiotherapy (Drs Yan, Liu)

Machine learning has great promise for applications in radiotherapy, ranging from diagnosis, image analysis, treatment design to follow-up. The current goal of the physics team’s research in this area is to leverage the power of ML to address challenging issues in radiotherapy treatment design and quality assurance. The team’s effort focuses on ML research that can impact our current practice and improve patient treatment quality and outcome.

Dose Calculation/Plan Optimization (Drs Liu, Li)

Intensity-modulated radiation therapy and volumetric-modulated radiation therapy (VMAT) represent one of the most significant technical advances in radiation therapy since the advent of the medical linear accelerator. It allows the clinical implementation of highly conformal nonconvex dose distributions. However, these advances do not come without a risk. IMRT is not just an add-on to the current radiation therapy process; it represents a new paradigm that requires the knowledge of multimodality imaging, setup uncertainties and internal organ motion, tumor control probabilities, normal tissue complication probabilities, three-dimensional (3-D) dose calculation and optimization, and dynamic beam delivery of non-uniform beam intensities. Among all those factors, our group’s research interests focus on how to improve the dose computation accuracy while maintain the computation efficiency during IMRT/VMAT planning process.

Quality Assurance (Drs Liu, Li, Sun, Yan, Schwarz, Wu)

Quality assurance is essential in the safe and effective delivery of radiation treatment. Our group has collaborated with industry leaders in developing innovative ways to streamline the QA process. Both commercial products as well as in-house developed methods have been in use for periodic machine QA and patient-specific QA. We are constantly reviewing our QA programs to ensure safe radiation delivery and to increase QA efficiency.