Graduate Research Assistant - Medical Physics

Appointment of one (1) candidate with interest in becoming a board-certified medical physicist to conduct a Ph.D. level research project in the Department of Radiation Oncology at the University of Iowa Hospitals and Clinics. The successful candidate will be expected to complete his or her Ph.D. requirements on an exciting Artificial Intelligence-based MRI-guided radiotherapy treatment planning project. The overall goal of this research project is to develop a reinforcement learning neural network [1] to predict the Elekta Unity machine parameters for an optimal MRI-guided radiotherapy treatment plan. The machine parameters to be predicted include multileaf collimator (MLC) leaf positions and monitor unit (MU) values for each control point in a step-and-shoot delivery. Once the machine parameters are determined at each control point, the GPUMCD Monte Carlo algorithm [2] in the Elekta Monaco treatment planning system will calculate the treatment plan dose. An investigation will also be performed to development a framework to predict alternative plans optimized to different endpoints such that the user can select between multiple plan options based on a single image. An example would be selecting between a plan that prioritizes planning target volume (PTV) coverage or a plan that prioritizes rectum sparing for a prostate treatment.

The research project will require extensive python programming and the generation of a reinforcement learning network using machine learning libraries (pyTorch, Tensor Flow, etc.). The student will gain extensive experience in machine learning, treatment planning, dose calculations, and MRI guided radiotherapy. These skills will set the student up for a future in both academic and clinical medical physics.

              

The student will be admitted through the Biomedical Engineering Graduate Program at the University of Iowa and will be supervised and funded within the Department of Radiation Oncology. The start date for this position is flexible but must be by the 2021 fall semester at the latest (August 23, 2021).

1. Candidate is required to have received a Master Degree in Medical Physics or equivalent field from a CAMPEP-accredited Medical Physics Graduate Program prior to the start date. Please clearly indicate this in your application.

2. The selected candidate is required to additionally apply to and be accepted in the Biomedical Engineering (BME) graduate program at the University of Iowa as a Ph.D. student to be appointed in this position. This BME application can be submitted outside of standard BME graduate program application deadlines but only after selection for this position. The candidate will be subject to all requirements of the BME graduate program. See https://bme.engineering.uiowa.edu/graduate-program for details on qualifications and how to apply.

3. It is highly desirable for the candidate to have strong interpersonal and communication skills and it is also desirable that they have completed at least two (2) computer science and/or computer engineering courses. Other courses relevant to the research topics outlined herein may be considered at the time of application.

4. It is highly desirable that the candidate have interest and some experience in computer programming, numerical algorithm development, and the integration of artificial intelligence/machine learning for treatment planning.

Percentage of appointment: 50% (full-time graduate student)

Length of appointment: 4-5 years

Desired skills: Computer programming proficiency in Python, C languages, and/or MATLAB, exposure to neural networks, and computational optimization.

Application process: (1) Up-to-date CV, (2) cover letter (statement of purpose), (3) three letters of recommendation, (4) college transcripts (MS transcript must be from CAMPEP accredited program). Letters of reference are to be emailed directly to Dr. Joel St-Aubin at joel-st-aubin@uiowa.edu.

             

The successful candidate will be accepted into the Medical Physics Graduate Track program at the University of Iowa in the department of Radiation Oncology. In addition to the high quality training received through the courses provided through the Biomedical engineering program, the Medical Physics Graduate Track program provides clinical hands-on experience doing medical linear accelerator and patient specific quality assurance through a clerkship program. Thus, through the medical physics graduate track program, the successful candidate will receive the training and experience (both research and clinical) in preparation for a CAMPEP accredited medical physics residency program.

The Physics Division of the Department of Radiation Oncology at the University of Iowa has fourteen faculty medical physicists, three medical physics residents, seven dosimetrists, an engineer, and two IT specialists. Our main facility houses a total of 5 external beam treatment units: three Elekta Versa HD systems, a Gamma Knife Icon stereotactic radiosurgery system, and an Elekta Unity MRI-linac. Additional technologies include a 3 Tesla MRI scanner, 4-D PET/CT scanner, mobile C-arm for fluoroscopy and kV cone beam CT, six degree-of-freedom robotic patient positioners, and multiple optical image guidance systems. Special procedures include 4D imaging/treatment planning/delivery, frame-based and frameless stereotactic radiosurgery, stereotactic body radiation therapy, total body and total skin irradiation, intraoperative radiation therapy, and an active and diverse brachytherapy program (HDR, prostate seed implants, eye plaque implants, etc). Additional technologies are available for support in our four partner sites, including Varian and Elekta linear accelerators. The Medical Physics Division is actively involved in translational and collaborative research.

The University of Iowa Hospitals and Clinics is a tertiary care, teaching, and research center located in Iowa City, Iowa. We are consistently named to U.S. News & World Report as one of “America’s Best Hospitals”. Iowa City, a Big Ten University town, offers a lively community with robust cultural, entertainment, and recreational activities, and is ranked as number 20 on Livability’s Best Places to Live (2020).

[1]         Mnih V, Kavukcuoglu K, Silver D, Graves A, Antonoglou I, Wierstra D, Riedmiller M, “Playing atari with deep reinforcement learning,” https://arxiv.org/abs/1312.5602

[2]        Hissoiny S, Ozell B, Bouchard H, Despres P, “GPUMCD: A new GPU-oriented Monte Carlo dose calculation platform,” Med Phys 38 (2011) 754-764