Date/Time
Date(s) - 10/06/2022
1:55 pm - 2:55 pm
Location
Rhines Hall 125
Categories
Deanna Pafundi, Ph.D.
Associate Program Director
Mayo Clinic Physics Residency Program
Dr. Deanna Pafundi was born in St. Petersburg, FL and grew up in Zephyrhills, Florida. Spent undergraduate (Nuclear and Radiological Engineering, Summa Cum Laude, 2000-2004) and graduate school (Nuclear and Radiological Engineering Sciences under a DOE Grant, 2005-2009) days at the University of Florida. Graduate work was under Dr. Wesley Bolch working on the image-based pediatric hybrid computational phantom models. Worked at the Crystal River Nuclear Power Plant in summer of 2003 during their fuel reload. Worked for the Transuranium and Uranium Registries in the Summer of 2006. Accepted into the 3 year residency program at Mayo Clinic in Rochester, MN from 2009-2012. Joined the staff in 2012 after residency and stayed there until June 2019. Received the Mayo Fellows’ Association Honor of Excellence in Teaching in 2019. Opportunity to move back home to Florida and work at the Mayo Clinic in Florida. Started working there in July 2019. I am the associate program director for the physics residency program and at Mayo Clinic Florida I am also very active in AAPM. I have served on the Working Group for Imaging in Treatment Planning under the Therapy Physics Committee since 2015. 2016 vice chair, and now chair since 2017. AAPM Task Group reviewer for 6 years. Also, a member of both the AAPM Working Group for Treatment Response Assessment and Therapy Imaging Subcommittee. I am currently being reviewed for academic promotion into associate professor. I continue to research advanced imaging applications for radiotherapy applications, especially with the upcoming proton and carbon ion therapy center. I also maintain a heavy clinical practice with focus on AI applications in radiotherapy.
Abstract
Positron Emission Tomography (PET) image with amino-acid tracers such as 18F-DOPA has the ability to capture the cellular activities of gliomas with strong tumor-to-normal tissue signal with superior sensitivity to tumoral infiltration and aggressiveness. The level of the Standardized Uptake Value (SUV) has been utilized to define heterogeneity of the tumor and to guide radiation therapy target delineation for improved outcomes. Furthermore, complex signals can be extracted from the distributions of the SUV histogram and texture patterns to predict treatment response. These rich signals are highly dimensional and mineable but surrounded by a noisy system beyond the capability of conventional quantitative analysis methods.
Radiomics, an emerging field based on machine learning methods, has become a powerful tool for extracting quantitative signals from this noisy environment and studying the correlation between various characteristics of the SUV distribution.
This presentation will take you through over a decade’s worth of data collected from several prospective trials at Mayo Clinic integrating FDOPA PET into radiotherapy treatment planning, surgical resection/biopsy targeting, and treatment response predictions.