2020 Summer Student Medical Physics Symposium

Wed. Aug. 5 10:00 AM - Wed. Aug. 5 01:00 PM
Location: via Zoom

On August 5, 2020 the University of Winnipeg will host the annual Medical Physics Undergraduate Summer Student Symposium. This year it will take place via Zoom from 10:00am - 1:00pm.

This is a great opportunity for students to present the projects they’ve been working on. Participating students will be given a 15 minute time slot to present (12 minute presentation + 3 minutes for questions). An evaluation committee will judge the presentations and issue two cash prizes for the best presentation. We also welcome Dr. Ivan Klyuzhin from the University of British Columbia who will be our invited guest speaker for the event.

If you are interested in attending, registration is required. Please RSVP your attendance to medicalphysicssymposium@gmail.com. A password protected zoom invitation will be emailed closer to the date.

 

PRESENTER	SUPERVISOR	TITLE
Melissa Anderson	Dr. Melanie Martin	"Mouse Bed for In Vivo MRI"
Kaihim Wong	Dr. Melanie Martin	"Moving toward live non diameter measurements in mice"
Cameron Russell	Dr. Andrew Goertzen	"Characterization of a Time-of-Flight PET Detector System"
Jordan Krenkevich	Dr. Stephen Pistorius	"Machine Learning for Suppressing the Skin Response in Breast Microwave Sensing"
Gabrielle Fontaine	Dr. Stephen Pistorius	"PET scatter image reconstruction with CNN machine learning"

 

Dr. Ivan Klyuzhin

University of British Columbia

Novel image analysis methods in PET imaging driven by machine learning: their development and applications.

The function of tissues in a healthy state and pathology can be assessed in-vivo using Positron Emission Tomography (PET) imaging. PET is based on the use of radiotracers - radiolabeled molecules that have a specific target or metabolic pathway inside the body. Numerous radiotracers exist that can tag different receptors and processes in tissues, for example the glucose analog 18F-flurodeoxyglocose (FDG). Recent advances in the field of machine learning, aided by rapid improvements in computing power and greater availability of data, are opening new possibilities for advanced PET image analysis. Although PET has been used in clinical practice and research for decades, there is an emerging recognition that PET images may contain a wealth of information that can be mined using high-throughput analysis methods and used to better understand disease origins or to devise personalized treatment plans. This talk will briefly summarize the recent developments in PET image analysis methods, and consider several examples of using machine learning in PET studies of neurodegeneration and oncology.