Meet Yujing Zou

Redirect to an updated page, click here!

Hello, my name is Yujing! I am currently a masters student at the McGill medical physics unit. Currently in Dr. Shirin Enger’s lab, my research interests lie in the intersection of deep learning, image processing & analysis, and outcome prediction modelling in medical physics.

I recently finished my first year of the CAMPEP-accredited Medical Physics M.Sc. program at McGill University. In 2020, I graduated from a joint major in physiology and mathematics with a minor in physics also at McGill. Throughout my degrees, I have been inspired and drawn to interdisciplinary research where mathematical modeling and computational tools are used to uncover problems in medicine. My current research interest surround the topic of deep learning-based outcome prediction modelling using multiple diagnostic imaging modalities such CT, MRI, histopathological images and Ultrasound. More specifically, despite combination of different treatment modalities, such as radiotherapy including brachytherapy, surgery, chemotherapy, chemoradiotherapy, and improvement of treatment protocols and imaging techniques used for management of cancer, it remains clinically challenging to predict which patients will benefit from which treatment combination. An accurate method for predicting a patient’s likelihood of response could reduce unnecessary interventions, lower healthcare costs, and reduce side effects. This prompted our objective to examine how pre-treatment patient characteristics, through imaging data, influence treatment efficacy as measured by post-treatment response, therefore to build a treatment outcome prediction model.

Previously during my undergraduate studies, I have also worked on non-linear dynamics problems in biology, as well as Radiomics and auto-segmentation algorithms on medical images. For example, 1) I investigated a type of discrete mathematical model called Cellular Automata (CA) model and compared its predictability in real-time excitable media cardiac cell wave propagation data with a continuous model called FitzHugh-Nagumo (FHN) model in Dr. Gil Bub’s lab. A Convolutional Neural Network (CNN) model integrated with the CA is developed to increase the model’s predictability at higher time steps. 2) Radiomics is a field examining correlations between diagnostic image features and treatment outcomes in radiation oncology. In Dr. Jan Seuntjens’s lab, I compared lymphadenopathy outcome prediction power between 2D (i.e. central tumor slice image) and 3D (i.e. whole tumor volume images) features from head and neck cancer dual-energy CTs (DECT) data at 21 energy levels from 87 patients’ DECT scans, and understanding how energy levels benefit different feature types.

Outside of academics, I am extremely passionate about songwriting and music production technologies, as well as learning new programming tricks and languages. My favourite band right now is Vulfpeck!

I am beyond excited to organize and lead this workshop series & Hackathon with my colleagues! I hope to see many of you soon!

To learn more about me, see my LinkedIn, ResearchGate, Twitter, SoundCloud, YouTube.

-Yujing

Our inspiration

Yujing and Luca both work on multi-modal image analysis using deep learning for various purposes from tumor auto-segmentation to outcome prediction using multiple diagnostic imaging modalities. They noticed that a few of their colleagues that were new to the field experienced some common challenges that they had experienced themselves: like how to work with DICOM files, getting started with practical deep learning programming in Python with little to no theory knowledge background and so on.  This prompted them to establish a workshop series to help kickstart peoples’ research projects involving medical imaging analysis and deep learning. We quickly discovered that there was interest that far exceeded our lab group within the McGill community.

Welcome to McMedHacks!

The students of Dr. Enger’s lab at the McGill Medical Physics Unit and the Lady Davis institute are offering a free, beginner-friendly workshop on how to use Python for deep learning-based medical image analysis. This is an amazing opportunity for you to learn about active fields of research from experts in the field and fellow students with research experience.

Our interactive workshop will teach you the fundamentals of medical image analysis, covering multiple imaging modalities including histopathology, CT, and MRI. You will learn how to use some of the most popular deep learning architectures, such as U-Net for tasks like automatic segmentation of organs or diseases. 

At the end of this workshop you will have the necessary skills to kickstart a research project in deep learning-medical image analysis. What better way to put your skills to the test than a friendly challenge? At the end of our workshop series, we will be hosting a McMedHacks hackathon, where you can collaborate with your peers, medical professionals, and AI experts in the field to complete a medical research themed image analysis challenge of your choosing. No matter if you are getting your hands dirty for the first time or would just like to get an edge for your research, our workshop has something for you.