Skip to main content

Concordia researcher earns double recognition for her efforts on cancer prediction and diagnosis

Parnian Afshar is awarded a 2021 Borealis AI Global Fellowship
May 31, 2021
|
Parnian Afshar: “With the advances in machine learning and deep learning in particular, we can assist radiologists’ image inferring and diagnosing of diseases at earlier stages.”

Parnian Afshar, a PhD student at the Concordia Institute for Information System Engineering (CIISE), is working in the emerging field of medical imaging known as radiomics to increase the accuracy and prediction of cancer detection.

Radiomics uses deep-learning techniques to extract features from medical images that are otherwise not visible to the naked eye, thus enabling doctors to diagnose tumours earlier.

Afshar’s innovative work has just been awarded a Borealis AI Global Fellowship to further support her research in the 2021 winter and summer terms. The Royal Bank of Canada established the Borealis AI research institute, which promotes the latest developments in artificial intelligence and state-of-the-art machine learning.

Afshar was one of 10 graduate students selected for the $10,000 fellowship, which corresponds to a six-and-a-half month tenure through summer 2021.

Afshar received recognition earlier this year by winning the March Prix Relève étoile Louis Berlinguet for her paper, “3D-MCN: A 3DMulti-scale Capsule Network for Lung Nodule Malignancy Prediction,” published in Scientific Reports. The Fond de Recherche du Québec awards the Prix Relève étoile Louis Berlinguet monthly to promote outstanding research in the “Nature et technologies” category.

‘This field can lead to life-changing outcomes’

Tell us a little bit about your research.

Parnian Afshar: I started my PhD under the supervision of Dr. Arash Mohammadi in 2017. CIISE is a multidisciplinary department that fits my research at the intersection of engineering and medical imaging.

I started my research by developing radiomics and deep learning-based models for cancer prediction/diagnosis. In the past year, we realized that the same models could be modified to address the problem of COVID-19 diagnosis, which is of paramount importance in breaking the chain of infection.

Why is your research attractive to Borealis AI?

PA: My research is aligned with Borealis AI’s aims in that it involves exploring and developing machine learning and deep-learning techniques for the crucial application of medical imaging.

Borealis AI supports students who are doing research in machine learning and can demonstrate their ability to introduce real-world applications of this field.

Why is this field important for society?

PA: Cancer is the leading cause of death in Canada, partly because patients are diagnosed at advanced stages. Medical imaging is key to a cancer diagnosis. With the advances in machine learning and deep learning in particular, we can assist radiologists’ image inferring and diagnosing of diseases at earlier stages.

We can also determine the type of the disease and predict the treatment outcome. How this field can lead to life-changing outcomes is what motivates my research.

Is there a lot of AI work being done in your field at Concordia and in Montreal?

PA: Montreal is certainly a hub of AI in Canada, and Concordia has started supporting extensive research on this field. Concordia has a lot of scientists working at the edge of AI and machine learning, for sure.

What else will you be working on in the coming months and years?

PA: I want to continue my work with histopathological images and rich embedded information. Processing these images is challenging due to their very large resolution. Advances in AI can help to extract useful information from these images and facilitate disease diagnosis.
 

Read Parnian Afshar’s award-winning research paper.



Back to top Back to top

© Concordia University