Prostate, lung cancer ID filter created at University of Waterloo

Alex Wong, a professor of system design engineering, has discovered unique bio-markers that could help radiologists identify prostate and lung tumours from medical images.

Prof. Alex Wong says digital 'fingerprint' will make it easier to identify cancer on MRI, CT scans

Wong said radiologists rely on their experience to help them interpret the images taken during MRI and CT scans. (University of Waterloo)

New research out of the University of Waterloo could take some of the guesswork out of reading MRI and CT scans.

Alex Wong, a professor of system design engineering, has discovered unique bio-markers that could help radiologists better identify prostate and lung tumours on medical images.

Diagnosing cancer through an imaging screening is a very difficult process. It's actually very tough.- Alex Wong, University of Waterloo

"You could treat it as creating a digital fingerprint for characterizing tissue," Wong said, one that can be compared to a patient's MRI or CT scan. 

"If this particular fingerprint matches that of, let's say, a malignant tumour, then a radiologist can look at this and say, 'Well, these are very similar. Maybe this is something we should look further into.'"

Wong says images taken by MRI and CT scans don't tell the full picture, and can be difficult to read. (Liz Henry/Flickr)

Creating a fingerprint

Wong developed his fingerprints by feeding hundreds of pages of medical images into a computer, and then designing a program that would look for similarities. In this way, he was able to identify the characteristics unique to images of certain tumours.

The pictures don't really give a full story.- Alex Wong, University of Waterloo

"We can then extract this information – this quantitative, biomark information – and put it in front of the radiologist, so that they have a much better way of analyzing whether or not it's cancer."

Wong said this discovery would change the way radiologists examine medical images. He says the current method involves a lot of guesswork, relying on experience and expertise rather than computer-assisted interpretation.

Wong says his fingerprint technology will make it easier for radiologists to accurately and consistently identify tumours. (Jeff McIntosh/Canadian Press)

Educated guesswork

"Radiologists are very skilled clinicians that do their best," he said. "The problem here is diagnosing cancer through an imaging screening is a very difficult process. It's actually very tough, despite the fact they have many, many years of experience looking at it. 

"That's mainly because just visually the pictures don't really give a full story. So, what ends up happening is a radiologist has to rely on their many years of expertise, knowledge of anatomy and tissue, and doing an educated guess as to what they're looking at."

'We might discover things that people haven't thought of before."- Alex Wong, University of Waterloo

The result is that many tumours are correctly identified, but some are not. 

"From a visual perspective, some of these differences between a malignant tumour and healthy tissue and benign tumours are so subtle that ... there are certain situations where just looking at the MRI itself you might not even be able to visually see it, and that leads to misdiagnosis."

Improving diagnosis

Wong said digital fingerprints of different tumours will make it easier for radiologists to accurately identify tumours, ensuring that there is more consistency in cancer diagnosis. 

The system is also helping scientists better understand the physiology of prostate and lung cancers. 

"By leveraging this wealth of information, what we might be able to do – and what we're starting to do – is we might discover things that people haven't thought of before," he said.

His system is being tested and refined at Sunnybrook Health Sciences Centre in Toronto, and he hopes it will be commercialised in the future.

Comments

To encourage thoughtful and respectful conversations, first and last names will appear with each submission to CBC/Radio-Canada's online communities (except in children and youth-oriented communities). Pseudonyms will no longer be permitted.

By submitting a comment, you accept that CBC has the right to reproduce and publish that comment in whole or in part, in any manner CBC chooses. Please note that CBC does not endorse the opinions expressed in comments. Comments on this story are moderated according to our Submission Guidelines. Comments are welcome while open. We reserve the right to close comments at any time.