RMIT University researchers train AI to detect early-stage prostate cancer

An AI model has been trained using CT scans of asymptomatic patients, with and without prostate cancer, to spot the smallest irregularities.
Written by Aimee Chanthadavong, Contributor
CT scans-gettyimages-141089990.jpg
Image: Getty Images

Researchers at RMIT University have developed an AI model that can detect early signs of prostate cancer by analysing routine computed tomography (CT) scans.

Developed alongside clinicians at St Vincent's Hospital Melbourne, the AI model has been trained using CT scans of asymptomatic patients, with and without prostate cancer, to spot the smallest irregularities, features of disease, and where exactly to look for them, without the need to manually crop the images.

In training the AI model, RMIT said the technology performed better than radiologists who viewed the same images and were able to detect cancerous growths "in seconds".

"We've trained our software to see what the human eye can't, with the aim of spotting prostate cancer through incidental detection," RMIT's Dr Ruwan Tennakoon said.

"It's like training a sniffer dog -- we can teach the AI to see things that we can't with our own eyes, in the same way a dog can smell things human noses can't."

While CT scans are typically not suitable for regular cancer screening because of high radiation doses, the AI solution could be used when men are scanning for other issues such as bone and joint problems, the study suggested.

Head of CT in diagnostic imaging at St Vincent's Hospital Melbourne Dr Mark Page believes early intervention for prostate cancer is key to a better health outcome.

"Australia doesn't have a screening program for prostate cancer but armed with this technology, we hope to catch cases early in patients who are scanned for other reasons," he said.

"For example, emergency patients who have CT scans could be simultaneously screened for prostate cancer.

"If we can detect it earlier and refer them to specialist care faster, this could make a significant difference to their prognosis."

RMIT added the technology has the potential to also be integrated with other diagnostic imaging equipment like MRI and DEXA machines.

For its next steps, the research team said it would now look for interested commercial partners to develop software to further integrate the AI technology with hospital equipment for possible clinical trials.

Google and AFL use AI to help fans kick footy like a pro

Image: Google

Google and the Australian Football League (AFL) have launched Footy Skills Lab, an online AI platform designed to help people sharpen their footy skills.

The platform uses Google AI and a smartphone's camera to coach and provides tips on how people can improve their skills through activities in ball-handling, decision making, and kicking across three levels of difficulty. The platform can recognise, track, and score the movement of a ball when it is within a camera's frame. 

The coaching tips are provided via short videos by professional Australian Football League Women and AFL Wheelchair players, including Carlton's Madison Prespakis and Richmond's Akec Makur Chuot.

To access Foot Skills Lab, users need a smartphone with an internet connection, something to prop their phone up, a football, and space to move. 

The activities have also been designed to be accessible for people with visual and hearing needs through audio prompts, closed captioning, and by providing alternative training movements for wheelchair users.

Once activities are completed, a scorecard will be generated, which users can share with family and friends.

Related Coverage

Editorial standards