Advanced artificial intelligence techniques are being developed by a team of Murdoch University and PathWest researchers as a way to detect and diagnose cancer.
The Murdoch University team is working with PathWest anatomical pathologist Dr Jeremy Parry to train a specialised computer to detect abnormalities in lymph nodes that may or may not be malignant.
If successful, the team hopes their algorithms could help to discern malignant from benign changes, and detect nuanced early indicators of cancer by recognising patterns within data.
The computer will be trained using digitised whole-slide scans of lymph node tissues collected from WA patient samples.
Associate Professor Kevin Wong, who is one of four Information Technology researchers from Murdoch University involved in the project, said the aim was to assist and empower not replace the human pathologists who assess samples.
“The techniques we develop will work hand-in-hand with a human pathologist in validating and enhancing their decision about a case,” Professor Wong said.
The techniques will also provide a platform for humans and machines to learn from each other.
The second part of the project will see the team assess the value of using digitised whole-slide scans of tissue samples across the WA health system.
Currently the system for examining tissue samples involves putting them on glass slides so they can be viewed under the microscope. If a second opinion is needed, the slide must be physically transported to wherever the person giving the second opinion is based.
Scanning the slides to create a digital image means they can be sent and viewed instantaneously anywhere in the world.
“We are developing AI techniques to better learn and extract from the digital scans, the features that can help experts make better predictions,” Professor Wong said.
The project has been made possible by funding from the Department of Health’s Research Translation Project (RTP) program.The other Murdoch researchers involved in the project are Associate Professor Hamid Laga, Associate Professor Ferdous Sohel and their PhD student Ms Upeka Somaratne.