Using AI to predict cancer risks

The team behind the Cancer Risk Predictor: (from left) DataRobot data scientist Yeo Hwee Theng, DataRobot sales director Lim Kean Chye, SGH colorectal surgeon Ronnie Matthew, and Ms Allyson Ngoo, lead business analyst at Integrated Health Information
The team behind the Cancer Risk Predictor: (from left) DataRobot data scientist Yeo Hwee Theng, DataRobot sales director Lim Kean Chye, SGH colorectal surgeon Ronnie Matthew, and Ms Allyson Ngoo, lead business analyst at Integrated Health Information Systems.PHOTO: LIANHE ZAOBAO

Up to a third of colorectal cancer patients may experience a recurrence of the illness, said Dr Ronnie Matthew.

To predict the risk of cancer recurrence and survivability, the colorectal surgeon at Singapore General Hospital (SGH) and his teamconceptualised an artificial intelligence (AI) system.

The Cancer Risk Predictor pools local population data and analyses it to make such predictions, even identifying individuals who are at high risk of the disease recurring.

Said Dr Matthew: "We have a huge data set here at SGH, since we have the largest department in Singapore for colorectal cancer, and we can make use of that."

He said a shortage of clinicians, as well as clinical uncertainty and dilemmas, may compromise patient safety and reduce job satisfaction for these professionals. With the AI system, Dr Matthew said false positives and other risks cancer survivors face can be reduced.

A version of this article appeared in the print edition of The Straits Times on July 18, 2018, with the headline 'Using AI to predict cancer risks'. Print Edition | Subscribe