How Canadian AI is kick-starting new discoveries in health and science

Artificial is poised to rocket the fields of and health into the , and Canada could play a pivotal role.

Canada’s space agency is exploring ways to equip future with AI. The Canadarm2, which astronaut David Saint-Jacques will use to perform the first every “cosmic catch” of a space rocket loaded with supplies for the International Space Station, has been developed with AI capabilities so that it can perform certain tasks autonomously.

NASA has also equipped its new Mars rover with AI so that it uses its smarts to avoid hazards like the soft sand that the rover Spirit got stuck in.


READ MORE:
Canadarm2 to perform ‘cosmic catch’ when SpaceX rocket arrives at ISS with supplies

But while robotics is an obvious launching pad, AI is poised to transform health care in ways never thought possible.

Toronto-based company Winterlight has developed a tool that it says will enable doctors to assess Alzheimer’s in elderly patients more efficiently than the current method of comparing written tests taken months apart.

“What we’re trying to do with AI is trying to catch people in their natural sort of environment or their natural behaviour,” said Winterlight’s Frank Rudzicz. “So a lot of our early work involves a snapshot of 45 seconds of speech, and we’re able to get 92 per cent accuracy.”

 

The company’s program is being used in several nursing homes as part of a pilot project. Using a tablet, researchers show seniors pictures and ask them to describe the images.

Winterlight bases its data on the seniors’ speech patterns. And while the company is being careful to say it is not “diagnosing” dementia at this point, it is confident that the technology is “assessing” patients well enough to give doctors an extra tool that can enable them to start treatment earlier.

Also Read:  Google announces winners of its 2019 Science Fair

We also have to not hand too much autonomy over to the machines,” said Rudzicz. “I don’t think it’s going to happen anytime soon. But we have to make sure that when the AI says you have this probability of having this disease, we’re able to open the black box and say: ‘This is why the decision was made.” Because AI is not perfect; it makes mistakes.”

You might also like More from author

Comments are closed.