UW researchers use AI to diagnose different diseases
Posted Jan 30, 2024 08:00:23 AM.
Last Updated Feb 5, 2024 04:59:17 PM.
A new method to diagnose diseases using artificial intelligence (AI) is coming out of the University of Waterloo – with the goal of aiding doctors and clinical institutes when evaluating patients.
A team of researchers out of the local university as well as McGill University and the National Research Council of Canada was led by the Canada Research Chair in medical imaging and AI, Dr. Alexander Wong.
“We actually created a new system called Trustworthy Deep Learning Framework for Medical Imaging Analysis. (TRUDLMIA) This framework uses a number of different strategies that we’ve come up with that allows us to build AI that’s not only high performing but also trustworthy,” said Wong.
Wong said it can be used for a variety of diseases like COVID-19, pneumonia, and melanoma.
The new system aims to support healthcare professionals by enhancing computer-aided diagnosis.
The challenges currently with existing methods are mainly about trustworthiness and how trustworthy are the results being given to clinicians. There can also be challenges in interpreting existing data.
“Our key thing is, at the end of the day, when a doctor is trying to make a diagnosis they want to use systems they can trust and if there is any kind of hesitation then we’re not going to see wide spread adoption for AI for clinical support,” said Wong.
“The predictions it provides are a lot more aligned and trustworthy then some of the other systems that are out there.”
The system is currently still in its testing phases, but Wong said they hope to release it soon to doctors or clinical professionals. They are working on ways to present information better so doctors can interact with the system easily.
“Our underlying goals is the hope is that this system will be adopted in different clinical institutes by doctors to be able to support them in what they need to do which is provide a great level of care for their patients and also by having AI we hope that it’s able to help them not only better, but better in terms of seeing more patients.”