Ground-breaking research by computer scientists at RMIT University has developed an artificial intelligence model that can hear the effects of Covid-19 in the sound of a forced cough, even when people are asymptomatic.
While the idea is not new, scientists at Cambridge University and MIT have been investigating the concept, the work at RMIT’s School of Computing Technologies could pave the way to a new generation of diagnostic mobile phone apps as part of a pre-screening process.
Study lead author Dr Hao Xue, a research fellow at RMIT, said they’d overcome a major hurdle in the development of a reliable, easily-accessible and contactless preliminary diagnosis tool for Covid.
“This could have significant benefit in slowing the spread of the virus by those who have no obvious symptoms,” he said.
“A mobile app that can give you peace of mind during community outbreaks or prompt you to seek a Covid test – that’s the kind of innovative tool we need to better manage this pandemic.
“It could also make a significant difference in regions where medical supplies, testing experts and personal protective equipment are limited.”
Xue said their method could also be extended for other respiratory diseases.
“With just a little tweaking and suitable data we could use this to test for Tuberculosis or other respiratory illnesses, or even design it for combined multi-diseases detection or classification system,” he said.
Their research is supported by Australian Research Council.
Major advance in AI training
While it’s not the first cough classification algorithm developed, the RMIT model outperforms existing approaches and has another major advantage that makes it more practical to use across different regions – the way it learns.
Study co-author Professor Flora Salim said previous attempts to develop the technology, including at MIT and Cambridge, relied on huge amounts of meticulously labelled data to train the AI system.
“The annotation of respiratory sounds requires specific knowledge from experts, making it expensive and time-consuming, and involves handling sensitive health information,” she said.
“Using a narrowly-targeted data set – such as cough samples from one hospital or one region – to train the algorithm also limits its performance outside that setting.”
Salim said it was this limitation that had proven a challenge for this technology’s practical application in the real world, until now.
“What’s most exciting about our work is we have overcome this problem by developing a method to train the algorithm using unlabelled cough sound data,” she said.
“This can be acquired relatively easily and at larger scale from different countries, genders and ages.”
The team are open to collaborating with potential partners on developing the technology and expanding its application for a range of respiratory diagnostic tools.
Their study, ‘Exploring Self-Supervised Representation Ensembles for COVID-19 Cough Classification’ is available now as a pre-print. It will be presented at prestigious data science conference KDD 2021 in Singapore this August.