AI cough analysis tool able to ‘hear’ Covid-19
MIT scientists have created an AI tool that can listen for Covid-19 in coughs.
The researchers employed AI engines developed to spot Alzheimer’s, using a dataset of over five thousand samples to train and test the machine learning algorithm.
Although other teams have been working on similar projects, none has achieved the success of the MIT project. The University of Cambridge’s Covid-19 Sounds attempt reported an 80 per cent success rate in July, but MIT’s model achieved sensitivity of 98.5 per cent, rising to 100 per cent for asymptomatic subjects.
The team said in a summary, “AI techniques can produce a free, non-invasive, real-time, any-time, instantly distributable, large-scale Covid-19 asymptomatic screening tool to augment current approaches in containing the spread of Covid-19. Practical use cases could be for daily screening of students, workers, and public as schools, jobs, and transport reopen, or for pool testing to quickly alert of outbreaks in groups.”
If it passed regulatory approval following further testing, the tool could be used to deliver daily nationwide Covid-19 testing.
The research was published in the IEEE Open Journal of Engineering in Medicine and Biology .