Artificial Intelligence, or AI for short, has come a long way over the past decade. Although there’s still doubt it will cross over the bridge from super computing to personality in our lifetime. However, it has become an indispensable tool for doing tasks beyond the mental capacity of humans. MIT News thinks artificial intelligence could detect asymptomatic COVID infections. Now that would be a step forward, don’t you think?
How We Would Benefit If AI Could Detect Asymptomatic Infections
People with asymptomatic COVID-19 don’t display the signature signs that make the disease readily identifiable. They could super spread the virus without either party knowing it. The only way to detect it is a physical test, but they may have passed the infection on beforehand.
Now if AI could detect asymptomatic infections, then this could diminish the spread of the pandemic. That’s according to Brian Subirana, a research scientist in MIT’s Auto-ID Laboratory and his colleagues. They published a paper on a working model on September 29, 2020. They say this has 98.5% Covid-19 sensitivity, and 94.2% specificity.
The researchers wondered if a forced cough over a cell phone would identify both symptomatic and asymptomatic COVID infections. Then they assembled a database of 5,320 infected volunteers who sent recordings to their website.
How Brian Subirana and His Colleagues Developed Their Model
Our un-artificial intelligence did not cope well with their terminology. However, there’s a link below to their research paper if you would like to know more. They believe their work provides a practical daily screening tool for students, co-workers, diners out, and so on.
While our human systems can’t detect the subtle nuances of COVID-19 coughs, the research team trained their supercomputer to recognize them. This embraced several strings that analyzed vocal cord strength, frustration as opposed to happiness and calm, and changes in lung and respiratory performance.
Then they added an overlay to assess muscular degradation that’s also a common biomarker for the disease. They proved their hypothesis. And they believe ‘pandemics could be a thing of the past. If pre screening tools are always on in the background and constantly improved.’
Related
Asymptomatics Carried Same Amount of Virus
New Definition of Close COVID-19 Contact
Preview Image: Overview of AI Discriminator
