According to many public health experts, testing for COVID-19 is key to stopping the pandemic. Testing leads to quick identification, isolation and treatment of COVID-19 cases, including those who came into contact with infected people. It stops the spread of the virus.1
Unfortunately, widespread testing in the U.S. has been limited by logistic and personnel problems, lab capacity, cost and the availability and reliability of the tests themselves. That’s why researchers at several universities and medical centers are exploring the use of smartphones to detect COVID-19, through users’ speech and coughs.
Apps designed to discern the presence of COVID-19 vocally are underway at Carnegie Mellon University in Pittsburgh,2 the École Polytechnique Fédérale de Lausanne in Switzerland3 and the University of Cambridge in the U.K.4
Many of the app creators emphasize that their immediate priority is less an accurate diagnosis for users than collecting enough voice samples from those with and without the virus to develop an algorithm that can accurately diagnose COVID-19 cases in the future.5 They say their devices are works in progress.
While users help build the vocal signature database there is also something in it for them. Users of Carnegie Mellon’s COVID Voice Detector app6 will receive “results, which are presented as a sliding scale to suggest how concerned we should be,” according to Digital Trends.7
How Can Someone ‘Sound’ Sick?
Doctors have reported that COVID-19 exerts unique and distinctive effects on the way patients speak and cough, which forms the basis for the voice-driven AI diagnostic tests.8 According to Bhiksha Raj, a Carnegie Mellon professor who worked on the COVID Voice Detector project:9
“Coughs of people with infected lungs sound different. The elongated vowels like ‘aaaa’ not only sound different, but the duration to which they can be kept up is reduced. When you speak a long string of numbers, followed immediately by the alphabet, you’re being asked to speak for an extended period of time.
COVID patients tend to get breathless, have runny noses, somewhat sore throats, and tire easily. All of these will affect the way they speak — particularly when they must speak for longer durations.”
Professor Cecilia Mascolo at the University of Cambridge, U.K., where the COVID-19 voice driven app is part of a project called EAR,10 agrees that the voice of someone with the virus can be different. She told Horizon magazine:11
“[S]ome research papers … indicate the cough that comes with Covid-19 has some specific features – it has been described as being a dry cough (with some specific distinguishing features that allow it to be identified).
Having spoken to doctors who are treating people (Covid-19 patients) in hospitals, there may be some changes to their voice, their patterns of breathing or the way they catch their breath as they talk like they are exhausted. We are looking at all of these things by asking participants to record themselves breathing and reading sentences out loud.”
Because of the vocal aberrations, the voices and coughs of users can be used diagnostically. After downloading the Carnegie Mellon COVID Voice Detector app, users are instructed to cough several times, pronounce a number of vowel sounds and recite the alphabet into the microphone of their smartphone or a computer.12 They then receive a score that indicates the likelihood of a COVID-19 infection.
Limits to a COVID-19 Voice Test
The Carnegie Mellon researchers who developed the COVID Voice Detector caution that the tests, though quick and easy, cannot be a substitute for medical advice; results must be verified by medical professionals and/or backed by the CDC. According to Raj:13
“The app’s results are preliminary and untested. The score the app currently shows is an indicator of how much the signatures in your voice match those of other COVID patients whose voices we have tested. This is not medical advice.
The primary objective of our effort/website at this point of time is to collect large numbers of voice recordings that we could use to refine the algorithm into something we — and the medical community — are confident about.”
Raj urges users of the COVID Voice Detector app to not make medical decisions based on results because they could be endangering themselves.14
Before a user of the Carnegie Mellon app creates their profile, they have to check a box on the website that says, “I understand that this is an experimental system which is still under development … It is not a diagnostic system. It has not been verified by medical professionals. It is not FDA or CDC-approved, and must not be used as a substitute for a medical test or examination.”15
Still, such disclaimers may not be enough. Rita Singh, a computer science professor at Carnegie Mellon who has worked with vocal signature apps to detect diseases like Parkinson’s, told the BBC:16
“It doesn’t matter how many disclaimers you put up there — how clearly you tell people that this has not been medically validated — some people will take the machine as the word of God.”
Google has also restricted apps like the COVID Voice Detector17 to make sure there is no misinformation. False positives could waste precious resources and false negatives could have fatal consequences, admits Singh, so great caution must be taken.18
The Pros and Cons of Artificial Intelligence
From FitBits to mobile cardiac-monitoring devices, many are increasingly relying on instruments for health measurement and feedback. Technology and machines rather than clinicians have replaced a lot of our health monitoring information and even our own self-appraisals.
According to Mascolo, who is working on a COVID-19 voice detector at the University of Cambridge:19
“The technique of listening to the body is actually very difficult for humans to acquire without a lot of training, but machines are much better at it. Artificial intelligence technologies like machine learning can identify features or patterns in a sound that the human ear cannot …
Our big vision … is for machine learning algorithms to be linked to wearable devices and smartphones so it can automate the diagnosis of disease through sound. Most of us might have a doctor listen to our body’s sounds periodically, but what happens if you have something that can listen to you continuously. It could be a new form of diagnostic.”
But, there is a darker side to smartphones and AI capturing your personal data — the potential for invasion of privacy and unabashed surveillance.
In March 2020, the White House Office of Science and Technology Policy began assembling a task force of tech and AI companies to “develop new text and data-mining techniques that could help the science community answer high-priority scientific questions related to COVID-19.”20
The task force of 60 included tech giants like Facebook, Google, Twitter, Uber, Apple, IBM, Microsoft and Amazon.
The risks of such companies data-mining your texts and cellphone data range from targeted marketing based on your personal data to literal spying on someone’s whereabouts and personal contacts. The scepter of “vaccination certificates” and totalitarian national tracking systems also cannot be ruled out.
The voice app makers seem to be aware of the risks. A notice on the Carnegie Mellon COVID Voice Detector reads:21
“We take your privacy seriously … We do not ask you for your name, address, or any other identifiers, although you may elect to provide an email address if you choose to do so for password recovery purposes.
Will Voice-Based COVID-19 App Help the Pandemic?
Not all technology and medical experts laud the work on voice-based COVID-19 apps underway at academic centers like Carnegie Mellon University, the University of Cambridge and École Polytechnique Fédérale de Lausanne. Ashwin Vasan, a professor at Columbia University Medical Center, articulated his doubts to Futurism:22
“‘Despite what could be a well-intentioned attempt by a bunch of engineers to help during this crisis, this is not exactly the messaging we want to be out there,’ he cautioned.
‘That somehow there is a nifty new tool we can use to diagnose coronavirus, in absence of the things we really need much more of, actual test kits, serologic testing, PPE for frontline healthcare workers, and ventilators for critically ill patients.
Let’s keep the focus on that, especially when our leaders in Washington seem unable to meet those most basic needs … Anything else is just a distraction.'”
The AI voice-based apps also show the difference in style between the slow-moving science/medical world and the fast and loose culture of Silicon Valley, wrote Digital Trends:23
“For some time, it has been evident that there is a conflict between Silicon Valley’s famous mantra of ‘move fast and break things’ and the real world’s need for empirical proofs, proper testing, and verification.
Put simply, technologists’ belief in hacking together tools and pushing them out into the world, safe in the knowledge that agile methodology will let them tweak and hone until the finished product works as well as hoped, doesn’t always mesh well with … peer review and clinical testing. This challenge is particularly pronounced in the field of medicine.”
Developers Acknowledge App Challenges
Developers of voice-based COVID-19 apps acknowledge the limits of the diagnostic tools, especially as they are in the early stages. According to Raj, the COVID Voice Detector “is, at best, currently a triage tool that could prompt people to seek out further medical opinions.”
But the app developers also see positive and worldwide public health applications if the algorithms are proven and the inventions succeed. Raj told Digital Trends:24
“‘If this works, we will have a very simple and easy way of monitoring millions of people,’ Raj said. ‘Not only can we get instantaneous evaluations, but also look at longitudinal trends among subjects who use it repeatedly.
This could provide a way tracking health outbreaks in general in future — particularly ones that affect voice.’
For this reason, Carnegie Mellon plans to share the data it gathers with other researchers around the world, encouraging them to work on it either collaboratively or independently. The one catch? ‘We will need the assurance from anyone who uses our data that they will not be protecting or commercializing any IP that comes out of it either,’ [said Raj]”
Others at Carnegie Mellon who have worked on the COVID Voice Detector project echo Raj’s optimism. According to Futurism:25
“‘I’ve seen a lot of competition for the cheapest, fastest diagnosis you can have,’ said Benjamin Striner, a Carnegie Mellon graduate student who worked on the project, in an interview with Futurism. ‘And there are some pretty good ones that are actually really cheap and pretty accurate, but nothing’s ever going to be as cheap and as easy as speaking into a phone.’
That’s a provocative claim in the face of the global coronavirus outbreak, and particularly the widespread shortages of testing kits. But Striner believes that the team’s algorithm, even though it’s still highly experimental, could be a valuable tool in tracking the spread of the virus, especially as the team continues to refine its accuracy by collecting more data.'”
Devices, apps or testing methods that can get the COVID-19 pandemic under control have positive potential, as most are eagerly waiting for the world to get back to the way it was before social distancing and the fear of COVID-19.