Chris Mitchell

---

The apps aim to analyze The Sound of coughing and other noises, in order to determine whether a user is likely to get infected

Two leading universities are trying to the development of apps, the lists of users to cough and votes, the predictions, whether they're infected with the coronavirus.

But the two projects take different approaches to privacy.

The Cambridge University effort trying to keep it voluntary, anonymous, but says that this work is currently limited.

Meanwhile, a team at Carnegie Mellon University says it is important that users register themselves, but it had to temporarily go offline.

The two initiatives are independent of each other.

Both of them are based on Machine Learning , Artificial Intelligence of computers to analyze of large amounts of data, patterns that can be used to solve problems.

In this case, The Goal is to distinguish in the situation, the Covid-19 of other diseases, including the flu.

Both teams to confirm that the resulting software will replace The Need for further medical tests.

data quality

University of Cambridge.

members of The Public are invited to breathe and to be tested for cough in a computer microphone, as well as information on age, gender, approximate location, and whether you have recently positive for the coronavirus.

you will then be asked to read The Following sentence Three Times : "I hope my data can help to manage The Virus pandemic. "

"The Goal is to collect enough data to check whether these sounds we are able to diagnose people, the Covid-19 and perhaps even The Stage of the disease," explains Prof Cecilia Mascolo .

"If this works, we could perhaps services such as the British NHS 111 helpline are available to help. "

In his first day, about 1,200 people recordings made, 22 of which said they had recently tested positive.

The Team hopes to have a product in less than two months time.

"The analysis will not take too long, but it all depends on the quality of the data we collect," Prof Mascolo.

At present, The Project is limited to the collection of samples on a website, but via a smartphone app.

This is in part because Apple and Google limit that can be published, coronavirus-like applications in their stores, and this effort has yet to Qualify to.

The Cambridge team hopes that its app will be released for download in The Coming days

"Which app is better, because it can go Back To The Volunteers in The Following days, and ask them to make recordings," says Professor Pietro Cicuta, another member of The Team .

This is not possible, on the website, he adds, without compromising the users anonymity.

Unintended consequences

This is built on the events in the Pittsburgh-based University.

It briefly went live on 30. March. The users were asked to cough, the data set vowel sounds, and recite The Alphabet , as well as details about yourself.

At The End of The Process , the tool, an indication of how likely you are to Covid-19.

But the researchers began a process of rethinking was required.

"It doesn't matter how many Disclaimer you put it - how much can you not tell the people that this validated medical - Some People take, The Machine as The Word of God," explains Dr. Rita Singh.

"If a system says a person is, the Covid-19 is not mandated that you have it, it can kill the person.

"And if he says a healthy person, you have, and you are going to be tested, they can take up valuable resources that are limited.

"So, we have very little room for error Either Way , and advise on the nature of the presentation of the results, so that these risks disappear. "

A note on the Covid Voice detector ' s Website says it is currently offline

she still hopes the data-collection aspect of The Service back online before The End of This Week .

The Plan is to register users, without their names. But in contrast to Cambridge, the effort, The Volunteers required to create an account with Links to your E-Mail address.

Prof Singh says, this is necessary to provide the users with revised feedback tool is more accurate - for example, if someone moves in a high-risk group.

"The Other thing is that we add The Right to be forgotten in all seriousness," you.

"So you should have the opportunity to come Back To us with The Line , years down, press the button and say I want to delete every single sample of My Voice . "

equipment restrictions

While both projects are optimistic about their prospects, another expert in the KI-based detection concerns.

"With The Midlands and London suffer from the worst Covid-19-outbreaks in the UK, the regional differences in The Way people mean The Sound , Some areas of excessive influence on the KI-model, unless carefully controlled in the data could have," said Chris Mitchell , chief executive of Audio analysis.

"is The different challenge of a purely technical nature.

"Picking up detailed respiratory sounds for the expert analysis is complicated, without using specialist microphones, and both tests require the patient to record yourself with a smartphone [or PC]. "

privacy, coronavirus pandemic, artificial intelligence, cambridge, apps, university of cambridge

Source of news: bbc.com