Cough analysis app could detect COVID-19 by sound alone

Posted:
in iOS edited October 2020
Researchers at MIT have used machine learning to create software capable of detecting whether a person has caught COVID-19 by analyzing their cough, a development that could eventually result in an iPhone app for daily checks.

The UK's NHS COVID-19 app for iPhone uses the Apple-Google exposure notification API to help combat the coronavirus.
The UK's NHS COVID-19 app for iPhone uses the Apple-Google exposure notification API to help combat the coronavirus.


So far, the iPhone has lent itself to helping users determine if they are at risk of the coronavirus by coming into close proximity to someone carrying the virus. If a new discovery by MIT researchers is developed further, the iPhone may end up being able to do more to dampen the spread of the virus further.

A paper from the team published in the IEEE Journal of Engineering in Medicine and Biology claims an AI model was created that could tell the difference between asymptomatic people and those who are healthy, via analysis of recordings of forced coughs. The model is claimed to be accurate 98.5% of the time when listening to recordings of people confirmed to have had COVID-19, as well as 100 percent of asymptomatic cough recordings.

The team collected more than 70,000 recordings via a website where the public could record a series of coughs via their smartphone or other devices, at the same time as filling out a survey about their symptoms, if they were confirmed to have the virus, and other details. The recordings resulted in around 200,000 forced-cough samples, including 2,500 of those confirmed to have COVID-19 or were asymptomatic.

Combining the 2,500 confirmed samples with another 2,500 randomly selected from the data set, the AI model was trained, then tested. The researchers claim the results revealed "a striking similarity between Alzheimer's and COVID discrimination."

The AI framework was based on one that existed for Alzheimer's research, and determined it could pick up four biomarkers relating to vocal cord strength, sentiment, lung and respiratory response, and muscular degradation specific to COVID-19.

The team are now working to create a pre-screening app it intends to distribute for free, based on the AI model, as well as working with a number of hospitals to enlarge the cough recording pool, for further training.

It is suggested by the team that such cough analysis could be implemented into smart speakers and digital assistants to perform daily assessments. This naturally would depend on the devices involved having sufficiently good quality microphones as well as handling the necessary privacy issues, not to mention the assistance of companies like Apple and Amazon to implement, making it unlikely albeit altruistic.
«13

Comments

  • Reply 1 of 46
    mac_dogmac_dog Posts: 1,066member
    I wonder if it can distinguish between a flu cough and a COVID cough. Hmmm…
    steve_jobsaderutterforgot usernamejony0
  • Reply 2 of 46
    jas99jas99 Posts: 147member
    So this would allow on demand, free, instantaneous testing for anyone with the right phone or microphone. This is actually pretty big news, especially if an app could be developed quickly. 
    jony0
  • Reply 3 of 46
    22july201322july2013 Posts: 3,564member
    Perhaps MIT could write an app, but it seems more appropriate for MIT to give the cough-detection code to Apple and Google (who could pay MIT) to enhance their COVID API, so that the governments deploying their COVID apps can access that feature, so that users don't need to install multiple apps. After all, this is a worldwide problem and I don't think MIT wants to get involved in deploying apps for 200 different countries.
    PetrolDavegregoriusmmac_dogforgot usernamejony0
  • Reply 4 of 46
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    SpamSandwichcat52NotoriousDEV
  • Reply 5 of 46
    Please put all efforts into simple, preventive universal 1.5m detection.
    Use BT, UW, LiDAR, NFC, TOF, AirDrop proximity detection and all other high-profile technologies to their extreme.
    How hard can it be now that the Covid API already uses a form of proximity-detection and the Measurement app has sub-millimeter precision ?
    edited November 2020
  • Reply 6 of 46
    fred1fred1 Posts: 1,108member
    It would be great if really worked, but allow me to doubt. So it was the cough that determined that people have COVID and not the list of symptoms? Hmmm. 
    cat52
  • Reply 7 of 46
    GeorgeBMacGeorgeBMac Posts: 11,421member
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2....

    Actually, no, they didn't say that.   It's just more propaganda and disinformation.
    h4y3sforgot usernameStrangeDaysjony0
  • Reply 8 of 46
    GeorgeBMacGeorgeBMac Posts: 11,421member
    Please put all efforts into simple, preventive universal 1.5m detection.
    Use BT, UW, LiDAR, NFC, TOF, AirDrop proximity detection and all other high-profile technologies to their extreme.
    How hard can it be now that the Covid API already uses a form of proximity-detection and the Measurement app has sub-millimeter precision ?

    Proximity (and duration) tracking has never been a technical roadblock.
    The roadblocks are mostly political.   Some fear others knowing where they've been and who they've been with more than they do dying themselves or killing others.
    forgot usernamejony0
  • Reply 9 of 46
    GeorgeBMacGeorgeBMac Posts: 11,421member
    fred1 said:
    It would be great if really worked, but allow me to doubt. So it was the cough that determined that people have COVID and not the list of symptoms? Hmmm. 

    I can see that an app could determine that a cough was NOT due to Covid.   But to determine that it IS due to Covid when a dozen other things could be causing it is not reasonable.
    aderutter
  • Reply 10 of 46
    But still declared covid by hospital so they get paid.....what a bunch of crap reported 
    cat52NotoriousDEV
  • Reply 11 of 46
    fred1fred1 Posts: 1,108member
    Kuyangkoh said:
    But still declared covid by hospital so they get paid.....what a bunch of crap reported 
    Please tell me you don’t still believe that all hospitals get paid more by the government if they report a case as being COVID. This was debunked in April:
    https://www.snopes.com/fact-check/medicare-hospitals-covid-patients/


    Dogpersonaderutterforgot username
  • Reply 12 of 46
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2....

    Actually, no, they didn't say that.   It's just more propaganda and disinformation.
    WRONG.  Stop spreading fake news.  “Comorbidities Table 3 shows the types of health conditions and contributing causes mentioned in conjunction with deaths involving coronavirus disease 2019 (COVID-19). For 6% of the deaths COVID-19 was the only cause mentioned. For deaths with conditions or causes in addition to COVID-19 on average there were 2.6 additional conditions or causes per death.”  https://www.cdc.gov/nchs/nvss/vsrr/covid_weekly/index.htm?fbclid=IwAR1qC9bXRTmggpniTBAvlzvAZ-wpZdAsCoFlRwiujGQy01DJlvTM6aOr4HQ  Yes this means 96% or more people who died with COVID-19 WERE ALREADY SICK WITH AN AVERAGE OF 2.6 OTHER ILLNESSES!  

    Not only that, but note the big drop in influenza cases, because hospitals have a financial incentive to pretend it is actually COVID-19.  Further, most heart conditions are being classified as COVID-19 deaths, this is the problem when government gets involved in healthcare!  
    edited November 2020 cat52SpamSandwich
  • Reply 13 of 46

    fred1 said:
    Kuyangkoh said:
    But still declared covid by hospital so they get paid.....what a bunch of crap reported 
    Please tell me you don’t still believe that all hospitals get paid more by the government if they report a case as being COVID. This was debunked in April:
    https://www.snopes.com/fact-check/medicare-hospitals-covid-patients/


    Snopes is HARDLY a reliable site for facts.  Did you read about its founders?  *rolls eyes*
    cat52SpamSandwich
  • Reply 14 of 46
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    Yes, this sounds like unsupportable nonsense.
    cat52
  • Reply 15 of 46
    GeorgeBMacGeorgeBMac Posts: 11,421member
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2....

    Actually, no, they didn't say that.   It's just more propaganda and disinformation.
    WRONG.  Stop spreading fake news.  “Comorbidities Table 3 shows the types of health conditions and contributing causes mentioned in conjunction with deaths involving coronavirus disease 2019 (COVID-19). For 6% of the deaths COVID-19 was the only cause mentioned. For deaths with conditions or causes in addition to COVID-19 on average there were 2.6 additional conditions or causes per death.”  https://www.cdc.gov/nchs/nvss/vsrr/covid_weekly/index.htm?fbclid=IwAR1qC9bXRTmggpniTBAvlzvAZ-wpZdAsCoFlRwiujGQy01DJlvTM6aOr4HQ  Yes this means 96% or more people who died with COVID-19 WERE ALREADY SICK WITH AN AVERAGE OF 2.6 OTHER ILLNESSES!  

    Not only that, but note the big drop in influenza cases, because hospitals have a financial incentive to pretend it is actually COVID-19.  Further, most heart conditions are being classified as COVID-19 deaths, this is the problem when government gets involved in healthcare!  

    LOL.... Nice spin!
    Did you get that nonsense from FauxNews?

    By the way, did they tell you that even NON-Covid deaths have increased by 100,000 this year?   They are taking that to mean that Covid has killed even MORE than the nearly 1/4 millions Americans reported.
    forgot usernameStrangeDaysjony0
  • Reply 16 of 46
    GeorgeBMacGeorgeBMac Posts: 11,421member
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    Yes, this sounds like unsupportable nonsense.

    Like most propaganda, it has a grain of truth to it.   In this case, yes, those who get seriously ill or die from the virus tend to have other conditions as well.   So they spin that to try and convince the fools that the virus really is a Democratic hoax.
    forgot usernameStrangeDaysjony0
  • Reply 17 of 46
    mac_dogmac_dog Posts: 1,066member
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    So what? We should simple disregard it? More like “Trollingforall.”
    GeorgeBMacmacplusplusDogpersonStrangeDaysjony0
  • Reply 18 of 46
    Even if I were to charitably believe this claim (I have asthma and currently bronchitis from the wildfires in September) by the time something like this might potentially be FDA-approved, the pandemic will most likely be over.

    This app (if it ships as one) could result in a lot of needless testing, or false negatives, hard to say either way with accuracy.
    SpamSandwich
  • Reply 19 of 46
    Mike WuertheleMike Wuerthele Posts: 6,837administrator
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    Yes, this sounds like unsupportable nonsense.
    If I had died from my COVID battle earlier this year, my "pre-existing conditions" on my death certificate would have been "ionizing radiation exposure" and "Potential asbestos exposure," both from more than 20 years ago. Another pair of pre-existing conditions that would have been listed on my death certificate would have been "military vaccinations" and "steel plate in the left leg" from 1996. both unrelated. A fifth, depending on ICD-9 or ICD-10 is "ex-Submariner" which ended in 1999.

    So, four or five "pre-existing conditions," zero of which would have had to do with COVID.

    Funnily, if I got hit by a car tomorrow, four or five would be on my death certificate as pre-existing conditions, despite the car being what did me in. No matter what finally kills me, be it being decapitated in a sword battle in a warehouse with lightning all around, or just keeling over from one too many Old Bay potato chips, all these things will be on my death certificate.

    Other common pre-existing conditions include such winners as "ever pregnant," "ever had an ear infection," "ever had a broken bone," "ex-smoker" and so many more. Ever had your tonsils out? That counts. So does an appendix removal, a vasectomy, a tubal ligation, and so, so much more. All counted by the CDC.

    Both of you are misinterpreting what the CDC said by accident, not having read the original materials and going by what you've heard, or by believing what a venue has told you because I'd rather not consider the ramifications of either of you willfully misinterpreting what the study says. 

    Dial it back a bit. There's a lot in this thread that is over the commenting guidelines, including the use of the term "fake news" or a derivative.

    AI will not be your political battleground.
    edited November 2020 mattinozh4y3sAngmohDogpersoniHyGeorgeBMacforgot usernameStrangeDaysdewmejony0
  • Reply 20 of 46
    This is an utter joke.  CDC says most people have 2.6 preexisting conditions who die, they don’t die from SARS-Cov2.  The cough could be from many other things.  
    Yes, this sounds like unsupportable nonsense.
    If I had died from my COVID battle earlier this year, my "pre-existing conditions" on my death certificate would have been "ionizing radiation exposure" and "Potential asbestos exposure," both from more than 20 years ago. Another pair of pre-existing conditions that would have been listed on my death certificate would have been "military vaccinations" and "steel plate in the left leg" from 1996. both unrelated. A fifth, depending on ICD-9 or ICD-10 is "ex-Submariner" which ended in 1999.

    So, four or five "pre-existing conditions," zero of which would have had to do with COVID.

    Funnily, if I got hit by a car tomorrow, four or five would be on my death certificate as pre-existing conditions, despite the car being what did me in. No matter what finally kills me, be it being decapitated in a sword battle in a warehouse with lightning all around, or just keeling over from one too many Old Bay potato chips, all these things will be on my death certificate.

    Other common pre-existing conditions include such winners as "ever pregnant," "ever had an ear infection," "ever had a broken bone," "ex-smoker" and so many more. Ever had your tonsils out? That counts. So does an appendix removal, a vasectomy, a tubal ligation, and so, so much more. All counted by the CDC.

    Both of you are misinterpreting what the CDC said by accident, not having read the original materials and going by what you've heard, or by believing what a venue has told you because I'd rather not consider the ramifications of either of you willfully misinterpreting what the study says. 

    Dial it back a bit. There's a lot in this thread that is over the commenting guidelines, including the use of the term "fake news" or a derivative.

    AI will not be your political battleground.
    I almost died from the regular flu last year and I personally know multiple people who’ve died from COVID. The fact is, people with severe underlying conditions are the ones most at risk. 99+% people suffer few or mild effects and fully recover. COVID is not the “end of the world” virus it has been portrayed as in the media and the subject itself is political, no matter how much that might be disputed.
    cat52
Sign In or Register to comment.