Giving the Humble Stethoscope an AI Upgrade Could Save Millions of Kids (ieee.org)

the_newsbeagle writes: The stethoscope is a ubiquitous medical tool that has barely changed since it was invented in the early 1800s. But now a team of engineers, doctors, and public health researchers have come together to reinvent the tool using adaptive acoustics and AI. Their motivation is this statistic: Every year, nearly 1 million kids die of pneumonia around the world, with most deaths in sub-Saharan Africa and South Asia. The death toll is highest among children under the age of 5. The researchers, from Johns Hopkins University, designed a smart stethoscope for use by unskilled workers in noisy medical clinics. It uses a dynamic audio filtering system to remove ambient noise and distracting body sounds while not interfering with the subtle sounds from the lungs. And it uses AI to analyze the cleaned-up signal and provide a diagnosis.

Comments

Definitely needed

[110010001000]

The main reason kids in impoverished nations are dying of pneumonia is the lack of electronic AI stethoscopes. Hopefully these guys have a startup and start shipping soon.

Re:Definitely needed

[Anonymous Coward]

I certainly hope they will 3D print them from privately mined asteroids.

Re:Definitely needed

[Rick Schumann]

You're being sarcastic, right? It probably takes, what, maybe half an hour on the outside to teach someone the basics of using a stethoscope, being a very simple device with no moving parts, needs no power, and is very durable, and most of all, very very cheap to produce? Not everything needs a gods-be-damned AI in it.

Re: Definitely needed

[Seewhatidonehere]

Its like the japanese self flushing singing toilet. Advancement at its brightest, usefullness at its best, been around for ten years and never sold much.

Re:Definitely needed

[turp182]

Regarding pricing, medically approved scopes start at over $100 (in the US). There are cheaper ones available.

Half an hour would be plenty of time for someone to learn how to hear a living person's heart beat. Verify life (scope is needed for that).

But what heart murmurs? I had one as a kid and I have a cat that has a different one (various levels and variants, for mamals).

What about identifying irregular heart activity?

To actually treat someone via stethoscope analysis requires a lot of time and education.

Re:Definitely needed

[oh_my_080980980]

Hey ass-hole:

"You wake up one morning to discover that your child is ill: His forehead feels hot to the touch, and his rapid breathing has a wheezing sound. You live in Malawi, where your health care options are few. When the local clinic opens, you wait for your turn with the solitary clinic worker. She’s not a doctor, but she’s been trained to identify and handle routine problems. "

"She puts on a stethoscope and presses its chest piece against your son’s front and back to carefully listen to his lungs. Through the windows, open in the heat of the day, come the sounds of people talking, the thrum of a generator, and the roar of a moped on the main road. The health worker strains to hear. " https://spectrum.ieee.org/biom...

This is not remotely the problem anywhere else. So in 3rd world countries were people who are not doctors treat people in noisy conditions, this would help.

Do you fucking understand now. Take your fucking head out of your ass.

Re:Definitely needed

[Rick Schumann]

You know what? WHATEVER. I think shit like this is just a solution in search of a problem, especially in poor countries that would be happy to get just plain old-fashioned stethoscopes and actual doctors. Everything has to get more complicated and more expensive because I'm sure the profit margin on a plain-old-stethoscope is small, so let's make some fancy thing that needs power and if it breaks you can't fix it so we can charge more and make more profit. Don't even pretend that isn't a major portion of the motivation behind things like this.

Re:Definitely needed

[Ol Olsoc]

The main reason kids in impoverished nations are dying of pneumonia is the lack of electronic AI stethoscopes. Hopefully these guys have a startup and start shipping soon.

That and those things that remove water vapor and generate enough water for the whole village. We need to send those over as well.

Re:Definitely needed

[Ol Olsoc]

You're being sarcastic, right? It probably takes, what, maybe half an hour on the outside to teach someone the basics of using a stethoscope, being a very simple device with no moving parts, needs no power, and is very durable, and most of all, very very cheap to produce? Not everything needs a gods-be-damned AI in it.

From what I can figure, this is just a stethoscope with a mic on thesensor side, and another mic on the other side to cancel out ambient noise. Okay.

Re:Definitely needed

[Ol Olsoc]

You know what? WHATEVER. I think shit like this is just a solution in search of a problem, especially in poor countries that would be happy to get just plain old-fashioned stethoscopes and actual doctors. Everything has to get more complicated and more expensive because I'm sure the profit margin on a plain-old-stethoscope is small, so let's make some fancy thing that needs power and if it breaks you can't fix it so we can charge more and make more profit. Don't even pretend that isn't a major portion of the motivation behind things like this.

Sounds to me like a noise cancelling mic and amp. Technology that's been around for years. Even cynical old me sees that as a good thing.

Even jaded old me couldn't resist saving children. I think it's kinda a genetic predisposition.

Practical old me tells us we better figure out how to feed all of the children we save though.

Re:Why would it need to change?

[PPH]

untouched stethoscope on display on his desk

Of course the one he actually uses he keeps stored in the refrigerator.

Hubble Stethoscope?

[Lionfire]

This article sounded a lot more interesting when I misread the title.

Re:Hubble Stethoscope?

[WankerWeasel]

I too misread the title. Saw Hubble, thought "How would they point it at Earth?" and read on. Not til close to the end did I realize that's just the name.

Re:Hubble Stethoscope?

[h33t l4x0r]

I read "Humble Stethoscope" but I imagined it coming bundled with a bunch of Steam games.

Re:Why would it need to change?

[Anonymous Coward]

From what I've seen of doctors, they store it in their ass, next to their head.

Re: My AI's AI

[Seewhatidonehere]

You are a plague to this planet, wanting your fucking AI to rule over the world. However. 1. Cramming a microphone into a stethoscope to alleviate unwanted noise is far from AI. My earphones can do it. Of course only when they are powered by extra battery. 2. Creating 100 dollar price equipment to replace a 5 dollar one will not save more people. 3. To alleviate outside noise, just need a quiet room. 4. When we will build AI, the stethoscope will become immediately obsolete in comparison with the handheld star trek tricorder.

Re: My AI's AI

[Seewhatidonehere]

Oh and the million kids die becouse they dont get the necessary medication and clean water to swallow it with and solid food to eat so they would get better.

Re:Why would it need to change?

[rtb61]

If you are sending a child to an unskilled worker in a noisy medical clinic, chance that a better far more expensive stethoscope will fix the problem, well, fucking less than zero. See problem right fucking there, 'UNSKILLED WORKER', a bloody witch doctor would be better than sending them to an unskilled worker, who can diagnose one thing and one thing only with a better stethoscope, any other condition and you are fucked.

Cool stuff

[JustAnotherOldGuy]

This is the stuff AI (or what passes for AI) can help solve- all sorts of fiddly problems that can benefit from the introduction of a "smart tool". I'm all for smarter gadgets and diagnostics that can help give regular/untrained people the ability to deal with various problems.

I mean, HELLO, this is what computers are meant to do- to help us do things we couldn't otherwise do.

Sure, maybe the wizards at the Mayo Clinic won't use it, but they aren't the target audience. I can see where this could be useful in all sorts of circumstances. On the battlefield, for one, but also in places where people trained to decipher the sounds heard through a stethoscope are far and few between.

It's like the super-simple AEDs (Automated External Defibrillator) that you see in offices and stores- they're simple enough that almost anyone can use one to restart a heart. My office has one and looks pretty straightforward to use.

Gadgets like a smarter stethoscope could help save some lives, and that's a good thing.

Re: Cool stuff

[Seewhatidonehere]

Some. Yet the title clearly claims millions.

Re:Cool stuff

[pz]

The summary takes a swipe at mature technology (the stethoscope) just because it is mature. Mature does not mean bad, nor ineffective. On the contrary, mature means "we have figured out all the ways to make this better." Stethoscopes are awesome instruments: they take no power, last indefinitely, and can be used to diagnose a wide range of diseases in skilled hands.

Now the last part is the important one: in skilled hands. The proposed instrument is for unskilled hands to help diagnose disease where doctors are scarce. It will need batteries. It will need protection from theft. It will cost a bundle. It may have annual licensing fees. It will surely need an internet connection. Now, is that really something that a sub-Saharan nation will be able to afford in its remote clinics?

Or would that same money be better put into training its workers as health care technicians specifically skilled to diagnose pneumonia through traditional stethoscopes? No licensing. Stethoscopes are dirt cheap. Once learned and continually exercised, the training doesn't go away, doesn't stop working, doesn't need batteries. Some of those trained workers would be inspired to become physicians. Those new physicians would help ameliorate the local shortage. Long term, that appears to be the far better choice.

If I were in charge of a remote village and given the option of (a) getting a shiny new device with all of the strings mentioned above, or (b) having the brightest one of my residents trained on hearing the difference between healthy and diseased lungs, you can bet which option I'd take.

Cool stuff, as the parent poster states, yes. Appropriate use of technology when a simpler, better solution exists? Perhaps not.

Here's where this kind of technology is actually useful: in comparatively rich places (not sub-Saharan Africa) where there's an under-served population at the physician level, but there are still skilled health care workers. Like Western inner city clinics staffed by physician assistants. PAs can use these tools to augment their existing skills to help triage patients more efficiently and accurately, sending those who are in need of more advanced services to regional hospitals.

But even then, such clinics are cost-sensitive, so pricing becomes an issue, and any company manufacturing a product that can affect the outcome of human health needs to carry substantial liability insurance. Those costs are reflected in the purchase price of such devices. I'm having a hard time thinking automated diagnostic machines are going to be inexpensive.

If the proposal were to, instead, create a small, ultra-fast antibody-based device that would take the results of a throat swab or cough to look for evidence of standard bacterial or viral lung pathogens, now you're talking. That kind of solution has the potential for being inexpensive and globally applicable.

Re:Cool stuff

[swillden]

It will need batteries.

Sure. Batteries are cheap.

It will cost a bundle.

There's no reason to expect this to be the case. It's a device specifically targeted at very poor regions of the world and there's nothing about it that requires expensive hardware. It requires a lot less hardware than is in the typical low-end smartphone that sells in India for $40.

It may have annual licensing fees.

How much do you want to bet me that it doesn't?

It will surely need an internet connection.

There is absolutely no reason for it to need an Internet connection.

Or would that same money be better put into training its workers as health care technicians specifically skilled to diagnose pneumonia through traditional stethoscopes? No licensing. Stethoscopes are dirt cheap.

But training is very expensive, and you're talking about very complex training.

Re:Cool stuff

[yarbo]

AEDs don't restart hearts. A flatline rhythm, AKA asystole, is not a shockable rhythm. The AED will figure out if it's a shockable rhythm though, but it's important I ruin all medical shows and movies.

AEDs are great if people know how to use them and know what else to do (check pulse, delegate someone to call 911), run chest compressions while someone else attaches pads, etc...

You should take a class though, because you don't want to have to figure everything out for the first time during an emergency. You get hands on experience with a trainer and mannequins and learn to handle and recognize some other emergency situations.

Re: Cool stuff

[JustAnotherOldGuy]

Some. Yet the title clearly claims millions.

What's your point?

Out of billions, "some" could easily be "millions".

Re:Cool stuff

[JustAnotherOldGuy]

I find your arguments less than compelling.

Is it a perfect solution? No.

Is it likely to be a useful, cost-effective way to help the target audience? Yes, I think so.

No one is saying that this is the end-all be-all solution to solve this problem, but it can help and I think it's obvious that it has a lot of potential. Just because it's not perfect or doesn't meet your ideal criteria isn't enough of a reason to not use it.

I remember when people said the same about cellphones- "They're too expensive, not useful enough, you'd have to construct cell towers all over the entire country, they need batteries, there are already payphones everywhere..." etc etc etc.

And yet most of us carry one every day and never think twice about it.

Re:Cool stuff

[JustAnotherOldGuy]

AEDs don't restart hearts.

You're correct, and that was my mistake. I should have said that it can help normalize an irregular heartbeat (or whatever term is accurate).

As for training on them, I agree 100%. An emergency situation is not the best time to learn anything.

I opened ours up at lunch one day and went through the manual- it looks very straightforward to use. I think most people could get the gist of it in a 15 or 20 minute session.

It's a perfect example of a somewhat-smart gadget that can help save lives, even in relatively untrained hands. And that's why I think a "smart stethoscope" is a great idea too. There's no real downside to it that I can think of.

WhoopeeShit

[AndyKron]

And it will only cost ten times what a normal one would AND the yearly software license will be so affordable!

Re:WhoopeeShit

[AmiMoJo]

If it saves you needing to pay a doctor for their time then it's probably going to be worth it.

Charities will buy them to give to staff.

Re: WhoopeeShit

[Errol backfiring]

On the contrary, we love them so much that we rather spend money on a doctor and an affordable stethoscope than on a stethoscope alone.

Re: WhoopeeShit

[jrumney]

Have you priced stethoscopes recently?

Re:My AI's AI

[Rick Schumann]

Yo, dawg..

Re: My AI's AI

[Seewhatidonehere]

Wake the fuck up kid and smell the coffee. Your hand is hanging into your pissbowl. Build a starship into that fucking stethoscope it still wont save millions of kids without alleviating the reasons causing them pneumonia in the first place.

Humble upgrade to humble stethoscope

[codeButcher]

All those years sine the 1800s and they haven't been able to upgrade it with a simple heating pad? I've had some stethoscopes on my chest and back that could give me pneumonia all on their own.

Let's play ....

[houghi]

... bullshit bingo.

It has "AI" and "for the children" in the subject, so we are starting strong today.

Depends on your pathology

[DrYak]

I've even had one arrogant turd of a doctor have a highly polished untouched stethoscope on display on his desk... unused.

You know a stethoscope serves usually to listen, e.g. to your lungs or your heart.

If you're constantly going to the doctor to pester him about this weird skin rash that you are regularily getting on your penis, the stethoscope will be of no use.

(And about the polishing : we are supposed to rub it with disinfection before and after each single use. Of course, it's going to look pristine and polished.
Or would you prefer if we used it to help you exchange every possible virus and bacteria among all patients coming to the practice ?!)

Re: Depends on your pathology

[Anonymous Coward]

Talking about penises and then polishing has me.... confused.

Not to rain on your parade

[nospam007]

This would replace the doctor, not the stethoscope.
He's the weak link.

How to fix medical care

[AHuxley]

Send your doctors to medical school on merit.
Graduate the best and only the best.
Ensure anyone who wants to work in your nation as a medical doctor is a qualified professional.
Everyone sits the same exams and has to pass the same exams to be approved.
Stethoscope skills are part of the needed years of approved study.
After years of hard work and study a doctor enters the profession.
Review the work done by doctors and ensure peer review is done.

The reason why that education is important is that not every condition is going to be the AI expected "pneumonia".
Skill is needed to detect many other conditions and what the best treatment is.

Re:How to fix medical care

[drinkypoo]

You have it completely backwards. The stethoscope won't have a bad day, it doesn't have ear wax, and it doesn't have to go to school. And the data will be produced through analysis, and it will have heard more conditions than any doctor. Doctors expect pneumonia. Computers don't. They process signals and match patterns.

In the future, doctors may be rare, and involved in only the very strangest and most complex cases, while nurses with advanced diagnostic equipment handle the routine stuff. And the computers will learn from the doctors, and health care will improve as a result.

As for the beginning of your comment, there is a world-wide shortage of doctors at the moment. I don't know how it works in other countries, but in this one the AMA has made it difficult to become one in a lot of irrelevant ways, which is to say they don't improve overall quality. Washing someone out because they don't perform well in an ER environment when they might be a perfectly good practitioner in other contexts, for example. Not every doc needs to work in the ER. We cannot survive your plan.

Re: How to fix medical care

[kenh]

We need to start accepting the good enough. Better to be seen by an OK doctor right away rather than be seen by the best doctor in 6 months

Do you know what they call the student at the bottom of the class in medical school? Doctor.

I am not aware of this being an actual alternative for anyone, most people have access to exactly one doctor, the one that is close to their house and accepts their insurance...

Re:How to fix medical care

[AHuxley]

Re "they don't perform well in an ER environment"
Why should any nation have to accept doctors who cant learn and who cant study?
Why could not see a very different set of people everyday in the ER and put their education to some use?
Thats the way that sorts out the professionals. Peer review and constant education and learning.
The good quality doctors can do all that. People who cant need to look for another profession.
A nurses with advanced diagnostic equipment is not a doctor.
That is a way to sort people entering a hospital and to get some basic information ready for the medical experts.

Re: My AI's AI

[drinkypoo]

"You are a plague to this planet, wanting your fucking AI to rule over the world. "

Humans are doing a shit job.

"Cramming a microphone into a stethoscope to alleviate unwanted noise is far from AI. My earphones can do it. Of course only when they are powered by extra battery."

That's not the clever part.

"Creating 100 dollar price equipment to replace a 5 dollar one will not save more people."

It also replaces the physician.

Granted, it only replaces them for the one task, but there are other technologies which replace them for other tasks. And as it turns out, a doctor with advanced aids to diagnosis is better than either a computer or a doctor alone.

" To alleviate outside noise, just need a quiet room."

There's no such thing in a busy disaster shelter, or on the side of the road, or in a war zone.

"When we will build AI, the stethoscope will become immediately obsolete in comparison with the handheld star trek tricorder."

Tricorders require additional sensor technology, not just signal processing. Also, we may never build real AI, or it may make US obsolete.

ok....

[argStyopa]

....it's not a bad idea to upgrade the concept of stethoscopes, but I think there's still a value in the basic tool that is (essentially) impervious to damage, climate, immersion, AND DOESN'T NEED A BATTERY. *Particularly* in that undeveloped remote-care situation they envisage in the OP.

AI needed to combat a noisy clinic?

[kenh]

with most deaths in sub-Saharan Africa and South Asia. The death toll is highest among children under the age of 5. The researchers, from Johns Hopkins University, designed a smart stethoscope for use by unskilled workers in noisy medical clinics.

Did anyone consider the possibility of simply examining children in quiet rooms before they decided to throw a few million dollars in research funds for a ten thousand dollar "pneumonia detector" to be deployed in some of the most impoverished locations in the world? I bet the local practitioners could think of better uses for thousands of dollars than a battery-operated tool to replace a $5 stethoscope... like vaccines, a refrigerator, etc...

Re:AI needed to combat a noisy clinic?

[oh_my_080980980]

Come on that's way to obvious of a solution. Don't you know we're suppose to throw tech at a problem with a simple solution.

Re:What?

[kenh]

I'm sure there are more issues at play for saving lives in sub-Saharan Africa than a lack of AI stethoscopes.

Well, we know they also need One Laptop Per Child and balloon-based Internet...

Re:Hi Everybody! I'm Doctor Nick!

[kenh]

"Unskilled worker" in this context means anyone that hasn't gone to medical school.

As for the Democrat plan in the US to "outlaw" private healthcare so everyone has equal access to substandard care is a non-starter. Supporters argue that they have finally figured out a way to provide quality healthcare for all at an affordable price in a timely manner, but they can't explain why no other socialized medical care plan on the planet has achieved that goal. Every time you point to a fundamental flaw they claim "well, we will make sure that doesn't happen, but when you ask how, all you get is crickets...

It's a grand idea, but medical care is expensive in terms of manpower and resources, and expanding access does not reduce the need for either.

AOC actually claimed that the US would save money on funerals with improved access to healthcare... Apparently, in her mind people with universal healthcare coverage are immortal, since they won't have a funeral. (Delaying a funeral isn't a savings, it's a deferral - nothing more.)

Re:AI Doesn't Exist

[kenh]

My HAL9000 would disagree with you!

Idiots of the world unite...

[oh_my_080980980]

Really? Yeah because an AI stethoscope won't be expensive and will be easy to service in sub-Saharan Africa. Did you read the bit about "developing strategies for countries with limited resources" or was your head up your ass because you're a fucking engineer who's never been to sub-Saharan Africa. FYI given the many illnesses that plague people living in sub-Saharan Africa broad spectrum antibiotics would be the right treatment. But hey what do doctors know compared to IEEE engineers who have never treated anyone in sub-Saharan Africa. Fucking morons.

"Given the global burden of these respiratory ailments, the World Health Organization has developed strategies for countries with limited resources. The guidelines for pneumonia [PDF] diagnosis minimize dependence on any technological tools, and instead rely solely on observed symptoms of shortness of breath, cough, and rapid breathing. In hopes of saving lives, the WHO recommends antibiotic treatment for all children with these symptoms, with the result that half the children who get treated for pneumonia don’t really have it. This approach puts unnecessary costs on communities, and the unnecessary medication contributes to the growing problem of antibiotic-resistant bacteria.

We propose a technological solution that builds on the stethoscope, which has hardly changed since its invention in the early 1800s."

"Today’s electronic stethoscopes typically cost around US $500, making them far too expensive for many health workers in the developing world. The Johns Hopkins smart scope is designed to be significantly cheaper, with affordable electronic components and low-cost power and computing options, in hopes that it can be useful for low-resource communities."

Dependency Inversion

[holophrastic]

No food, no water, no medicine, no doctors, no "quiet please, I'm diagnosing your dying child". So the solution is to replace the most reliable medical tool in the entire industry, a solid device that can be thrown around and has zero dependencies other than the patient, and someone to wield it, and we're going to replace it with a computer.

May I remind you:

No food, no water, no medicine, no doctors, no quiet, no electricity, no tech support.

Re: My AI's AI

[Ol Olsoc]

The difference is that we're at least on the same planet with Milla Jovovich.

Not knowing whothe good Ms. Jovovich is I googled her name. Well now! It's a good thing I'm not hooked to the Brainwaves to speech machine in the last article, or my wife would be wondering about the dialog coming from my office.