Cass

To add to this, looking at figure 4, why is their "average" heartbeat so messed up? That's not what a normal average heartbeat looks like. P is too flat, Q is too big, R is blunted, and there's an extra wave between S and T that's not supposed to be there at all. If their "healthy patient" ECGs were bad enough to produce this mess on average, it's no surprise their AI had no trouble telling the data sets apart.

(For comparison, the "CHF beat" looks a lot more like a healthy heartbeat.)

As a doctor as opposed to an AI researcher, so many of the choices this study makes are baffling to me.

First of all, why just one heartbeat? You never capture just one heartbeat on an ECG anyway, and "Is the next heartbeat identical to the first one?" is such an important source of information, it seems completely irrational to exclude it. At least pick TWO heartbeats. If you're gonna pick one random heartbeat, how do you know you didn't pick an extra systole on accident? (Extra systoles look different, and often less healthy, than "normal" heart beats, as they originate from different regions of the heart.)

Secondly, why heart failure and not a heart attack? One definition of heart failure is "the heart is unable to pump sufficiently to maintain blood flow to meet the body's needs," which can be caused by all sorts of factors, many of them external to the actual function of the heart - do we even know for sure that there are ANY ECG changes definitely tied to heart failure? Why not instead try to detect heart attacks, which cause well-defined and well-researched known ECG changes?

(I realize AIs that claim to be able to detect heart attacks already exist. None of the ones I've personally worked with have ever been usable. The false positive rate is ridiculously high. I suppose maybe some research hospital somewhere has a working one?)

Cooking for cats is actually quite hard, and very easy to screw up. The kinds of meat humans usually eat don't contain everything your cat needs. Unless what you're cooking for them is actual mice, be very sure you know what you're doing. You have to add taurin and calcium, among others.

On the web, there's usually a ratio of readers/lurkers to commenters of somewhere between 100 to 1 and 1000 to 1. So if 30 people comment on this article, that means 300 to 3000 are reading it. That means your second degree connections need to be two or three orders of magnitude higher.

If you want a fully qualified GP-track resident, there's probably no way to compress the learning process much further down than the four years of American med school. German med school takes six and a half years to teach the same amount of knowledge, and that's already a brutally challenging course of study. (At least it was to me, a reasonable driven straight-A student.)

Looking at the American med students I know, getting through med school in four years seems to require giving up all semblance of a private life and driving yourself to the edge of exhaustion/over the edge into a mental break-down. I don't think there's a way to do it in three years or even less.

That said, there might be a market for specialists with less training than that. There's no real reason to make e.g. an aspiring orthopedic surgeon sit through the entirety of med school, including all the dermatology/pediatrics/gynecology etc stuff.

But as another commenter mentioned, you'll have a hard time getting the regulatory bodies to accept your "eight months to orthopedic surgery" students as fully qualified doctors

I'm always baffled by the idea that we need to get people living in space for the human race to survive climate chance or other catastrophes. What do you think we could possibly do to the Earth that would make it less habitable than space? Even in the worst, worst case scenario, the Earth's gonna be left with some amount of oxygen-containing atmosphere, arable soil, gravity, and survivable if possibly uncomfortably hot or cold temperatures. Any station we could possibly build to withstand the temperatures and vacuum of space would be much more easily built on Earth.

It seems like this would be a trivial problem to solve. In the extremely unlikely case that some guest requests access to or deletion of their data in your physical guestbook, just ask for the approximate date of their stay, look through the book for the relevant page, (presumably entries are sorted by date), make a copy for them (with the data of anyone else that wrote on the page censored out), and stick the original in the shredder. Done.

Yeah, it's gonna take you ten, fifteen minutes, but how often is it going to happen, realistically?

I'd never realized this, even after a year of living as a pedestrian in the UK, but this finally answers my question of why the pedestrian lights take so freaking LONG to turn green here. It may be intended as a pedestrian-friendly measure, but it FEELS like an intense annoyance, and the natural result is that everyone just crosses on red.

The common cold is caused by a virus, not by bacteria. You can get the bacteria that cause Legionnaire's disease building up in the fluid in air conditioning units and then getting vaporized into the air, but I don't see how spraying more bacteria around is going to stop you from breathing in the ones the AC is blowing into your face.

I'd be very curious about which bacteria you have in your spray, and which bacteria they're supposed to be displacing.

Research/academia can certainly be kind of screwed up, but I doubt that any ethics board in a developed country is going to okay a study that involves tricking people into a spinal tap. And all the consent forms patients are asked to sign before agreeing to a procedure, the ones that tell you about the procedure and the risks in excruciating detail, are as much for the protection of the doctor as the patient. That way, if the patient ends up paralyzed, at least they can't claim you didn't warn them when the case goes to court.

So unless you were dealing with a single rogue Mad Scientist, I'd assume that was an error.

An EKG machine cannot tell you whether you're fine or not. Not all heart attacks show up on EKGs, which is why the hospital will do a blood test for Troponin, too. A home Trop test would do what you want, but it can take several hours for the test to turn positive, and the hours between the onset of pain and the blood test showing that you're having a heart attack should really be spent attached to a cardiac monitor and in the presence of trained medical personnel, i.e. in hospital.

As far as I'm aware, computers are better at diagnosing a very small and very specific subset of diagnoses. If I'm wrong about this and there's a computer out there who can diagnose patients with better or equal accuracy than the average physician, please link me to the study proving so immediately, because I desperately want this in my life, and so do ninety percent of my colleagues. (There's a stubborn subset who wouldn't trust a computer if Asclepius himself descended from the heaven to sing its praises. The rest of us have ten times more work than time and would gladly welcome the help.)

As it is, we can't even seem to find a program that can analyse an ECG better than an overtired and exhausted resident at 4 am, which is really quite a low bar to cross.

Glucose, insulin and potassium are already readily available on every ambulance, so all you'd need is a clinical trial proving that the mixture is effective and safe to use for heart attack patients. The costs of a clinical trial seem well within the reach of private philanthropy, especially since this isn't an experimental drug where you'd have to start from scratch--all three of those meds have been extensively studied both separately and in combination. (Glucose and insulin is a combination commonly used to treat high potassium levels, for example.)

If you're taking Warfarin, you have to get your INR checked every two to three weeks. It's really not practical to do that at three different labs every time. Also, this isn't some exotic edge case that only applies to a few people, but an extremely common medication. We're talking about huge numbers of patients here. If all their tests had to be triple-checked, the costs would be enormous.

Oh hell yes. The main one peaople ask me about right now is Marcumar/Warfarin, a drug many patients with atrial fibrillation have to take for the entire rest of their life, despite the fact that it has awful side effects and requires them to go to their doctor every two to four week to have a blood test done. Many patients are quite excited to hear that there's now new alternative drugs (NOAKs) that let you skip the biweekly blood draw. I get asked about Pradaxa, Eliquis and Xarelto so often, my standard Marcumar pitch now includes a whole paragraph about how those alternatives are just as bad. This is getting harder to sell as new studies are coming in and my arguments against the new drugs are steadily being reduced to "They're a lot more expensive." Patients really do NOT want to hear about how they have to get stuck with needles every two weeks so their doctor can save a few Euro a month in the prescription budget.

The whole NOAK advertising scheme seems like a huge win for the pharma companies to me. (Possibly also for the patients. Not so much for the affordability of our medical system.)

All these "back to the dark ages" worst case scenarios seem to assume that we'll wake up one day and realize that every single one of our antibiotics has stopped working on anything. That's not how it's going to happen. It only takes a few years for most strains of bacteria to lose their resistant traits, and we're going to have years of undeniable warning signs. (As opposed to the deniable warning signs we have now.)

What's actually going to happen: We're going to have an epidemic of some sort of drug-resistant bacteria in the Western world--maybe that incurable TB that's going around, maybe something else. A few thousand Westerners are going to die--people the Western world actually cares about, as opposed to them poor people in them third world countries that are always dropping like flies anyway.

There'll be a media hysteria similar to the Ebola crisis. People will suddenly realize that antibiotic resistance isn't just a problem affecting poor people in third world countries, but that it is now a problem threatening them and their children.

The resulting media blitz will educate even the average idiot about the importance of taking the full course of antibiotics even if you feel much better already. Antibiotic use on lifestock will be banned. Western governments will pressure China and India into stopping the unrestricted sale of antibiotic. Western doctors are finally going to realize that prescribing antibiotics like candy just to shut a patient up is a mistake even if the patient is yelling really loudly. Developing a new antibiotic is sudenly going to be a lucrative or at least PR-relevant strategy for Big Pharma, instead of a likely fincancial loss. Hospitals are going to be forced to actually deal with the problem of MRSA. (Right now, many patients that should be tested and isolated aren't tested because there's no room to isolate them in case of a positive result.)

Within a few years, drug resistance will taper off sharply, without antibiotics ever losing their use except in the case of a few rare resistant diseases, which will be tested for and isolated. Problem solved.

These are all things we could do right now, if we wanted to, we just don't care about the problem enough. The moment it becomes big enough to interest the majority of the population, we know how to solve this.

A coworker who worked as a doctor in Afghanistan and Myanmar for about a year told me that a third of the medications they had in the hospital, and two thirds of the medications you could purchase in pharmacies contained no active ingredient. People kept dying on him because an antibiotic would randomly do nothing, or lysis medication turned out to consist of colored NaCl. He said that most of those medications were delivered from China and India because of the low prices in those countries. This is anecdata, obviously, but I'd be very careful purchasing drugs from countries with low oversight. Fake drugs are an enormously profitable market.

I also like this quote from a New York Times article:

“If I have to follow U.S. standards in inspecting facilities supplying to the Indian market,” G. N. Singh, India’s top drug regulator, said in a recent interview with an Indian newspaper, “we will have to shut almost all of those.”

The unease culminated Tuesday when a top executive at Ranbaxy — which has repeatedly been caught lying to the F.D.A. and found to have conditions such as flies “too numerous to count” in critical plant areas — pleaded with Dr. Hamburg at a private meeting with other drug executives to allow his products into the United States so that the company could more easily pay for fixes. She politely declined.

http://www.nytimes.com/2014/02/15/world/asia/medicines-made-...

Having been in quite a few mildly scary crowd situations at concerts, "elbows akimbo" seems like the worst possible idea. The main force will be coming from the front and back, but your elbows will be out to the side, so people in front and back will still be able to crush you just as badly, while your arms will get trapped between the people standing beside you, giving you even less mobility than you'd otherwise have. It also seems like a good way to get your arms dislocated in a surge.

In my experience, elbows crossed in front of your chest will give you a little bit of extra breathing room, but even that doesn't do much.

Wait, what? In Germany, you're expected to tip your waiter ten percent for good service. Not tipping is only done for terrible service. Including in rural and traditional areas.

Sourse: German in a rural and traditional area

Parkinson's long list of possible symptoms includes both movement disorders and dementia.

"Later that evening, while walking home, she noticed bilateral foot drop and foot numbness, which caused her to trip and fall. She spent several hours lying on the ground before she was found."

This last sentence raises a lot of questions if you think about it. We're talking about a presumably healthy 35 year old woman here. (At least, if she had any prior medical problems, the article does not see fit to mention it.) This woman experiences numbness of both feet and and an inability to lift her feet ("foot drop"). She falls. It makes sense so far.

But why does she lie on the ground for hours? She might not have been able to walk, but she should have been able to crawl - the article makes a point of mentioning that muscle power at knees and hips was normal. Was she unconscious, and if so, why? Was she in debilitating pain, and the article just doesn't mention it? Why didn't she take her jeans off once she noticed her legs swelling so much that her jeans were cutting off blood circulation to her leg muscles (which sounds pretty unpleasant!)?

I know this is just supposed to be a short summary, but it rises more questions than it answers.

The pages will almost certainly be based on word count. I believe the standard is 150 words/page.

Word count is probably the best measure to base this on, because time spent reading is totally arbitrary and depends on external variables, like how fast a reader the buyer is, how distracted they are while reading, and how difficult the text is. Personally, I'd think it makes more sense to pay us per 1000 words of text read, but if they want to call 250 words of text a page, that essentially accomplishes the same thing.

You generally want to wait six weeks after a heart attack before implanting an AICD, because the patient's heart frequently recovers to the point where an AICD won't be necessary at all. The LifeVest is supposed to bridge those six weeks.

We don't put the patient's name on the EKG, so it's not directly identifiable, and we get their consent before submitting (if they're in any state to consent, which often they aren't), but I'm sure it does violate all kinds of rules and guidelines, if not actual laws.

For an unstable patient, our process is much the same - the EMTs will put in the IVs, and we'll bypass the ER to proceed directly to the cath lab. For stable patients, the EMTs will prioritize getting them to the hospital as fast as possible and often won't have the time to get the IVs into place before they arrive. (This part of the county has a ridiculous density of hospitals, so the ambulance ride is frequently not even five minutes long, which doesn't give the EMTs that much time to work.)

We've found that things go little more smoothly if we do some of the prep in the ER, where there's a little more space for people to undress the patient etc. without worrying about bumping into the expense catheter equipment or compromising sterility, but even stable patients aren't supposed to spend more than five minutes there before being rushed to cath.

I do wish we had some way to directly transmit EKGs and vitals. American hospitals seem to have more money available for electronic equipment like that. (On the other hand, in Germany, the whole hospital stay including the catheter and a day in the ICU, probably won't cost you more than 10.000 Euro, and usually less than that.)

A month ago, a friend who'd worked as a doctor in South Africa for decades was interning with me to learn about the procedures in our German ER. I remember the first time a patient with a heart attack came in while she was there, and the speed of the proceedings just blew. her. mind.

What happens, if things go well, is this:

The EMTs transmit a picture of the electrocardiogram to us[1]. We confirm that it's a STEMI (the kind of heart attack where speed matters the most, as opposed to an NSTEMI, which isn't quite as urgent), and alert the catheter lab and the attending cardiologist. Half the ER personnel drops everything else to prepare for the patient's arrival.

The patient arrives. We transfer them to our gurney. One nurse gets the patient's clothes off (with scissors if necessary) while another nurse attaches electrocardiogram electrodes and a third shaves the puncture site for the catheter.

The ER physician puts in an iv line and draws blood while informing the patient of the treatment plan. The cardiologist takes a very quick history of the patient and confirms the STEMI with the second electrocardiogramm. We wheel the patient up to the catheter lab and get started.

Altogether, if everything goes well, it's something like five to twelve minutes from the moment the patient comes through our doors to the moment we've got the patient on the operating table.

[1]By taking a picture of it on their phone and transmitting it via Whatsapp, which is just about the ugliest, hackiest and probably borderline-legal way of transmitting patient data imaginable, but it gets the job done.

If the watch can learn to detect irregular pulse/atrial fibrillation (hell, even if it only recognizes it a fraction of the time) it's going to have a huge impact on strokes, too.

This might actually be partly generational, if you, like me, grew up using a calculator. I must've done that test with probably about a hundred patients by now, and I have to tell you, the average halfway mentally competent seventy year old can do it faster than me. It's particularly embarrassing when they make a mistake early on, and then I can't just look down at the test sheet in my hand for the correct answers to see if the rest of the numbers are correct.

That said, like other commenters have said, it's supposed to be a little bit hard. This exercise is part of the mini mental state examination [1], which is intended to be a screening test for the early stages of dementia, so you want to challenge people a little. The math part is actually the second-hardest task - the most difficult part for people with even very mild cognitive problems seems to be drawing the two interlocking pentagons.

1: http://en.wikipedia.org/wiki/Mini%E2%80%93mental_state_exami...

I'm in Germany, my tax rate is ~40%, and we do get all those benefits.

As a non-technical user with a newish Lenovo laptop, is there some way I can make sure I'm not affected by this?