Slashdot Mirror
Build Your Own ECG
Jason writes "I finally finished documenting my $4 home made electrocardiograph (heart monitor). If anyone is interested or wants to build one for themselves, please come by and take a look. Makes me wonder why medical care costs so much. :)"
PLEASE be careful with ECG or EEG circuits, especially if you're planning to use an oscilliscope to see the wave or a data acquisition board to log the data. The pads and the gel used to adhere them to the skin and lower the resistance to get a good signal can sometimes cause current to flow into the body, especially if the circuit is not optoisolated.
Normally, the skin resistance is high enough to make the current flow negligible; however, when the pads are on, the resistance in the path is very low, and you could seriously injure or kill yourself if even a small amount of circuit flows 'back' through the electrodes.
Professional ECG machines usually have a lot of protection circuitry on their front ends (the instrumentation amplifiers) as well as between the amplifier and the ADC/output circuits to prevent this from happening. This is obviously even more critical in line- (i.e. 110V or 220V-) operated devices.
Karma: Excellent Birds (mostly as a result of listening to Laurie Anderson)
Makes me wonder why medical care costs so much.
Damn near everything used in a hospital has to be certified to be used for medical purposes. People's lives are at stake, and you have to be sure that your device operates within tolerances, doesn't crash, doesn't electrocute people, etc. It costs money to think of every possible problem and design a device around that. Also, as other people have said, the people that run these things are some of the most educated people in the world. Try to start a hospital with your $4 device and let me know how it goes.
EXACTLY! Just what I was thinking when I saw this.
.1 amp to kill you dead, and about .01 amp can interfere with normal heart operation. Normally, skin runs about 10 to 100 kohms resistance - to get 10 milliamps you would need about 100 to 1000 volts delivered across the chest.
Kids, DO NOT TRY THIS AT HOME.
Real medical gear has full galvanic isolation - that means there is NO current path that goes from the patient's body to the equipment - the signals pass through either an isolation transformer, an optocoupler, or a capacitive coupling. That way, any ground leakage in the equipment won't fry the user.
It takes about
When you put the gel on, you reduce the resistance to a few hundred ohms. Now you need only a volt.
Normal consumer equipment can have "leakage currents" - places current shouldn't be flowing but is. You hook your home-brew circuit up to the printer port on your PC, and maybe you are OK. Then one day, while screwing around with it, a cap starts to fizzle in your power supply, or maybe you reach up to adjust your monitor, or maybe you put your foot on the ventilation register. Then you get to start (posthumously) on the 6 o'clock news.
At a MINIMUM, you should power the circuit with a nine volt battery, and communicate with the PC via an opto-isolated RS-232 link.
Even better, splurge and get the real medical isolation amplifier modules. Yes, they will cost a bit more than US$4, but then, if that is all the value you place on your life....
On second thought - go for it! And make sure you clip the ground lead off your computer's power cord while you are at it. And do it in the bathtub - that will help shield the fnord rays out.
www.eFax.com are spammers
There are several reasons healthcare is so expensive.
1. Litigation. Does the phrase "malpractice insurance crisis" ring a bell?
2. A side effect of (1.) is something called "CYA medicine". Ever receive a chest CT because your heartburn just might be a pulmonary embolism? It happened to me just recently.
3. Failed accountability. This one takes a bit more explaining. Ever bother to look at your bill? Of course not. Why? Because chances are, the insurance company pays it. Aha! You say. What if I'm not insured? Well then, many people who aren't insured "spend down" and go on Medicaid. Once more, nobody looks at what Medicaid is being billed, except for the hardworking beurocrats (cough)bull***(cough). Only the very narrow slice of the population that is "self paying" actually looks at a bill (more on thatlater) You would think that insurance companies would be on guard for their bottom line, but corporate inefficiency is often no better than government inefficiency.
4. Complex and inefficient billing. Health care is one of the few businesses where you receive service at a single location, yet billing goes from subconractors directly to insurance companies or patients. Worse yet, billing from some contractors takes weeks, or even months. Yuck! Imagine if every business worked like this. Imagine getting your car fixed, and you get bills from the mechanic, the parts department, and the oil supplier spread out over 2 months. It's not just inconvenient. It actually hinders your ability to make financial plans because you don't know what's coming. And why don't you know what's coming? That leads us to...
5. ...Secrecy. That's right. Secrecy. Try to call up a hospital and ask them for their price list. Chances are, you'll get the same answer I got: "That's on a computer and it's confidential". I was transferred to a manager who had her phone on voicemail. In retrospect, I should have known I was in trouble when the phone tree had "press 2 if you're an attorney". This is probably one of the biggest reasons healthcare costs too much. Sure, there are several hospitals within driving distance, but if I think I am going to need an exam that is likely to involve half an hour with a doctor, some medication, and an x-ray... I have no idea who charges the least for an x-ray, or what the hourly billing rate is for a doctor, or what the average examining time is for diagnosing a condtion. We have more accountability at the garage than we do at the hospital (Chilton's guides, posted labor rates, etc).This alone is probably the single biggest factor driving up healthcare costs. Lack of pricing information makes comparison impossible, resulting in a virtual monopoly even though there are multiple companies. So, what did I do? I gave up and paid a price that I could not verify as accurate because I knew that the only way to get the price list was to make a federal case out of it, and spend 100 times more in legal fees than my bill was.
6. Vested interests. You can call me a conspiracy theorist if you like, but I think various interests want the price of healthcare to skyrocket so that they can use that as an excuse to socialize it. The corporations actually secretly like the idea of socialized medecine, because then they get to become government agencies. If you are a corporate sleazeball, the next step up is to become a government sleazeball; the perks are just that much better. You can just hear them salivating.
Want to fix healthcare? Fine. Require providers to give one bill in a timely manner--no pass-throughs to subcontractors. Require providers to post price-lists online if they have a website, or to make price-lists available to the local libraries. Require employers who insure their employees to provider high deductable insurance. There should be no claims or forms until annual out-of-pocket costs exceed 10% of your annual pay. Place a cap punitive damages, as many have suggested.
For all intensive purposes, "whom" is no longer a word. That begs the question, "who cares"?
Well, issues of "medical care" and "medical equipment" costs being two rather separate things(one of the biggest costs in medical care is liability insurance, probably followed by administrative overhead)...
...it might be because the expensive REAL version won't kill you when there's a lightning strike nearby, or when someone touches the case after building up a static charge, or something shorts out in the computer half...the list goes on. It only takes a few microamps to stop your heart- it's all in the path the current takes. Having those nice electrodes in the right places, making great electrical conduct with your skin...well, umm...you should get the picture.
Medical equipment is designed to be 'bulletproof' in almost every way- there's a standard, for example, for medical-rated Edison plugs and sockets.(Edison plug = US electrical plug). It's VERY heavy duty, makes really good contact, has excellent stress relief on the cord, etc...because something VERY important might be using it, like an artificial heart pump in an operating room, or a dosage machine for an IV, or a ventilator. The REAL version also can't crash or stop working- so, for example, if it has a computer, the instruction code, the chips...everything is heavily tested. Jokes aside, the Pentium math dividing bug is a perfect example of why you can't just use "anything" for medical equipment. What if that bug caused the heart monitor to display the wrong heart beat rate? Electronics used for medical equipment get a LOT more testing- lives are at stake. Same idea behind the MIL specs, although with MILSPEC stuff, the idea is more that the military really abuses the crap out of stuff ON TOP of similar concerns as medical stuff.
All of the above are why you often see these days disclaimers from chip makers that say "this device is not certified for use in life support equipment" and such. The statement often extends to industrial automation- "situations where malfunction may result in injury or death", stuff like that. Ie, "don't use this where if it screws up, it dumps 10 tons of molten steel on a bunch of steel workers' heads."
Please help metamoderate.
http://saintaardvarkthecarpeted.com
Carousel is a lie!
Where computer-automated reading of ECG is much more effective is in the automated defib units that you might find in airplanes, malls, ambulances, etc. The computer is good enough to know if a jolt of electricity would be helpful in correcting the problem.
Of course, the expense and the value of ECG is in the physician interpretation. Likewise, an aspirin in the hospital will cost you dollars instead of cents due to administration costs, nursing costs, insurance, etc.
What has really helped ECGs is fax machines and now the internet. If a health care professional has a question regarding a waveform, they can get it to someone who can give an accurate reading. Soon, if not already, physicians will be able to see the ECG live (and in living color) of patients on planes...
And quit dogging the editors for posting this... We need more tech-related medicine news. It warms this geek MD's heart.
Davak
If you can, please mirror my mirror. I'm sure a large number of slashdot readers have servers available they can put to good use.
I am a third year medical student at a state medical school in the United States
This is a multifactoral problem. The average medical student graduates with about $180,000 debt (closer to $220,000 by the time it's paid off). Monthly payments can be about $1,000 per month.
Secondly, technology has largely supplanted physical diagnosis. Doctors have gotten sloppy in the past due to reliance on expensive diagnostic tests rather than relying on patient's history and their 5 senses (Greeks used to taste urine to diagnose diabetes). This problem is being addressed. Medical students are trained more like engineers in that cost to the patient (and by extension the system as a whole) is an important consideration in ordering tests. Of course, the health of the patient is paramount. For example, a good abdominal exam can obviate the need for an expensive CT scan.
The legal system in the US contributes in no small way to the cost of health care. Professional liabilty insurance ("malpractice insurance" to the laypublic) premiums can range upwards of $100,000 per year for high-risk specialty. An OB/GYN I know in Florida was recently offered $250,000 of coverage at a $216,000 premium. He is now practicing without coverage. Doctors pass these increased costs onto patients.
Lastly, medical equipment is held to tremendously high quality control standards. From my software engineering classes, I seem to recall that the importance of reliability testing was consistently invoked by mentioning areospace and healthcare equipment. If your $4 EKG misses one MI, you've got a big assed lawsuit on your hands (which we all pay for through increased costs).
My $0.02
--b
There was something like this in Scientific American a while ago, only it used an instrumentation amplifier instead of an op-amp, which would reduce common mode interference and give you a cleaner signal. Analog Devices makes a cheap one, AD620 which sells for around $5 and wouldn't raise the cost of the project too much. They also make an expensive one, AD624AD, around $20 with better gain and better common mode rejection, still not outrageously expensive. some diodes to give some protection against getting electrocuted by a power surge might be a good idea too, for those girly men who can't handle their electricity. you can buy real ekg electrodes cheap too. Of course, you'll blow your entire budget hiring a cardiologist to interpret the results.
Scientific American published such a circuit in their Amateur scientist area a few years ago... link to article .
It is a commonly referenced site.
No no... I appreciate the input of you guys that actually make the equipment function... the Biomed guys regularly save me (by keeping my equipment running). I couldn't do my job nearly as effectively if it wasn't for the biomedical engineers. I don't mean to make fun of anyone's hard work, and I'm probably being too hard on the people that program these things. Maybe it's the case that the interpretation software is simply in its infancy.
Perhaps I should expand on my initial comments. A previous poster pointed out that the first thing you are taught in medical school is to ignore the machine read... that's true, and medical students are still taught that way (I teach in an academic setting, and I teach my students the same thing).
As I understand the machine's algorithms (if somebody who programs these things wants to correct me, please do), they interpret the waveforms based on an ideal model, and attempt to interpret current-of-injury patterns, based on deviation from an expected baseline.
Many situations make the machine read useless (and to be fair, extraordinarily difficult)... any patient in a paced rhythmn (pacemaker, single or dual chamber), the machine will default, and not give a read. An excellent call for the engineer that designed the machine... reading injury on some paced rhythms can be very sticky, even for an experienced clinician. Some of these machines regularly read "digitalis effect".. a difficult call, particularly in a suspected ischemic or strain-induced ST depression. A noisy baseline (in a patient who's shaking, for instance)will often throw off the machine. Many patients who have known cardiac disease have EKGs that are difficult to interpret, and injury that can only be discovered based on comparison to a previous EKG.
What I'd like to see is a program that compares old EKGs to new ones, and automatically gives you a change summary (in addition to the tracings themselves, naturally). I could see that being very useful, particularly if it uses the previous EKG tracings to redefine "normal" for itself. That might help the over-sensitivity problem. Many people are walking around with tracings that are nowhere near the classic "normal," but are normal for THEM. What I'm saying is that I'd like to see a program with a dynamic "normal;" one it can redefine on the fly.
It's not to say that I don't read the machine's interpretation... I do, but I subordinate it to my own clinical interpretation. To be fair, I have the luxury of knowing the history... something the machine may never know, and as any physician will tell you, history makes the diagnosis 80% of the time; the tests are simply to confirm what you suspected all along.
Maybe if you frame it like that, EKG machine reads don't really need to be perfect... they are, after all, just an adjunct... A human still makes the decisions.
Even if a man chops off your hand with a sword, you still have two nice, sharp bones to stick in his eyes.
It's easy to say that it's possible to automate a lot of this stuff. It's harder to actually implement the automation. Philips owns Viridia this week, and their system is essentially the best on the market.
Unfortunately, the *best* in the market can only interpret QRS complexes. ECG techs still have to go into the system by hand, and label the PR, QRS, QT, and RR intervals. This is still quite an active area of research.
And for the record, they're nowhere NEAR 95% accurate. 70% would sound more accurate. And I would know, having been an ECG tech at a the local teaching hospital. The hospital spent $110,000 on the system and we were so unhappy with it that the engineering team came down from Massachusetts.
The rub? They're using a statistically-based FFT program, and their sample set is ~250 AHA ECG recordings. Humans who are good at ECG interpretation need to expose their neuro-fuzzy brains to at least 2,000 ECG's in order to know what they're doing. And they wondered why we were having problems.
When the head engineer got up to speak, he made an interesting comment. He said that when he was a student at MIT, a physician from Beth-Israel Deaconess Hospital came to the engineering department and asked if they could analyze ECG signals. They looked at them, and seeing their simplicity, said of course. Thirty years later, he's still working on the problem.
You'd think he would have tried something other than the FFT by now, but he hasn't. So much for thinking outside of the box.
--V--
Whether it be $4 or $10, it's a helluva lot cheaper than the roughly $38,000 that people around the country are paying for Lifepak 12 monitors. And I would know, being a paramedic. I actually had to do quite a bit of ECG processing during my stint at the local college to get a bachelors degree...I can't decide which is worse for ECG processing, Matlab (which I used) or Visual Basic. Matlab is nice, but at $3000 a pop, Visual Basic is the cheaper way to go. Matlab's built-in functions definitely helped tho (***cringe*** at the thought of implementing the FFT in VB).
Perhaps the biggest problem with ECG systems is noise filtering. Adaptive filtering has failed miserably; you'd think the engineers designing the systems would implement Bass and Treble-style filters for healthcare providers to use, but condescention has gotten the better of them. Noise makes interpreting an ECG sometimes impossible, yet no such filters. This presents a problem for me on a weekly basis - it is sometimes impossible to interpret the rhythm, because the filters aren't working properly. If they'd only let me adjust the lowpass and highpass cutoff's, I would feel SOOOO much better. And if they'd give me a Savitsky-Golay filter, I'd have a wet-dream....here's hoping.
Yes, it's true, and if you do enough searching from that link (you gotta go all the way down for your link) you'll find what happens if you're affected badly by colloidal silver: silver colored skin