Machine interpretation of EKGs

[sas5814]

I always read my own EKGs before looking at the machine read but I notice they machine interpretation often varies from mine and sometimes I can't find any criteria to explain the machine's reading. So I was curious as to others impression of the value and accuracy of the machine's interpretation.

[EMEDPA]

yup, trust your instincts. Just had an ekg that the machine read as V-tach(see in em photo quiz section). It missed the pacer spikes. certainly a different treatment for a fast paced rhythm vs vtach. hydration vs cardioversion...

[GetMeOuttaThisMess]

I used to do the same, read it myself first and then see if it points out anything hokie.

[UGoLong]

I try to figure out what the machine said what it said. It doesn’t work well with low voltage QRS or artifact. I don’t recommend ignoring the interpretation, but just to rule it in or out after you’ve made your own call.


Sent from my iPad using Tapatalk

[thinkertdm]

The machine just takes a voltage pattern and compares it to a number of stored patterns.  Unlike a machine, you are able to think a step ahead and see more complex patterns. in short, if you ignore the tracing and only read the machine interpretation, better have a good malpractice lawyer on speed dial.

[UGoLong]

The machine just takes a voltage pattern and compares it to a number of stored patterns.  Unlike a machine, you are able to think a step ahead and see more complex patterns. in short, if you ignore the tracing and only read the machine interpretation, better have a good malpractice lawyer on speed dial.

The same might be said about at least looking at the machine interpretation and reconciling it with your own.

Two heads (even if one is a less-capable machine) are better than one.


Sent from my iPad using Tapatalk

[ventana]

my machine loves to read afib..... even when not there...

[Guest JMPA]

Machine interpretation of EKGs ? Simple, don't do it.

 

[UGoLong]

Machine interpretation of EKGs ? Simple, don't do it.

 

I don’t, but ignore any input at one’s peril. And operating by rules in big type is a bit more dogmatic than I feel comfortable with.


Sent from my iPad using Tapatalk

[quietmedic]

It's a bit of an art... learning how to take the machine interpretation for what it is.  As others mentioned, with low voltages and noisy baselines, the machines go absolutely nuts and are useless. Other times, they might even pick up something subtle that I wasn't looking for, like an extremely borderline High QRS voltage by some criteria (for whatever it's worth).  The actual algorithms the machine uses to recognize things is complex, I have seen the sort of pseudo programming algorithm online somewhere, but if you just go through wikipedia or life in the fast lane, pretty much all the numbers are there, as well as the most common criteria used (LVH for instance has more than a handful of calculation methods, so it varies). As someone else mentioned, it's great to try to figure out in reverse why the machine is reading something... Usually you can figure out with one of the two sites above, but of course, part of the art is learning what to ignore...for example, EKGs tend to reeeeealy overread "anterior infarct" on a massive scale in my experience, so you kind of have to eyeball it and say, oh, I see what it was reading in V1 and V2, but it's nothing. Like I said, it's sort of an acquired art, it definitely helps to have a cardiologist or electrophysiologists around to run it by, so they can give you the real-world reality on what to ignore and what not to ignore.

[Guest JMPA]

here is the issue, ekg machine interpretation can lead to inherent bias. The ekg is to important a tool to skew interpretation. In my opinion a methodical approach with a clear mind and without assumption leads to the optimal interpretation.

[UGoLong]

here is the issue, ekg machine interpretation can lead to inherent bias. The ekg is to important a tool to skew interpretation. In my opinion a methodical approach with a clear mind and without assumption leads to the optimal interpretation.

To me, it’s like asking your student what he or she thinks and then critiquing the answer. As a preceptor, I really am not hung up on wanting to agree with a wrong conclusion by my students.

But to each his own.


Sent from my iPad using Tapatalk

[bike mike]

I usually don't pay much attention to the machine's interpretation. Wrong too often. Can't tell you how many STEMI's I've been called to for someone in atrial flutter and the ECG reads anterior STEMI.