I was unable to upload this to the original thread by erin_jonesx but here is a brief explanation of artefact/noise.
On the top trace I have highlighted normal beats and you can see that they are spaced out normally (time is represented horizontally)
You may need to look closely.
On the bottom trace I have highlighted what I believe the device has counted as the heart rate and misjudged to be AF, due to their frequency. There are probably 3 x more of them.
It has actually included features of a normal cardiac cycle in the count. However, they have been amplified by movement, sweaty fingers etc, ie artefact.
I hope this is clearly explained here and I don’t mean to tell anyone anything they already know. My aim is to put minds at rest.
Later I will post a brief explanation of normal ECG features and also what AF looks like 💚
Written by
Fibfab
To view profiles and participate in discussions please or .
Artefats are all too common with Kardia as I understand that it has an acoustic function and can pick up all sorts of extraneous signals. I found that any electronic equipment could interfere with mine (a very early version I was given to test when thye first came to, UK ) and being any where near my computer was a no no. Whilst I can recognise simple arrhythmias on ECG I watched an established specialist in ECG interpretation destroy a room full of doctors once and learned that it is all too easy to get it quite wrong even for supposed experts.
Hi Fibfab, I don't think you have used a Kardia? Kardia doesn't count from the 'picture', it counts the electrical signals from the heart which can clearly be seen if you convert to PDF. You can have a lot of 'frizzle' showing on your recording but only the genuine R wave will be indicated. If there is too much artefact you get a message to that effect depending on your settings. This is why it is considered reliable by cardiologists and EPs.
I was prompted to get a Kardia by feeling pauses in my pulse which I thought just be faint beats which I missed. My Kardia confirmed pauses up to 5 secs after which my medication was reduced and I was given the choice of a pacemaker or ablation based on my recordings and reports of presyncope. I don't think that would have happened if Kardia recordings were interpreted as you suggest.
Kardia has an education section on the app demonstrating common rhythms and ahrrythmias.
I have a Kardia which I use to demonstrate to patients. We have them in our department.
All of these signals, artefact included, will be picked up by it and the interpretation is affected by this. This can be seen on the trace above as the device has miscalculated the heart rate to be 160bpm.
I actually think they’re very good but they aren’t perfect and do not replace a proper 12 lead ECG. There is still much to learn about their place in diagnostic cardiology.
I marked the “picture” to show the signals which represent artefact and intrinsic rhythm.
Thanks Fibfab
DrDave01 is the inventor of the Kardia and responds here occasionally. I hope he will respond to this thread because I am puzzled that the Kardia reported “possible atrial fibrillation” when it doesn’t seem to be: my experience has been that when Kardia reports that, it always is AF. It seems I am wrong.
It would be good to hear when Kardia might be reporting false positives for AF.
This is erin-jonesx ‘s thread referred to by Fibfab
Apologies, I stand corrected on the use of the term artefact. I was attempting to explain that there had been a miscounting of the QRS complexes due to an error from the size of the other ECG components.
Again, just trying to put minds at rest that this does not show AF.
I know I have a something wrong with my t wave as my gp said I had a heart attack but cardiologist said no you haven’t.
I will wait for part 2 and hopefully be educated 😂
Thanks DrDave01
Just to be clear, is it the reduced height of the QRS here which has caused the software to report AF in error, as well as the doubled rate?
The context of our concern here was that a new member aged 18 was unwell with episodes of tachycardia. We advised her to buy a Kardia and within 48 hours her Kardia was reporting rates up to 236 with possible atrial fibrillation. Naturally she was concerned. Is there a way the Kardia reporting software can be tweaked to address the problems thrown up here?
Does Kardia ever report “Atrial Fibrillation” rather than “Possible Atrial Fibrillation”? If not, how should users interpret the results?
I find Lead II recordings much better than Lead I for picking up my periodic atrial flutter. I wonder if Lead II recordings are promoted enough in your Kardia literature.
Content on HealthUnlocked does not replace the relationship between you and doctors or other healthcare professionals nor the advice you receive from them.
Never delay seeking advice or dialling emergency services because of something that you have read on HealthUnlocked.
Update on Kardia picking up inverted t waves 
Kardia Accuracy
Help interpreting a Kardia report.
Using the Alivecor Kardia Mobile post ablation
