AI can now identify atrial fibrillation through sinus rhythm from The Lancet Published:August 01, 2019

[holyshedballs]

Atrial fibrillation is a substantial health-care challenge and is considered to be a global pandemic, as prevalence rates have increased greatly 1 and atrial fibrillation-related hospitalisations outnumber those of major cardiac conditions such as heart failure and myocardial infarction. 2 Atrial fibrillation confers an increased risk of stroke and mortality; it therefore needs to be detected not only to manage the arrhythmia but also to prevent comorbidities and death. 3 A 10-second, 12-lead electrocardiograph (ECG) in current clinical practice is unlikely to reveal possible atrial fibrillation if not present in this short monitoring time. Silent or undetected atrial fibrillation is common and the few screening methods available are demanding in terms of time and resources. Continuous monitoring by means of loop recorders is often indicated, particularly in case of embolic stroke of undetermined source (ESUS). 4 Novel and user-friendly wearables to identify arrhythmias have emerged with recent digital advances: wearable ECG technology using automated photoplethysmography algorithms have shown feasible and accurate cardiac rhythm detection and can aid in monitoring the dynamic burden of time spent in atrial fibrillation, 5 while mobile atrial fibrillation applications are available for patients and health-care professionals for education and guidance in management. 6

1.Chugh SS Havmoeller R Narayanan K et al.

Worldwide epidemiology of atrial fibrillation: a Global Burden of Disease 2010 Study.

Circulation. 2014; 129: 837-847

2.Gallagher C Hendriks JM Giles L et al.

Increasing trends in hospitalisations due to atrial fibrillation in Australia from 1993 to 2013.

Heart. 2019; (published online April 1.)

DOI:10.1136/heartjnl-2018-314471

3.Kotecha D Breithardt G Camm AJ et al.

Integrating new approaches to atrial fibrillation management: the 6th AFNET/EHRA Consensus Conference.

Europace. 2018; 20: 395-407

4.Sanna T Diener HC Passman RS et al.

Cryptogenic stroke and underlying atrial fibrillation.

N Engl J Med. 2014; 370: 2478-2486

5.Dorr M Nohturfft V Brasier N et al.

The WATCH AF trial: smartwatches for detection of atrial fibrillation.

JACC Clin Electrophysiol. 2019; 5: 199-208

6.Kotecha D Chua WWL Fabritz L et al.

European Society of Cardiology smartphone and tablet applications for patients with atrial fibrillation and their health care providers.

Europace. 2018; 20: 225-233

7.Attia ZI Noseworthy PA Lopez-Jimenez F et al.

An artificial intelligence-enabled ECG algorithm for the identification of patients with atrial fibrillation during sinus rhythm: a retrospective analysis of outcome prediction.

Lancet. 2019; (published online Aug 1.)

dx.doi.org/10.1016/S0140-67...

8.Kottkamp H

Human atrial fibrillation substrate: towards a specific fibrotic atrial cardiomyopathy.

Eur Heart J. 2013; 34: 2731-2738

9.Martinez-Selles M Masso-van Roessel A Alvarez-Garcia J et al.

Interatrial block and atrial arrhythmias in centenarians: prevalence, associations, and clinical implications.

Heart Rhythm. 2016; 13: 645-651

10.Chua W Purmah Y Cardoso VR et al.

Data-driven discovery and validation of circulating blood-based biomarkers associated with prevalent atrial fibrillation.

Eur Heart J. 2019; 40: 1268-1276

11.Roselli C Chaffin MD Weng LC et al.

Multi-ethnic genome-wide association study for atrial fibrillation.

Nat Genet. 2018; 50: 1225-1233

12.Hannun AY Rajpurkar P Haghpanahi M et al.

Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network.

Nat Med. 2019; 25: 65-69

[Teresamarie]

Great job in tagging a fib with the need for longer heart monitoring. I’ve had two strokes and am frightened to have another that may cause great damage. I need to look further into this. Just got to get the energy to do it! I hit the wall yesterday!

[helvella Administrator]

Thanks for posting.

My thoughts:

What will be done for those found positive for a-fib? Let me guess, statins and a beta blocker. I hope that doesn't sound too jaundiced but it seems realistic. What would be the best possible intervention?

The intro refers to wearables (I assume Apple iWatch but likely some other products as well - I see one trial used the Samsung Gear Fit 2). I couldn't quite understand if wearables are superior to this AI technique? I think so because they don't just identify those susceptible to a-fib but also exactly when including duration (so long as the person is wearing a charged device).

[holyshedballs in reply to helvella]

I see this as another step in patients taking control of their own health and being able to challenge GPs regarding AF.

One of the reasons GPs want to leave patients on insufficient doses is the fear of AF. Being able to monitor the heart should give the GP some reassurance that the patient is competent to monitor their own health and to some degree counter that argument.

The NHS is also going in that direction - advising GPs etc. to use patient patient generated data .

And that is whole new debate!

[holyshedballs in reply to helvella]

I was looking on line for a wearable AF monitor. I saw the apple watch and thought ill look at that. I looked for another when I saw the price - from £399 for the 5 and from 329 for the 4 with AF plus the watches need a modern iphone to hoop up to.

I looked at fitbit but they aren't doing AF monitoring right now. They said they are trying to develop a different method of tracking AF because they think that current wearable AF monitors give too many false positives. techradar.com/uk/news/fitbi...

I don't know if that is true.

[Hidden in reply to holyshedballs]

I have heard of Alivecor (I think) and it’s not quite so expensive. Can use with a smartphone.

[holyshedballs in reply to Hidden]

I use alivecor with my iphone 5s. I think its quite good for what it is.

[helvella Administrator in reply to holyshedballs]

I bought one of the least expensive monitor "watches" - a Xiaomi Mi Band 2 - for around £20. Within its limits, it is quite good but definitely not at afib monitoring standard.

However, I did read that the company is producing a more expensive (towards £100) and much more sophisticated model which just might fit the bill. Unfortunately, it was not going to be under the Xiaomi branding - and I can't remember what brand they were going to use.

One thing I like is that mine has a fantastic battery life - none of this recharge every day business!

[helvella Administrator in reply to holyshedballs]

I've found it! The company name is Huami.

They have launched the Amazfit Smart Watch 2 and, more relevantly, the Amazfit Health Watch. Sorry about the none-too-special website at the end of this link:

gsmarena.com/huami_unveils_...

Xiaomi have also launched the HiPee Smart ECG Wizard. Not a watch, but a device which enables simple ECGs to be taken and transfrred to a phone (or whatever). Potentially very useful. And a very oddly chosen name.

c.mi.com/thread-2335183-1-0...

Seems to be about $58.

[Hidden in reply to helvella]

I understand blood thinning is one of the main treatments - in people assessed to be high risk of stroke. As the fibrillation (wibbling eratic muscle movments) can cause clots to form inside the heart, which can migrate to the brain causing stroke.

‘If you have atrial fibrillation, you are five times more likely to have a stroke. And atrial fibrillation contributes to just under 20% of all strokes in the UK.’ From stroke.org.uk