July 1, 2025 • 21 mins
Dr Ramesh Nadarajah leads a groundbreaking trial using machine learning to detect signs of atrial fibrillation (AF)—a hidden heart condition that dramatically increases stroke risk. Hear how new tech is helping doctors intervene earlier, and what it could mean for the future of stroke prevention.
Views and experiences shared in this podcast belong to the individuals and do not constitute medical advice. For medical advice please contact your pharmacist, GP or dial 111. In an emergency please dial 999.
This podcast was recorded and produced by Under the Mast– creative audio productions and was presented by Caroline Verdon
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:07):
Medical breakthroughs,the research journey.
Hello and welcome.
I'm your host, Caroline Burden,and you are about to join me on
a journey into the fascinatingworld of medical breakthroughs,
but not just any breakthroughs.
We are diving into thepersonal stories, the setbacks.
(00:30):
That you will not believe.
This moment behind the cutting edgeresearch happening right here at
Leeds Teaching Hospitals NHS Trust.
Coming up in this episode,
the AI algorithm is called FIND af.
It was developed by ourselvesat the University of Leeds.
We developed and checked 2million people in the uk.
(00:50):
We've now done that in 9 millionpeople in other countries.
So this algorithm could gomuch further internationally.
Globally and it can actuallybe something that we export.
Can an algorithm predict a stroke?
What started as an idea in Leeds couldchange the face of healthcare worldwide?
My name is Ramesh Raja.
I'm a academic clinical lecturer at theUniversity of Leeds and a cardiology
(01:13):
registrar at Leeds Teaching hospitals.
So atrial fibrillation is, uh,an abnormal heart rhythm that
comes from the top of the heart.
It's very common affects about 1.5million people in the uk, and it's
becoming more common so that thereare now more cases diagnosed of atrial
(01:34):
fibrillation each year than the four mostcommon causes of cancer put together.
Wow.
Yeah, so it's very common and it isimportant, so it causes symptoms like
breathlessness, the feeling of your heartracing, uh, chest discomfort, fatigue,
but also, and probably more importantly,it increases your risk of stroke by
(01:55):
five fold, increase your risk of heartfailure, and increase your risk of death.
All of these things, ofcourse, very important.
Um, and when you put that all together,you have a very common condition,
which is becoming more common and isimportant in terms of for the individual
and for the health system as a whole.
Um, and really.
(02:15):
What we in this study are looking atis one of the big issues with atrial
fibrillation, which is it causes allthese problems, but it still causes
increased risk of stroke, increased riskof heart failure, increased risk of death,
whether or not it causes you symptoms.
So it can sometimes be abit of a silent killer.
Um, you can have no symptoms but haveatrial fibrillation and be at risk.
(02:38):
Um, so it's, it's, it's estimatedthat between a quarter and a
third of people that have atrialfibrillation, uh, are undiagnosed.
And
how treatable is it, you say it increasesthe risk fivefold if you were to treat it.
How does, how does the risk look then?
(02:58):
Yes, there's lots of treatmentsfor atrial fibrillation.
Uh, the ones that are aimed atreducing your risk of stroke,
uh, are tablets, which areeither once a day or twice a day.
Very well tolerated.
A number of different available optionsthrough the NHS, um, uh, and with a.
(03:21):
We can reduce that risk of stroke bytwo thirds if we find people with this
abnormal heart rhythm and we treat them.
And so that is, uh, what is behind thenthis, this study using ai and you say
AI today, and people think about Ja, youknow, DPT and is it gonna steal my job?
And are we gonna haverobots in the future?
And all the rest of it.
(03:42):
Um, it brings acrossa, a lot of questions.
How, how are you using it in this study?
So the, the problem we find is thatpeople, uh, have this abnormal heart
rhythm, but are undiagnosed and weactually have in the health service
lots of different tools that we canuse to diagnose the problem types of,
uh, devices that will take an ECG, uh,and show you your heart rhythm and.
(04:09):
Atrial fibrillation.
The question is, who shouldwe give those devices to?
Uh, they cost money.
Uh, they inconvenience people byhaving to take their recordings.
So we want to find the rightpeople, and that's what we're
using AI for In this study.
We are, uh, using AI in datathat's routinely available in.
(04:30):
Each of our, uh, GP records and usingthat to identify people at risk.
And it's those people at risk thatwe are offering the opportunity to
have, uh, rhythm recording to lookfor this abnormal heart rhythm.
And so if, um, if the results of thisstudy were to mean that actually this
would become routinely used, is thissomething that would replace, uh, you
(04:54):
know, a, a, a GPS role within this?
So as you say, people are always.
Potential implications of ai,uh, um, things such as chat
in very
this simply to helphealthcare professionals.
(05:19):
And more efficiently.
Uh, because at the moment we have noavailable tool in the NHS to allow
us to identify these people at risk.
What we're doing is trying to providethat information to primary care
healthcare professionals so thatthey can use it to inform their care.
So not replacing, hopefullyaiding, uh, and improving care.
(05:40):
And so how, how does the study work?
How many, how many peopledo you need to be involved?
Yeah.
So first of all, we, the, the,the algorithm, the AI algorithm is
called Find af, and that's becauseit's future innovations in novel
detection of atrial fibrillation.
And actually it was developed byourselves at the University of Leeds.
So this doesn't comefrom a private company.
(06:00):
This was developed through funding forthe British Heart Foundation by myself.
Doctors, a team of statisticians,a team of data scientists, all done
with the aim of improving care.
Um, and when we made it, we madeit, uh, using information from
(06:21):
over 2 million people in theUnited Kingdom, uh, all anonymized.
Uh, but this was information thatis available in primary care,
but we could use to identify.
Just to be clear as well,the AI is not changing.
It's not some sort of multi-headed hydra.
It's a fixed thing.
We trained it.
(06:41):
It's what, it's, it's not gonnachange, it's not gonna adapt,
it's not gonna do other things.
It's built for this purpose.
Uh, and it was built by scientiststo try and make things better.
Um, and now what we're doing in thisstudy with funding the Be Foundation
and from charities importantly.
We're actually taking it from theresearch setting, which is we did it in
(07:02):
data and actually to make a difference.
So in this study, uh, we have putthe algorithm into primary care
records, into the records that sitin your GP practices, and we've
identified people at risk and we'veidentified 2000 people at risk across.
And we've invited those individuals,um, to take part in the study.
(07:26):
Now, of course, whatdoes the study involve?
Is it a lot of work?
It really isn't at all.
We've tried to make this study as easyas possible for people to take part in.
Um, so we've deliberately, uh,centered the study in areas that are,
uh, are full of poor communities.
They're full of communities with ethnicminorities and places where historically,
(07:46):
uh, heart disease has been a. Uh, whathappened to the study is the AI algorithm
runs through the records, identifies theat-risk people, and those individuals
are invited by a text and by a letter.
If they want to take part andthey send the letter back to us.
Um, we actually send them a littlehandheld device, which put your thumbs
(08:08):
on, and they take the recordingsas they go about their daily life.
Four times a. So they neveractually have to leave their
house to take part in the study.
It all happens very remotely to tryand make it as convenient as possible.
Of course, you know, you have to takeyour recording a few times a day, and it's
easy enough to forget that once or twice.
(08:28):
But this little handoutdevice is very light.
It can fit in your handbag.
Um, it should be something that youcan do as you go about your daily life.
And you don't have to cometo see us in hospital.
You don't have to go forlots of, uh, invasive tests.
It's all done simply through that.
And of course at the end of thisstudy, once people have undergone
this period of ECG monitoring, uh,we then provide them with the result.
(08:51):
We tell them whether we found, uh,atrial fibrillation or we didn't.
And if they, we do find atrialfibrillation, we make sure that
their primary care, um, healthcareprofessionals aware and that
they're put on the right treatments.
And what's the study?
So the goal is to, uh, one, improvedetection of this ab abnormal heart
(09:13):
rhythm because as I said, a thirdof people who have it aren't aware.
So we know there's a, an unmet needthere, and we're trying to address
it by actually getting peoplediagnosed and gets them on the right
treatments before it causes them harm.
And also within the study, uh, weknow that different individuals
are at different levels of risk.
(09:33):
Uh, and what we want to understand is.
Is there a point at which the risks, uh,is so high that, uh, this will be the
most effective approach to use in the NHS?
This is all funded by research charitiesat the moment, but we want to find out
what's the optimum risk level at which weshould provide people with these tools.
Uh, that will then be something that wecan actually translate and use in the
(09:57):
NHS that will be effective in terms of.
Get them the right treatment,but also cost for the nhs.
So we're for the nhs.
And if it, you know, did get the greenlight and it did get rolled out, what
sort of timescale are we looking at tosort of see it, uh, in use generically.
(10:17):
Yeah, so actually in parallel to thisstudy going on, we've actually been
working with bodies within the N chestto see how we can translate this.
So we've done, we've done anumber of important steps.
One is we did a lot of interviews with,with GPS and people in, in primary care
to understand one, is it useful to them?
Which we found resounding Yes.
Two, how would theylike it to be deployed?
(10:37):
How do they want us to use it in NHS?
As we use that information we've puttogether the sort of pathway that would
follow, and actually we're currentlyworking, uh, with, uh, providers of
these, uh, GP records in primary care toactually make the algorithm available.
So we're working with, uh.
(11:03):
I mean, it, it sounds reallyinteresting and like it could actually
have really quite a significantimpact, um, upon people's health.
Yeah, absolutely.
So I think it's really importantthat people are aware that we did
this research one 'cause we'rea team that include doctors who.
Actively looking to make a differenceto care right now, not in 10 years,
(11:25):
not in 20 years, but right now.
And thus, even though we were using ai, wemade a lot of efforts to make sure that it
was implementable right here, right now.
And also that was scalable.
So, uh.
98% of the UK populationare registered with a gp.
That's a lot of people, right?
And all of those people havea gp, uh, medical record.
(11:48):
And this algorithm can interact withevery single one of those records.
So what we're aiming for is to makesomething that's truly scalable across
the country to help everyone in the, and.
And I think it's, um, it's reallyinteresting that, as you say, this has
(12:11):
come from the NHS, it's come from supportfrom charities, and it's come from LEED
University, leeds teaching hospitals asopposed to from a, an organization, a
business trying to, um, you know, helppeople but also benefit financially.
And of course, uh, we, we knowthat, uh, uh, it's important that
(12:33):
we have companies trying to do that.
Yeah.
Private companies coming inwith their medications, with
their devices, et cetera.
That's important for our innovation andhow we drive healthcare going forward.
But very much it, this, uh, thiswhole program of work, this aim to
improve has come from within, withinthe N Charities, informed by done by.
(12:58):
Scientist for.
The pure reason of tryingto make things better.
So it comes with the bestintentions, and that's why I
think that's really important.
You know, you hear all the time nowwith AI about ethics boards, about
are they doing the right thing?
Are they, are they, uh, doingsomething that's, uh, for
profit rather than for benefit?
I think we can be very clear thatwe, we say with a very clear heart,
(13:20):
we're doing this for people's benefit.
Uh, and that's why, uh, we help
this.
And how long is this, um, piece of,you know, research work from start
to hopefully, you know, soon finish?
Yeah, I mean this, this researchwork has been years in the making.
(13:41):
Uh, so, um, the fundingapplication to this went back
in to develop the algorithm.
Find AF was back in 2018.
We started working on it in 2020.
This study, uh.
Uh, started, uh, over a year ago now.
(14:02):
Uh, so this has been over half a decadein development to get to this point.
And my role in this was actuallythe, uh, it was my PhD funded
by the British Heart Foundation.
Uh, that led to thedevelopment of the algorithm.
So I was taken outta my clinical workas a doctor for a few years to do this.
To work with ai to work with, uh,primary care records to try for.
(14:35):
Part of my role as half clinician,half academic has been then trying
to drive this forward with a reallystrong, uh, multidisciplinary team of
lots of really important healthcareprofessionals and scientists as a
team trying to bring this forward.
That must be personally really fulfilling.
Absolutely.
It's the best part of the job.
(14:55):
Being able to look after patientsin, uh, in my NHS job, but also do
stuff that I actually think can makea stepwise change, a paradigm shift
change, uh, and a change at scale, uh,is, is really rewarding and seeing, uh,
the fruits of years of hard research.
To maybe actually make a differenceto people is super rewarding.
(15:16):
And I, and I imagine you, you are seeingthat now with, with the trial patients who
actually they have been diagnosed with af
Absolutely.
That's been a, a really rewardingthing in that as part of the study,
which obviously research, we actuallyhave found AF for some people and made
sure they're on the right treatment.
And actually we get really goodfeedback from patients, one
who've taken part in the studyand in terms of diagnosed happier.
(15:44):
They know they're on the right treatment.
They know they're being protected.
So often, you know the pointthat you see your GP is the point
where there is something wrongas opposed to future planning.
And I suppose this makes a ahuge difference 'cause it's
preventing those hospital stays,the heart attacks, the strokes.
Absolutely.
I think we can all see the NHS as understrain, uh, in from us within who are
(16:09):
take, who are part of the system to alsofor all the patients who come and see us
on a daily basis are constantly tellingme how busy things are and how, and
how and how under strain the system is.
And so, uh, and a clear, a clear, um.
Shift focus things when
(16:33):
before.
We'll know somebody's had a stroke,and we'll know how debilitating that
can be for the individual, how muchthat can affect the family, how it can
affect everything, responsibilities,financial stuff, et cetera.
It's not just a, a single thing.
This affects your life.
So if we can stop that for an individual,that's a hugely rewarding thing for us as
(16:53):
clinicians, but also a hugely beneficialthing for the NHS, both in terms of,
uh, the provision of that care, but alsoin terms of the cost of that care and
for the cost for society as a whole.
And obviously you, you, you arehoping that this will get the rollout.
And let's say once this has the rollout,do you have any other projects that you
are already thinking about for the future?
(17:17):
So, um, heart disease in general.
Is still alongside cancer, the, thenumber one cause of, of death in the,
in the United Kingdom, including thenumber one cause of premature death,
uh, in, in the United Kingdom and thecause of, cause of hospitalization and
quality of life, uh, being impeded.
So one can look at what we've done herefor atrial fibrillation and with fine af.
(17:40):
As effectively the first in a seriesof a number of these, uh, tools and
projects that we're running, we're alsodoing something very similar in heart
failure because, for example, in heartfailure, that's again super common.
More than a million people diagnosedin the uk, but you wouldn't believe
that eight in 10 of those peopleare only diagnosed after they're so
sick they have to come to hospital.
(18:01):
So that's a clear unmet need that wewant to diagnose people earlier before
they're so sick they come to hospital.
So we're doing a very similar thing, whatwe did with Find EF with heart failure.
We've made find hf where we'reagain using primary care records at
scale, identifying at-risk people.
And we're shortly hopefully gonna bestarting a study where we're doing
a very similar, uh, study, but fortrying to protect heart failure.
(18:28):
Diseases which severelyaffect people's lives.
So hopefully in years to come, you'llsee this as just a starting point of how
we then shifted everything and improvedearly diagnosis in the community to
make patients better and live longer.
That's phenomenal.
I hadn't even thought about thefact that you could then take the,
the technology and then transposeit for, for different illnesses.
(18:48):
Absolutely.
So the, the, the team.
In terms of the humans that go intomaking this, this, these technologies,
the principles behind how we do it andhow we test it, how we make sure it's
safe to be used in, in people we can, wecan replicate in different disease areas.
That's what we're currently doing.
Uh, and we know there are many, manydiseases that are diagnosed late in, in
(19:10):
the disease course at the point at whichthey really affected people's lives.
And we wanna try and get ahead of that.
And we want to try andshift everything we know.
That's what the government wants as well.
They want things in the community.
They want digital, not analog.
They want prevention, not not treatment.
And we want to try and address thosethrough a series of these partnerships
between clinical work, research work,charities, funding us, the national
(19:34):
law, health research funding us,and then taking this into the n hs.
And presumably it, it, it hasthe power to go further than the
NHS and actually go worldwide.
Yeah.
So that's something.
So Arthur, first goal, I'm a NHS Doctor.
Arthur first goal was toimprove care in the NHS.
Uh, but.
We have over the last five yearsdeveloped a really strong collaborative
(19:58):
network across many countries actually.
And so actually the find AF algorithmthat I spoke about that we, um,
developed and checked in 2 millionpeople in the uk, we've now done that
in 9 million people in other countries.
So this algorithm.
Much further internationally,globally, and it can actually
(20:19):
be something that we export.
This is, this is UK made,NHS University made.
We can export this to improve careworldwide, but also to bring back revenue
for, for the NHS, for the universities.
So a, a huge impact almostof unimaginable scale.
So that's what we hope to achieve.
We have to get there first.
But we, you know, we, we, we, we aim,we aim big because we aim to try and
(20:42):
improve things across the country, thechest, and then we aim to, to, uh, take
that model and apply elsewhere as well.
And, and be able, and you know, I saidatrial fibrillation is very common.
For example, this isthis one disease area.
1.5 million in people in the uk it'sabout 63 million people globally.
So that's a huge number.
We're talking about huge numbers here,and so we really are looking to do
(21:04):
things at scale, make big benefits,and potentially make a big difference
on a national and international scale.
But of course we need to get that first.
It's been five years development sofar, uh, and we need to keep going.
Coming up on our next episode,the advance now is that we have
personalized cancer vaccines.
So people who have been diagnosed withcancer, a sample of their tumor is taken,
(21:27):
the genetic material can be extracted,and the particular mutations of that
individual's cancer can be identified.
And the vaccine developed that ispersonalized in that it's directed
at targeting those antigens that werepresent in that particular patient.
And you can hear more about thebreast cancer vaccine from Professor
(21:47):
Chris twelves on our next episode.
Medical Breakthroughs.
The research journey is anunder the mask audio production.
Medical breakthroughs,the research journey.
Hello and welcome.
I'm your host, Caroline Burden,and you are about to join me on
a journey into the fascinatingworld of medical breakthroughs,
but not just any breakthroughs.
We are diving into thepersonal stories, the setbacks.
(00:30):
That you will not believe.
This moment behind the cutting edgeresearch happening right here at
Leeds Teaching Hospitals NHS Trust.
Coming up in this episode,
the AI algorithm is called FIND af.
It was developed by ourselvesat the University of Leeds.
We developed and checked 2million people in the uk.
(00:50):
We've now done that in 9 millionpeople in other countries.
So this algorithm could gomuch further internationally.
Globally and it can actuallybe something that we export.
Can an algorithm predict a stroke?
What started as an idea in Leeds couldchange the face of healthcare worldwide?
My name is Ramesh Raja.
I'm a academic clinical lecturer at theUniversity of Leeds and a cardiology
(01:13):
registrar at Leeds Teaching hospitals.
So atrial fibrillation is, uh,an abnormal heart rhythm that
comes from the top of the heart.
It's very common affects about 1.5million people in the uk, and it's
becoming more common so that thereare now more cases diagnosed of atrial
(01:34):
fibrillation each year than the four mostcommon causes of cancer put together.
Wow.
Yeah, so it's very common and it isimportant, so it causes symptoms like
breathlessness, the feeling of your heartracing, uh, chest discomfort, fatigue,
but also, and probably more importantly,it increases your risk of stroke by
(01:55):
five fold, increase your risk of heartfailure, and increase your risk of death.
All of these things, ofcourse, very important.
Um, and when you put that all together,you have a very common condition,
which is becoming more common and isimportant in terms of for the individual
and for the health system as a whole.
Um, and really.
(02:15):
What we in this study are looking atis one of the big issues with atrial
fibrillation, which is it causes allthese problems, but it still causes
increased risk of stroke, increased riskof heart failure, increased risk of death,
whether or not it causes you symptoms.
So it can sometimes be abit of a silent killer.
Um, you can have no symptoms but haveatrial fibrillation and be at risk.
(02:38):
Um, so it's, it's, it's estimatedthat between a quarter and a
third of people that have atrialfibrillation, uh, are undiagnosed.
And
how treatable is it, you say it increasesthe risk fivefold if you were to treat it.
How does, how does the risk look then?
(02:58):
Yes, there's lots of treatmentsfor atrial fibrillation.
Uh, the ones that are aimed atreducing your risk of stroke,
uh, are tablets, which areeither once a day or twice a day.
Very well tolerated.
A number of different available optionsthrough the NHS, um, uh, and with a.
(03:21):
We can reduce that risk of stroke bytwo thirds if we find people with this
abnormal heart rhythm and we treat them.
And so that is, uh, what is behind thenthis, this study using ai and you say
AI today, and people think about Ja, youknow, DPT and is it gonna steal my job?
And are we gonna haverobots in the future?
And all the rest of it.
(03:42):
Um, it brings acrossa, a lot of questions.
How, how are you using it in this study?
So the, the problem we find is thatpeople, uh, have this abnormal heart
rhythm, but are undiagnosed and weactually have in the health service
lots of different tools that we canuse to diagnose the problem types of,
uh, devices that will take an ECG, uh,and show you your heart rhythm and.
(04:09):
Atrial fibrillation.
The question is, who shouldwe give those devices to?
Uh, they cost money.
Uh, they inconvenience people byhaving to take their recordings.
So we want to find the rightpeople, and that's what we're
using AI for In this study.
We are, uh, using AI in datathat's routinely available in.
(04:30):
Each of our, uh, GP records and usingthat to identify people at risk.
And it's those people at risk thatwe are offering the opportunity to
have, uh, rhythm recording to lookfor this abnormal heart rhythm.
And so if, um, if the results of thisstudy were to mean that actually this
would become routinely used, is thissomething that would replace, uh, you
(04:54):
know, a, a, a GPS role within this?
So as you say, people are always.
Potential implications of ai,uh, um, things such as chat
in very
this simply to helphealthcare professionals.
(05:19):
And more efficiently.
Uh, because at the moment we have noavailable tool in the NHS to allow
us to identify these people at risk.
What we're doing is trying to providethat information to primary care
healthcare professionals so thatthey can use it to inform their care.
So not replacing, hopefullyaiding, uh, and improving care.
(05:40):
And so how, how does the study work?
How many, how many peopledo you need to be involved?
Yeah.
So first of all, we, the, the,the algorithm, the AI algorithm is
called Find af, and that's becauseit's future innovations in novel
detection of atrial fibrillation.
And actually it was developed byourselves at the University of Leeds.
So this doesn't comefrom a private company.
(06:00):
This was developed through funding forthe British Heart Foundation by myself.
Doctors, a team of statisticians,a team of data scientists, all done
with the aim of improving care.
Um, and when we made it, we madeit, uh, using information from
(06:21):
over 2 million people in theUnited Kingdom, uh, all anonymized.
Uh, but this was information thatis available in primary care,
but we could use to identify.
Just to be clear as well,the AI is not changing.
It's not some sort of multi-headed hydra.
It's a fixed thing.
We trained it.
(06:41):
It's what, it's, it's not gonnachange, it's not gonna adapt,
it's not gonna do other things.
It's built for this purpose.
Uh, and it was built by scientiststo try and make things better.
Um, and now what we're doing in thisstudy with funding the Be Foundation
and from charities importantly.
We're actually taking it from theresearch setting, which is we did it in
(07:02):
data and actually to make a difference.
So in this study, uh, we have putthe algorithm into primary care
records, into the records that sitin your GP practices, and we've
identified people at risk and we'veidentified 2000 people at risk across.
And we've invited those individuals,um, to take part in the study.
(07:26):
Now, of course, whatdoes the study involve?
Is it a lot of work?
It really isn't at all.
We've tried to make this study as easyas possible for people to take part in.
Um, so we've deliberately, uh,centered the study in areas that are,
uh, are full of poor communities.
They're full of communities with ethnicminorities and places where historically,
(07:46):
uh, heart disease has been a. Uh, whathappened to the study is the AI algorithm
runs through the records, identifies theat-risk people, and those individuals
are invited by a text and by a letter.
If they want to take part andthey send the letter back to us.
Um, we actually send them a littlehandheld device, which put your thumbs
(08:08):
on, and they take the recordingsas they go about their daily life.
Four times a. So they neveractually have to leave their
house to take part in the study.
It all happens very remotely to tryand make it as convenient as possible.
Of course, you know, you have to takeyour recording a few times a day, and it's
easy enough to forget that once or twice.
(08:28):
But this little handoutdevice is very light.
It can fit in your handbag.
Um, it should be something that youcan do as you go about your daily life.
And you don't have to cometo see us in hospital.
You don't have to go forlots of, uh, invasive tests.
It's all done simply through that.
And of course at the end of thisstudy, once people have undergone
this period of ECG monitoring, uh,we then provide them with the result.
(08:51):
We tell them whether we found, uh,atrial fibrillation or we didn't.
And if they, we do find atrialfibrillation, we make sure that
their primary care, um, healthcareprofessionals aware and that
they're put on the right treatments.
And what's the study?
So the goal is to, uh, one, improvedetection of this ab abnormal heart
(09:13):
rhythm because as I said, a thirdof people who have it aren't aware.
So we know there's a, an unmet needthere, and we're trying to address
it by actually getting peoplediagnosed and gets them on the right
treatments before it causes them harm.
And also within the study, uh, weknow that different individuals
are at different levels of risk.
(09:33):
Uh, and what we want to understand is.
Is there a point at which the risks, uh,is so high that, uh, this will be the
most effective approach to use in the NHS?
This is all funded by research charitiesat the moment, but we want to find out
what's the optimum risk level at which weshould provide people with these tools.
Uh, that will then be something that wecan actually translate and use in the
(09:57):
NHS that will be effective in terms of.
Get them the right treatment,but also cost for the nhs.
So we're for the nhs.
And if it, you know, did get the greenlight and it did get rolled out, what
sort of timescale are we looking at tosort of see it, uh, in use generically.
(10:17):
Yeah, so actually in parallel to thisstudy going on, we've actually been
working with bodies within the N chestto see how we can translate this.
So we've done, we've done anumber of important steps.
One is we did a lot of interviews with,with GPS and people in, in primary care
to understand one, is it useful to them?
Which we found resounding Yes.
Two, how would theylike it to be deployed?
(10:37):
How do they want us to use it in NHS?
As we use that information we've puttogether the sort of pathway that would
follow, and actually we're currentlyworking, uh, with, uh, providers of
these, uh, GP records in primary care toactually make the algorithm available.
So we're working with, uh.
(11:03):
I mean, it, it sounds reallyinteresting and like it could actually
have really quite a significantimpact, um, upon people's health.
Yeah, absolutely.
So I think it's really importantthat people are aware that we did
this research one 'cause we'rea team that include doctors who.
Actively looking to make a differenceto care right now, not in 10 years,
(11:25):
not in 20 years, but right now.
And thus, even though we were using ai, wemade a lot of efforts to make sure that it
was implementable right here, right now.
And also that was scalable.
So, uh.
98% of the UK populationare registered with a gp.
That's a lot of people, right?
And all of those people havea gp, uh, medical record.
(11:48):
And this algorithm can interact withevery single one of those records.
So what we're aiming for is to makesomething that's truly scalable across
the country to help everyone in the, and.
And I think it's, um, it's reallyinteresting that, as you say, this has
(12:11):
come from the NHS, it's come from supportfrom charities, and it's come from LEED
University, leeds teaching hospitals asopposed to from a, an organization, a
business trying to, um, you know, helppeople but also benefit financially.
And of course, uh, we, we knowthat, uh, uh, it's important that
(12:33):
we have companies trying to do that.
Yeah.
Private companies coming inwith their medications, with
their devices, et cetera.
That's important for our innovation andhow we drive healthcare going forward.
But very much it, this, uh, thiswhole program of work, this aim to
improve has come from within, withinthe N Charities, informed by done by.
(12:58):
Scientist for.
The pure reason of tryingto make things better.
So it comes with the bestintentions, and that's why I
think that's really important.
You know, you hear all the time nowwith AI about ethics boards, about
are they doing the right thing?
Are they, are they, uh, doingsomething that's, uh, for
profit rather than for benefit?
I think we can be very clear thatwe, we say with a very clear heart,
(13:20):
we're doing this for people's benefit.
Uh, and that's why, uh, we help
this.
And how long is this, um, piece of,you know, research work from start
to hopefully, you know, soon finish?
Yeah, I mean this, this researchwork has been years in the making.
(13:41):
Uh, so, um, the fundingapplication to this went back
in to develop the algorithm.
Find AF was back in 2018.
We started working on it in 2020.
This study, uh.
Uh, started, uh, over a year ago now.
(14:02):
Uh, so this has been over half a decadein development to get to this point.
And my role in this was actuallythe, uh, it was my PhD funded
by the British Heart Foundation.
Uh, that led to thedevelopment of the algorithm.
So I was taken outta my clinical workas a doctor for a few years to do this.
To work with ai to work with, uh,primary care records to try for.
(14:35):
Part of my role as half clinician,half academic has been then trying
to drive this forward with a reallystrong, uh, multidisciplinary team of
lots of really important healthcareprofessionals and scientists as a
team trying to bring this forward.
That must be personally really fulfilling.
Absolutely.
It's the best part of the job.
(14:55):
Being able to look after patientsin, uh, in my NHS job, but also do
stuff that I actually think can makea stepwise change, a paradigm shift
change, uh, and a change at scale, uh,is, is really rewarding and seeing, uh,
the fruits of years of hard research.
To maybe actually make a differenceto people is super rewarding.
(15:16):
And I, and I imagine you, you are seeingthat now with, with the trial patients who
actually they have been diagnosed with af
Absolutely.
That's been a, a really rewardingthing in that as part of the study,
which obviously research, we actuallyhave found AF for some people and made
sure they're on the right treatment.
And actually we get really goodfeedback from patients, one
who've taken part in the studyand in terms of diagnosed happier.
(15:44):
They know they're on the right treatment.
They know they're being protected.
So often, you know the pointthat you see your GP is the point
where there is something wrongas opposed to future planning.
And I suppose this makes a ahuge difference 'cause it's
preventing those hospital stays,the heart attacks, the strokes.
Absolutely.
I think we can all see the NHS as understrain, uh, in from us within who are
(16:09):
take, who are part of the system to alsofor all the patients who come and see us
on a daily basis are constantly tellingme how busy things are and how, and
how and how under strain the system is.
And so, uh, and a clear, a clear, um.
Shift focus things when
(16:33):
before.
We'll know somebody's had a stroke,and we'll know how debilitating that
can be for the individual, how muchthat can affect the family, how it can
affect everything, responsibilities,financial stuff, et cetera.
It's not just a, a single thing.
This affects your life.
So if we can stop that for an individual,that's a hugely rewarding thing for us as
(16:53):
clinicians, but also a hugely beneficialthing for the NHS, both in terms of,
uh, the provision of that care, but alsoin terms of the cost of that care and
for the cost for society as a whole.
And obviously you, you, you arehoping that this will get the rollout.
And let's say once this has the rollout,do you have any other projects that you
are already thinking about for the future?
(17:17):
So, um, heart disease in general.
Is still alongside cancer, the, thenumber one cause of, of death in the,
in the United Kingdom, including thenumber one cause of premature death,
uh, in, in the United Kingdom and thecause of, cause of hospitalization and
quality of life, uh, being impeded.
So one can look at what we've done herefor atrial fibrillation and with fine af.
(17:40):
As effectively the first in a seriesof a number of these, uh, tools and
projects that we're running, we're alsodoing something very similar in heart
failure because, for example, in heartfailure, that's again super common.
More than a million people diagnosedin the uk, but you wouldn't believe
that eight in 10 of those peopleare only diagnosed after they're so
sick they have to come to hospital.
(18:01):
So that's a clear unmet need that wewant to diagnose people earlier before
they're so sick they come to hospital.
So we're doing a very similar thing, whatwe did with Find EF with heart failure.
We've made find hf where we'reagain using primary care records at
scale, identifying at-risk people.
And we're shortly hopefully gonna bestarting a study where we're doing
a very similar, uh, study, but fortrying to protect heart failure.
(18:28):
Diseases which severelyaffect people's lives.
So hopefully in years to come, you'llsee this as just a starting point of how
we then shifted everything and improvedearly diagnosis in the community to
make patients better and live longer.
That's phenomenal.
I hadn't even thought about thefact that you could then take the,
the technology and then transposeit for, for different illnesses.
(18:48):
Absolutely.
So the, the, the team.
In terms of the humans that go intomaking this, this, these technologies,
the principles behind how we do it andhow we test it, how we make sure it's
safe to be used in, in people we can, wecan replicate in different disease areas.
That's what we're currently doing.
Uh, and we know there are many, manydiseases that are diagnosed late in, in
(19:10):
the disease course at the point at whichthey really affected people's lives.
And we wanna try and get ahead of that.
And we want to try andshift everything we know.
That's what the government wants as well.
They want things in the community.
They want digital, not analog.
They want prevention, not not treatment.
And we want to try and address thosethrough a series of these partnerships
between clinical work, research work,charities, funding us, the national
(19:34):
law, health research funding us,and then taking this into the n hs.
And presumably it, it, it hasthe power to go further than the
NHS and actually go worldwide.
Yeah.
So that's something.
So Arthur, first goal, I'm a NHS Doctor.
Arthur first goal was toimprove care in the NHS.
Uh, but.
We have over the last five yearsdeveloped a really strong collaborative
(19:58):
network across many countries actually.
And so actually the find AF algorithmthat I spoke about that we, um,
developed and checked in 2 millionpeople in the uk, we've now done that
in 9 million people in other countries.
So this algorithm.
Much further internationally,globally, and it can actually
(20:19):
be something that we export.
This is, this is UK made,NHS University made.
We can export this to improve careworldwide, but also to bring back revenue
for, for the NHS, for the universities.
So a, a huge impact almostof unimaginable scale.
So that's what we hope to achieve.
We have to get there first.
But we, you know, we, we, we, we aim,we aim big because we aim to try and
(20:42):
improve things across the country, thechest, and then we aim to, to, uh, take
that model and apply elsewhere as well.
And, and be able, and you know, I saidatrial fibrillation is very common.
For example, this isthis one disease area.
1.5 million in people in the uk it'sabout 63 million people globally.
So that's a huge number.
We're talking about huge numbers here,and so we really are looking to do
(21:04):
things at scale, make big benefits,and potentially make a big difference
on a national and international scale.
But of course we need to get that first.
It's been five years development sofar, uh, and we need to keep going.
Coming up on our next episode,the advance now is that we have
personalized cancer vaccines.
So people who have been diagnosed withcancer, a sample of their tumor is taken,
(21:27):
the genetic material can be extracted,and the particular mutations of that
individual's cancer can be identified.
And the vaccine developed that ispersonalized in that it's directed
at targeting those antigens that werepresent in that particular patient.
And you can hear more about thebreast cancer vaccine from Professor
(21:47):
Chris twelves on our next episode.
Medical Breakthroughs.
The research journey is anunder the mask audio production.
Popular Podcasts
On Purpose with Jay Shetty
I’m Jay Shetty host of On Purpose the worlds #1 Mental Health podcast and I’m so grateful you found us. I started this podcast 5 years ago to invite you into conversations and workshops that are designed to help make you happier, healthier and more healed. I believe that when you (yes you) feel seen, heard and understood you’re able to deal with relationship struggles, work challenges and life’s ups and downs with more ease and grace. I interview experts, celebrities, thought leaders and athletes so that we can grow our mindset, build better habits and uncover a side of them we’ve never seen before. New episodes every Monday and Friday. Your support means the world to me and I don’t take it for granted — click the follow button and leave a review to help us spread the love with On Purpose. I can’t wait for you to listen to your first or 500th episode!
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Special Summer Offer: Exclusively on Apple Podcasts, try our Dateline Premium subscription completely free for one month! With Dateline Premium, you get every episode ad-free plus exclusive bonus content.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.