AI-Driven Personalized Cardiac Care Ft Alfred Woo

Transcribe

[Marty]

Welcome to the Digital Health Transformers podcast, where we explore the ideas and innovations shaping the future of healthcare. I’m your host, Marty. And today we’re focusing on how artificial intelligence and personalized health technologies are transforming the way we understand, monitor, and manage cardiac care.

I’m joined today by Alfred Wu, the chief product officer at AliveCor, a company leading the transformation of cardiology through intelligent, patient-driven solutions. AliveCor’s mission is to save lives and transform cardiology by delivering highly personalized heart data to clinicians and patients anytime, anywhere. Their FDA-cleared CardiaMobile device is one of the world’s most clinically validated personal ECG solutions, setting new standards for accessibility, accuracy, and patient engagement.

Alfred, thank you so much for being here, and I’m so excited to talk to you.

[Alfred Woo]

Thanks, Marty. Happy to be here. Thrilled.

[Marty]

So, let’s start with your story. You’ve led product innovation in digital health for years. And I’m wondering, what drew you to AliveCor and its mission to personalize cardiac?

[Alfred Woo]

Well, it’s been a long story, but a little bit about me. I’ve been in technology here in the Bay Area for about two to three decades, primarily on the consumer technology space. And I think one thing you’ll learn in consumer technology, for a product anyway, is the user is the center of the universe.

Now, that’s great when you’re selling consumer products, but no one has ever said those words about healthcare. So, when I came to AliveCor about six years ago, it was an opportunity to sort of bring some of those principles from the consumer side into the healthcare space. My objective when I joined was, hey, we can empower the patient by utilizing some of the core principles from the consumer space.

That is prioritizing user experience, making things simple and satisfying, and bringing that directly to the consumer. That was kind of the whole motive of coming here. And I think when I arrived, it was the perfect time, because the potential of AI was becoming clear.

There was acceptance of biometric sensors and being biometrically monitored. So, the confluence of all of those trends made a perfect time to leverage AI sensors and data to help bring about this mission of making healthcare a lot more patient-focused versus clinician-focused or payer-focused.

[Marty]

Okay, got you. Yeah, AliveCor’s mission, it’s like to be a 24-7 virtual cardiologist for patients beyond the clinic. Okay, and then I’m wondering, how does that vision shape your approach here to product design and patient engagement?

[Alfred Woo]

Yeah, so if you try to break it down, there’s probably two primary problems you’re trying to solve for the patient. One is access or accessibility. You already mentioned that earlier today.

And one is intelligibility. What does it all mean? Typically, healthcare topics are pretty opaque to the patient.

So, when we develop products, we kind of follow those principles, and there’s probably four credos that we follow. One is ease of use. It must be fast, simple, easy to understand.

Simplicity always wins over complexity when you’re building something that is focused on the consumer or the user. You take a very digital-first approach, and this kind of speaks to that access problem. We maximize.

We use software, algorithms, and AI as much as possible because they’re always available. They don’t get tired. They can provide useful inputs 24-7, seven days a week, no matter where you are.

And then we back that up with clinical services so that when there’s an issue, when you require the extra level of care, we have that escalation path. Third, I think, is data is great, but actions are better than data. So, data can come from a lot of sources.

You can buy a cardio-mobile device. You can buy other wearables that give you data. But data alone is not useful.

We really want to focus on what does it mean and, probably more importantly, what to do with that information. So, that’s one use of our AI tools is to try to parse actions and suggest actions from the available data. I think the fourth piece of it is really trending toward continuous care.

So, healthcare should not be episodic. It should not be something you get only when you go see your doctor. It should be something that, with these tools and this technology, gives you a mechanism to continuously and passively manage your health all the time, even when you’re not in front of your clinicians.

So, those are the four things that we try to do with all our products and that keeps us focused on our core mission.

[Marty]

Yeah. And so, okay. So, yeah.

I mean, it is a big problem you’re solving. There’s a lot of patients that are experiencing gaps between their cardiology visits, too. And so, how is AliveCore closing that gap by bringing this real-time, personalized cardiac insights into the patient’s hands?

[Alfred Woo]

Yeah. So, the primary – the first problem is access, which you’re talking about. And just some statistics, 46% of counties in the U.S. don’t have a single cardiologist. If you go to rural areas, I think north of 80% are cardiologists-free. So, getting that cardiological help is hard. Even in populated areas and more urban areas, it can take many weeks and many months, right?

So, it’s very episodic and very hard to receive. So, with these tools, they can get assistance. A lot of times, with their data, with some AI, they either get escalation or they can get peace of mind that everything is fine.

So, that kind of solves the first line of support problem. The second thing is we can redefine what care is. If you think of care as the traditional sense of when you’re afraid of your doctor, yeah, that’s not going to happen all the time.

But if care could mean daily insights, it could be insight into your trends. It could be confirmation of either a problem or lack of a problem. It could be a nudge on what to do to improve your condition.

You define care that way, it can become a continuous thing, right? And that helps fill the gaps when you’re not in front of your doctor. So, I think in the end, if you want to sum it up, one of the goals that we strive for is we want to shift the focus of healthcare from the doctor’s role in treating sickness to the patient’s daily role in maintaining wellness.

So, a bit of a paradigm shift we’re looking at from the other side.

[Marty]

Yeah. Yeah. And so, okay, for wellness, I imagine predictive insights could be important.

[Alfred Woo]

Yeah.

[Marty]

And that’s where maybe AI can come in.

[Alfred Woo]

Yeah. Predictive and insights into the trends. A lot of times when you look at health data, it’s hard to understand what they mean.

But if I can sum up, oh, over the course of the week, you’re treating this way. That makes it much more intelligible and that makes it more valuable than just raw data.

[Marty]

And I was going to, yeah, I was going to kind of lead that thought into my next question, which is, so with AI and analytics playing such a central role in this process, how do you see artificial intelligence enhancing both the clinician’s workflow and the patient’s sense of control over their own heart health?

[Alfred Woo]

Yeah. And that’s a great question and that really goes to the core of our mission. So if you think of the resources, data and AI, at three kind of levels, there’s the raw data itself, the ECG reading, for instance, the blood pressure reading, whatever the data is, then there’s the insights you gain from it, a little understanding of what it all means.

And then there’s the actions that you take from it. So, of course, the Cardinal device records ECG data. We can tap into Apple Health and get data from other sensors.

We have direct integrations on our blood pressure sensor. So we have a wealth of data. And not only could you have like single points over time, we capture longitudinal data.

So that can be used with AI to help track trends, which is also a very powerful thing. The second piece, I think, is that is the insights, which is what does the data mean? And that could be AI that summarizes results for you, that tells you you’ve had AFib X times last week.

Your AFib incidents are getting more prolonged or getting better. And then certainly as we fuse the input from different sensors, from ECG, from blood pressure, in the future we get data from glucose monitors. We can provide a much more comprehensive and a richer set of insights.

Then the third piece is spurring action. Now that you know how you’re trending, what do you do? If it’s a concern, you should see a doctor for sure.

If you can’t get to your doctor, we can provide that clinical service. So when you look at, back to the original question, how does it benefit the patient or how it benefits the doctor? If you’re a doctor, getting that stream of data is very valuable.

I mean, people can take the ECG reading every day, a few times a day. You cannot do that by getting people to the clinic. The insights can because they can help them pick the signal from all the noise.

I think where the actions come in is that is what benefits the patient the most. Now that I know this stuff, what do I do with it? The key here, I think, is with the AI, all these can be provided 24-7, real-time, anywhere, anyplace.

Then, of course, it kind of fills that care gap between when folks are not in front of their doctors.

[Marty]

Yeah, awesome. Yeah, so I know that AliveCore is pioneering deep learning models in ECG interpretation, kind of to what we’re talking about using AI here. And so how do these tools that you guys are pioneering, how do these tools support clinicians in diagnosing and managing these complex cardiac issues?

[Alfred Woo]

Yeah, I think for the doctor, it’s got to be partly about efficiency. They can work faster, but it also, I think, improves their performance. It makes them better at their job because one thing AI is very good at is it’s very good at seeing patterns that the human eye can’t see.

So it can detect minute changes across thousands, even millions of readings, and detect the relevant pieces. Very hard for a human to do, even a trained human. And like I said, 24-7, seven days a week, never gets tired.

So as a doctor assistant, AI, I think, is wonderful. It could either give them an assessment that they can confirm, or in many cases, it points out things that they may not have seen at all because there may be subtle. So not only is it an efficiency booster, I think it gets better care for the patient because now their viewport is wider.

I think one thing about our clinical AI, particularly, is our CardioMobile 6L is the only FDA cleared personal six-lead ECG reader. So the more leads you have, the more data there is and more views you can get at the heart and the opportunity for great diagnostic value. So with this extra information and the AI that accompanies it, we can reduce the efficiency workload even more by giving a more comprehensive set of AI determinations that are faster and more accurate.

For the patient who probably can’t make as much sense from the ECG data, one thing that does do for them is it alerts them if there’s an issue. Because with heart disease, it’s often asymptomatic. It often has very indistinct symptoms.

Oh, I’m feeling tired. I don’t know what that means. But if we could text something and tell them, we have seen signs of this.

We can’t diagnose you because you’re AI, but you should definitely see your doctor. It kind of gets them off the couch and gets them through the consultation phase. And we’ve seen many cases where our users have said they have had indistinct symptoms for years.

They go to their doctor. They never catch it. They borrow a device.

They have recording. They have captured it. And now they’re getting treatment.

And I think that’s the most gratifying story that I hear. I hear that time and time again.

[Marty]

Yeah. And so, yeah, the trust aspect kind of stood out to me a lot in that response there. And so, yeah, the clinical adoption of AI is going to require trust.

And so with that in mind, how are you ensuring transparency, data security, and compliance as your technology evolves?

[Alfred Woo]

Yeah, that’s a very important topic, Marty. Especially for us because we are handling sensitive data, sensitive personal health information. Trust is paramount.

And so there’s kind of a couple levels of trust. One is security and privacy. And to conform, to be fully compliant, we are compliant with standards such as HIPAA, ISO 27001, SOC 2 DPR.

So that addresses kind of the privacy security part. Then I guess the second piece is the integrity of the AI. Like you all know, right?

Sometimes it hallucinates. It makes up stuff. Can you trust the AI?

So we’re taking it in a very piecewise fashion. I mean, we built our AI. We put guardrails around it.

We said, these are things that you can say. These are things that you can talk about. We put some pretty strict guardrails.

Over time, as AI improves, we’ll probably expand it. But since we have a network of cardiologists that work with us in all 50 states, any kind of high-impact decision that the AI may be leaning toward, we escalate to the clinician to make sure that we’re giving proper, appropriate, accurate information to the patient. So I think that’s how we’re structuring our AI usage.

Over time, we’ll expand it. But today, I think that gives us the best balance of performance, security, and patient benefit, and minimizing patient risk, which is paramount to us.

[Marty]

Yeah, yeah. I mean, I think it’s important because often I feel like AI and automation can sometimes seem impersonal. So knowing that there’s guardrails and knowing that there’s things in place to make sure that they can’t do bad things.

But, yeah, it can often seem very impersonal. And that leads me to another question, which is, how does AliveCorps make sure that this technology is going to strengthen the clinician-patient relationship between humans rather than replace it, keeping that more personal relationship?

[Alfred Woo]

Yeah, yeah. I think that’s also another great topic because I think, as you can imagine, people like that human connection with their doctor. They have a relationship with them.

They really don’t want just an AI doctor replacing it. And maybe that’s a convenient way to think of how AI would be used. But I think, realistically, in terms of putting these safety guardrails in place, in terms of what people want, the better way to think of how AI would be used is as a tool to drive individual empowerment.

So we give patients the agency to make conscious decisions, to act independently, and to take a more active role in their health care. It’s shifting the paradigm from being a passive passenger, doctor, tell me what’s wrong with me, treat me, to being an active partner in your heart health. And it’s very opaque, it’s difficult, but we’re giving the tools to take that role.

And without those tools, it’s not really possible for the layman to understand everything. But now, the balance of powers has shifted a little bit and they can be an active partner. I think that strengthens the relationship more than it breaks it down.

Rather than just, you’re the doctor, I’m the patient. Now I’m a partner in my health journey.

[Marty]

Yeah, I think that’s a great way of looking at it. And then what – so, okay, back to the clinician then. What feedback have you received from clinicians about how these tools are integrating with their workload, reducing their workload, integrating with their work and reducing their workload while improving the patient?

What feedback have you gotten from the clinicians using your stuff?

[Alfred Woo]

Well, I’ll start off by summing it up. The clinician enthusiasm for cardio is great. If you look at kind of our sales of cardio devices to consumers, these are over-the-counter devices.

You can buy them at Amazon, Best Buy. I mean, it’s a customer decision. A large percentage of these purchases are made because their doctor said, you should have this.

So, you know, doctors are on board and they think that it’s a benefit to the patient. So, you know, the enthusiasm already is very high. And what we hear time and time again is, you know, heart rhythms are largely unmonitored outside the clinic.

To get a heart rhythm measurement, yeah, there are more devices now. But traditionally, to get a clinical-grade one, you go into the clinic. By having this ability to monitor them continuously every day if they need to, getting an alert if you see something abnormal, I think that is a great benefit, especially because arrhythmias like atrial fibrillation, they’re not persistent typically.

If they are, you’re probably quite sick. But they come and go. So the ability for them to capture, detect, and then diagnose and treat without having the patient in there, I think it’s a benefit for everyone.

It’s a benefit for the doctor. It’s a benefit for the patient. And like I said, we hear so many stories about just that aspect of the product just making huge changes in the patient’s lives.

[Marty]

Yeah, awesome. And, yeah, yeah, cardiology really is kind of rushing towards – sorry?

[Alfred Woo]

Yeah, sorry, yeah.

[Marty]

Yeah, it seems like cardiology really is rushing towards this more virtual, continuous kind of care model that we’ve been describing in this podcast. And so then let’s talk about trends, the future. What do you see as the next significant step in taking this personalization even further in cardiac health?

[Alfred Woo]

Yep, yeah. So cardiac health is not isolated. It’s often co-morbid with other conditions.

I think something like three-quarters of our users have AFib, also hypertension. A third have diabetes. Sleep apnea is highly correlated with AFib as well.

So as there’s more sensor, more data that we can sort of fuse those inputs from not just ECG but other biometric signals as well, we can build a model that gives a more holistic view of the patient’s health and provide kind of more holistic recommendations on how they should manage or improve it. So I think that’s happening now. I think as we get more and more sensors out there, people accept the use of this monitoring.

There’s a huge opportunity to provide great personalization and, like you mentioned earlier, great predictive capabilities because we have so much more data available. I think the second thing is we talked about data, insights, and actions. It’s trending toward actions.

You know, tools that measure is kind of what we have today and what we had yesterday. I think what we’re trending toward is toward intelligence that acts, which is really kind of the end goal is don’t just tell me what my problem is. Tell me what I should do about it or, in some cases, do that for me even.

So it would actively protect you, coach you to improve your condition. And then, when needed, connect you to other health resources that provide the extended level of treatment that you need. So you can see all those together really help fulfill the mission of you’re not afraid of your doctor.

We’ve got you covered.

[Marty]

Awesome. Well, yeah, this was great. Alfred, thank you so much for joining us today.

I think AliveCore’s work demonstrates how intelligent, patient-focused tech can really extend cardiac care beyond the hospital and empower people, clinicians and patients, to act earlier and with greater confidence. So we really appreciate your insights on the future of heart health. And for anyone listening, to learn more about AliveCore and its solutions, visit AliveCore.com or Kardia.com.

That’s AliveCore.com and Kardia spelled with a K. Stay tuned for more conversations on how digital innovation is redefining healthcare delivery. And until next time, stay well.

[Alfred Woo]

Thanks, Marty. Enjoyed it. Thank you.