Dr. Ron Falk invites Dr. Anil Gehi, the Sewell Family-McAllister Distinguished Professor of Medicine at UNC to provide insight into how some wearable health products work. Dr. Gehi serves as the Director of the Clinical Cardiac Electrophysiology Service and Program Director of the Clinical Cardiac Electrophysiology Fellowship.
Their conversation explores the evolving role of wearable health technology, such as Apple Watches, in monitoring cardiac health, advancing patient care, and transforming the landscape of clinical cardiology.
00:00:00:00 – 00:00:28:09
FALK: Hello and welcome to season two of the Chair’s Corner from the Department of Medicine at the University of North Carolina, Chapel Hill. I am Doctor Ron Falk, chair of the department and the host of this podcast, where we’re dedicated to empowering patients like you with knowledge about your condition, enhancing your quality of life every step of the way.
00:00:28:11-00:01:22:19
FALK: Each and every day we’ve seen an explosion in devices. Anything from, you know, wristbands, things like Fitbits, smartwatches, the Apple Watch and Garmin watch. There’s, other devices that are wrist- worn. There’s rings that were a ring you might have heard of. There’s even devices that are embedded in clothing that can detect various biometric data. So let’s divide these then, just for simplicity, for now, things that one could wear on one’s wrist. Things that one could wear on one’s finger, and then a plethora of a whole bunch of other intriguing, technological things that I probably don’t want to buy but are out there for me to think about. So what do these devices actually measure?
00:01:22:20 – 00:02:49:18
DR. ANIL GEHI: Yeah. Great question. So I think you have to think about it from the point of view of what sensors are available. And then how can we embed them in a device. So I think some of the most basic sensors that are out there are, for example, an accelerometer, essentially it’s, crystal that’s in the device that has a change in electrical charge when it’s moved. And that can be then translated into movement. And we can measure things like, step counts and, and physical activity. If you combine that with the GPS, that’s in a lot of devices, you can even get more information. Things like track your distance, you can track altitude changes, you can track velocity, this early. So this can be very sophisticated.
And you could use it, to essentially measure activity of all sorts. Those are in all sorts of devices. They’ve been around in in watches. They’ve been in bands like Fitbits, for many, many years. I think another important sensor that’s in a lot of devices is this technology that we get a lot of use out of called photocephlography.
00:02:49:58
FALK: Oh, say that twice.
00:02:50:18 – 00:04:57:21
GEHI: Yeah. Right. We say we call it PPG for short, okay. And basically it’s the same idea as what’s in like, a pulse ox. Like you probably, or most people in your audience are aware of devices that go on the finger and measure your oxygen levels. Essentially what’s happening there is there is a device that’s shining a light.Usually it’s an LED light onto the skin, and it detects the pulsations in the skin. And it can take that pulsation information and use that to basically calculate heart rate. So it just measures the rapidity of the pulsations in the skin and can use that to calculate a heart rate. And that is used in all sorts of devices.If you think about, the wrist worn devices like an Apple Watch or Garmin watch, essentially you might notice, if you were to look at the underside, that it will shine a light onto your skin periodically and that light gets then, reflected back, to the watch, and it can sense the pulsations in the skin and use that to detect heart rate.So that can be very useful for measuring things like, slow heart rates, fast heart rates. We can take that and look at patterns of heart rates and detect various rhythm problems. And then you could use that heart rate information and also look at things like heart rate recovery, which is an important parameter for conditioning of a person, as they’re going through like a fitness program or a conditioning, program, we can look at something called heart rate variability to look at kind of changes in heart rate that are beat to beat.So the PPG essentially can be translated to many, many things. Another important sensor that’s embedded in some of these devices is, ECG. Okay. So rather than just look at the pulsation and calculate heart electrocardiogram. So rather than just look at the rate, you can actually look at the electrical activity.So essentially to do an ECG you need to have two electrodes touching the skin. And measure the electrical activity between those two electrodes. And go through some filtering, processing, to be able to basically produce an ECG, something like you might get in the, in the office, in multiple leads, we can do, with a single lead, and with particular devices.
00:04:57:23 – 00:05:32:00
FALK:Let me just go back for a minute for the audience. Your heart is a muscle. With four chambers and those chambers pump in an incredibly well organized fashion because of the electrical impulses that are produced by, this system, this electrical system that you are now describing, being able to be measured in the office with an EKG machine.But now, rather than all those leads just to electrodes on the skin.
00:05:33:23
GEHI: That’s right. So if we take two electrodes, and apply them to the skin, you generally need to be on opposite limbs or somewhat spread far apart such that the heart is in between. So they can’t just be two electrodes like on one limb, for example
00:05:49:00
FALK: On one watch.
00:05:49:20 – 00:06:26
GEHI: Right. It could be, it could be. So if you think about how an Apple Watch does an ECG, essentially one electrode is on the back of the watch touching the skin. And then to do an ECG, you have to actively put your finger on the dial. And when you put your finger on the dial, that’s essentially an electrode on your opposite arm.With the electrode on the, on the back of the watch. And so now you have two electrodes on opposite arms, and you can measure electrical activity between your arms, your heart is between your arms. And so you can therefore get a one lead ECG.
00:06:26:00- 00:06:51:00
FALK: And this is really an important concept for everybody listening to understand if you don’t use on an Apple Watch that part of the device that actually forces you to measure heart rate by putting your finger from the opposite hand, your opposite arm on that button, you’re not going to get an accurate reading.
00:06:51:07 – 00:08:05:14
GEHI: Exactly. So the way the watch and most watches will look at rhythm problems continuously is by using that first technology. I talked about PPG where what it’s doing is it’s looking at the pulsations in the skin and interpreting what the rhythm rate is, or sorry, what their rate is or the rhythm is by the regularity and the speed of those pulsations. It’s not actually looking at the electrical activity itself. Now, to measure the electrical activity itself, you need to have two electrodes and it can only be done kind of, actively. So point of care, you have to put your finger on it and then you get it. So you’re not getting continuous ECG with a watch. There are devices out there where you can get continuous ECG, but those are things like a chest strap.That’s you’re, that you have under your skin where there’s two electrodes on the chest, all the time and you can get an ECG that way. And so that for that reason PPG is not going to be as accurate, for, determining the actual rhythm. It’s going to be able to tell you rate and you can get some interpretation of rhythm by looking at the regularity of it. But it’s not going to be as accurate as say, an ECG
08:10:00- 00:08:22:54
FALK: And there are things that humans do to themselves or diseases do to humans that would alter the accuracy, I suppose, of these, of these measures.
00:08:23:00, – 00:09:07:00
GEHI: Right. So, PPG for example, again, one of the probably the most common sensor that’s out there, is quite good when you’re at rest.But as you become more active and your heart rate is going through more fluctuations or say there are some changes to your skin, maybe you’re sweaty, maybe your device is not contacting your skin as well. Maybe your skin is dry. I mean, those sorts of conditions are going to make the accuracy of the sensor less. Especially some people have irregularities of their heart rate that make it difficult for the device to detect heart rate.So there are a variety of conditions that can make PPG less accurate.
00:09:08:00-00:09:15:46
BOTH TALKING: And darker. Skin tones are one of those factors that can also alter. That’s right. Obesity. There are several that. That’s right.
00:09:16:00-00:09:58:15
GEHI:Temperature of your skin, darker skin tones. You know, emotion, and therefore not having good contact with the device, those are all going to make it less accurate.
And for that reason, many times, when you’re exercising, your heart rate may not be as accurately detected.
00:09:33:00-00:09:58:17
FALK: Let’s talk then. In addition to the ability to measure rate, to measure, an EKG equivalent, if you put your finger on the, on the dial or you wear a chest strap, another real measure that you folks think about is how fast the heart recovers, after exercise is stopped. Can you chat about with that?
00:09:59 – 00:10:49
GEHI: Yeah. So, so one measure of a person’s fitness is heart rate recovery. So there are a number of studies, dating back to when we were doing a lot of treadmill testing on patients. To look at recovery after exercise and demonstrated that a person who is less fit, less conditioned is going to have a slower heart rate recovery.So you can look, for example, at the time it takes for the heart rate to drop by to say, 20 points. And if it drops by, say 20 points in a minute, that’s very good heart rate recovery. If it takes longer, to drop, that’s poor heart rate recovery. And that can be an indication of a person’s physical fitness.And it’s one of the things that will measure if somebody, say, going through a cardiac rehab program.
00:10:50:00 – 00:11:18:00
FALK: Okay, let’s be a little bit more specific here, because I can feel our listeners are going to start freaking out here. So just let me just go through this more slowly in somebody who is a conditioned athlete, who’s running up and down the basketball court constantly, their heart rate from a peak during exercise and now at full rest should recover. How much over what period of time?
00:11:18:10 – 00:11:41:23
GEHI: Yeah. So it’s hard to give you a hard and fast number but give me a this. So an estimate. So if somebody is in very good physical condition their heart rate at peak to rest would recover by about 15 to 20 beats per minute, within a minute. And these are the kinds of things that somebody who’s really into fitness and say, a competitive athlete is going to monitor a little bit more.
00:11:42:01 – 00:11:53:01
FALK: Let’s talk now about one of your favorite subjects, and that is, heart rhythm. Because these devices can help us with that.
00:11:53:20- 00:12:54:08
GEHI: Yeah. And this is actually one of the major use cases for these devices is, again, using the PPG, you can monitor heart rate and you can look at changes in heart rate and changes in the regularity of the heart rate.So for example, if somebody’s heart rate is, say, going along at 60 beats a minute and then all of a sudden their heart rate jumps to 150 beats per minute. Something probably abnormal is going on. Boom boom boom boom boom boom boom. Exactly. And so, since, if you are wearing a device like a watch or a ring that can detect heart rate using a PPG type sensor, it can alert you that your heart rate has suddenly changed. And they call that attack arrhythmia. You can, you can if you have a watch like one of these devices, you can go and look and you can see, oh, look, there was a sudden change in heart rate. It’s taki arrhythmia that might be of concern.
00:12:55:00- 00:12:57:00
FALK: It went from a steady beat to an accelerated at a faster rate.
00:12:57:09 – 00:13:54:00
GEHI: Right. And it’s and it’s important to think about how quickly that change happened. So for example, if somebody exercises, it’s normal for your heart rate to gradually increase. But if it suddenly changes to 60 now it’s 150. That’s probably something abnormal. It’s not a person that’s just simply exercising or walking up the stairs. So and, and very similarly you can look at slow rhythms.
We call that Brady arrhythmias. So if somebody’s heart rate is normally 70, all of a sudden it’s now 30. That might be something abnormal. And a watch or some other device could alert you to that. There are other heart rhythm disturbances that a device can alert you to, atrial fibrillation being one of the most common. So atrial fibrillation is probably the most common heart rhythm disorder.
And its hallmark is that the heart rate is irregular. It has an irregular pattern.
00:13:54:25-00:13:56:00
FALK: So Anil, sound it out for us.
00:13:57:00-00:14:52:10
GEHI: Yes. Okay. So, normal rate is going to be bum bum bum bum. It might be faster, it might be slower, but it’s going to be steady. Bum bum bum. AFib might be bump bump bump bump bump bump bump bump bump bump.It’s irregular. And that irregularity is a hallmark of atrial fibrillation. Now it’s not only atrial fibrillation. There are other conditions that can cause an irregular heart rhythm. But AFib is the most common. And a device with which has a PPG sensor can therefore detect atrial fibrillation. And what we can talk about maybe a little bit later, about one of the ways we use these devices is that AFib is a condition that sometimes has no symptoms, and having a device that can detect a condition that has no symptoms may be something very useful.
00:14:53:00 – 00:15:09:00
FALK: Last but not least, all of these devices claim that they can tell you how you are sleeping. Whether you have sleep apnea, whether you’re restless, whether you’re having deep sleep, whether you’re having REM sleep, help!
00:15:09:34-00:16:21:14
GEHI: Yeah. Again, very, very controversial and hard to know how good these are because, again, these are direct to consumer devices. They, for the most part, are not studied in a medical type scenario.Of course, the gold standard for measuring sleep quality and things like sleep apnea is an in lab test with things like brainwave assessment, eye movement, muscle activity, ECG, pulse oximeter. You have to have all of that together to get a gold standard assessment of sleep. But these devices can estimate they look at things like heart rate, they look at movement.They can look at other indicators of sleep apnea. So for example, they can assess if they had information on pulse oximeter. They can look at changes in pulse oximetry during sleep. And they can come up with an perhaps a rudimentary assessment of sleep quality. They may claim much more, but until we have studies that compare these to the gold standard, it’s hard to really know what to make of it.
00:16:22:16 – 00:16:51:45
FALK: There’s a lot of other, perhaps, uses that may or may not have as much research behind it. So, for example, there’s a lot of buzz online about rings that can be used for women that are supposedly contract fertility, sleep data. They promise all sorts of potential, feedback information. What should we think about that whole group of possibilities?
00:16:52:00 – 00:17:35:01
GEHI: Yeah. It’s hard. It’s hard to know what to make of this. Remember, again, these are direct to consumer devices. They take the basic sensors. I’ve spoke of heart rate, maybe temperature, maybe movement, take that volume of information and try to interpret all sorts of stuff. They have their own proprietary algorithm that might say, try to make a determination on fertility, or they might make a determination on sleep quality, or they might make a determination on your stress level.
We don’t know what’s in the black box. We don’t really know what they’re doing. These have not been studied in any rigorous way, so it’s really hard to know what to make of it.
00:17:36:03 – 00:17:57:14
FALK: Anil, I’m walking down the street. I’ve been exercising. And, you know, I’m a little short of breath, so I’ve just paused for a minute, but I feel really good, and I. I start walking fast again, and all of a sudden my watch goes crazy. An alert of some kind. What should I do?
00:17:57:38 – 00:18:24:21
GEHI: Yeah. Great question. So, there are a variety of different kinds of alerts you can get.
But I think probably the things that people might sense most often is tachyarrhythmia alerts, meaning, the watch detected a fast heart rate.
00:18:14:00-00:18:15:50
FALK: What should I do about it, though?
00:18:16:00 – 00:19:24:30
GEHI: Yeah. So it sort of depends on how a person feels. If a person feels okay. Not in an emergent situation. So an emergent situation might be significant.
Chest pain, significant shortness of breath or feeling (both talking) If you feel fine, I think that’s something to take note of and alert your doctor. Because Tachyarrhythmias, fast heart rhythms can happen at any time, and they are something that sometimes have consequences. For example, a very common tachyarrhythmia is atrial fibrillation. Atrial fibrillation being, an abnormal heart rhythm where the heart is beating in an irregular fashion and it can have consequences like, putting somebody at risk for developing blood clots in the heart and strokes, it can have consequences of causing shortness of breath or lightheaded spells or chest pain.
And these are things that may not need to be addressed in an emergency, type scenario, but should certainly be, be followed up on with your doctor.
00:19:25:00-00:19:53:58
FALK: There are times when these devices are life saving. And let’s talk about some of those. If you have an elderly parent that’s living alone or even an elderly human who’s walking along and they fall, these devices actually have algorithms that can help talk to us about that.
00:19:54:20 – 00:23:38:40
GEHI: Yeah, I think and I think this is, some of the most fascinating things about these devices is how these are going to come into play with our kind of day to day, assessments of health and assessments of emergency situations. So you’ve probably seen how some of these devices have fall detector algorithms built in.And essentially they have, you know, a gyroscope, of some sort in the device that can measure changes in position, they can measure changes in movement. And if you feed all that information into an algorithm, eventually you can develop a fall detector. And these are actually quite good. In fact, there was a day that my mother fell at home and I was alerted by her watch, and I was able to call her and see that she was okay.And it was pretty impressive that these things work. And I think those are some interesting use cases that these devices are really going to revolutionize. And there can be many others that you can imagine. One thing that I’m really looking forward to is how this can revolutionize the determination of events that can be associated with the sudden death.Think about a patient who say, is in their home and has a cardiac arrest. So their heart suddenly goes into an abnormal rhythm. They pass out, and they would die. If somebody doesn’t see this, start doing CPR, call 911, and hopefully the paramedics get to them quickly. Many sudden cardiac arrest events happen in the home when somebody might be alone or a person with them is not able to do CPR.Maybe they haven’t been trained or whatnot. These devices could potentially revolutionize that because remember, if you have a device on your wrist monitoring your heart rate all the time. And of course, these devices have the capability of calling 911, perhaps that can detect an event and be able to interpret whether this is a life threatening condition and bypass all the steps that are necessary to activate the paramedics.That could potentially really revolutionize care. Of course, there’s a lot of things, a lot of steps that have to happen before that can be, you know, something that is in the real world. There’s a lot of organizations of systems that need to come into play for that to happen. Of course, we need to make sure that this is an accurate type assessment, but we can imagine ways that that can happen.In fact, think about other sensors that are in, in our, in our environment. Think about your smart speaker, your Amazon Alexa, in your kitchen or in your bedroom that can listen to noises in your environment. And it can actually there are studies showing that your Amazon Alexa smart speaker can hear agonal breathing. So agonal breathing is that sort of kind of labored, gasping breathing that happens when somebody is in a severe, emergency type situation. They might be having a cardiac arrest. They might have an abnormal breathing pattern. These devices can actually hear that. And you can imagine that you can take that incorporated into some other, say, a watch on their wrist, put all these things together. Machine learning algorithms could put all that stuff together. And say something bad is happening.I better alert somebody.
00:23:39:54-00:23:59:49
FALK: Although I worry that Alexa is listening to all sorts of things that I don’t want a lecture to listen to. Doctor Gehi thank you. So very, very much for sharing your incredible expertise on all of these wearable tech devices and how you have really illuminated this area for us incredibly well. Thanks so much.
00:24:00:00-00:24:01:13
GEHI: Well, thanks for having me.