Why You Shouldn't Trust Your Smart Ring

0:00 Like the thirsty cast members of The Bachelorette,

0:03 Samsung Aura and, if rumors are correct,

0:06 [music] Apple all want to put a ring on it.

0:10 And why not?

0:11 The promise of a smart ring is undeniably sexy.

0:14 A screen-free titanium hoop that tracks your vitals 24/7.

0:18 Who wouldn't want it?

0:19 But, like a reality TV marriage,

0:22 buying a smart ring is sometimes followed by regret.

0:26 If you've ever wondered why your smart ring thinks you were napping

0:29 while you were working and yet misses your PR at the gym,

0:32 that's because, turns out,

0:33 shrinking an entire lab's worth of equipment down to the size

0:36 of a piece of jewelry requires some serious engineering compromises.

0:40 Compromises that could be fat-fingering your biometric data.

0:44 Let's break these devices down, starting with the main sensor.

0:47 [music] PPG, or photoplethysmography, is a fancy name for shining light

0:52 into your skin and measuring how much bounces back.

0:55 In a hospital, they'll clip a sensor onto your fingertip,

0:58 yield transmissive PPG.

1:00 It's incredibly accurate because the signal is unobstructed.

1:03 Smartwatches, on the other hand, use reflective PPG,

1:06 where the sensor and the photodiode are on the same side.

1:09 Fun fact, by the way,

1:10 they use green light because your blood absorbs it like a sponge,

1:14 providing a high-contrast signal.

1:16 Depending on, you know, how far apart your photodiode is,

1:20 if it's, say, on the curvature of a ring,

1:22 you may actually kind of be between these modalities,

1:24 where you're somewhat like in a transflective approach.

1:28 You may see behavior of both like a reflective and a transmissive PPG.

1:31 Smart rings can occupy a weird middle ground.

1:34 Because the sensors wrap around your finger,

1:37 light travels further through your skin than it would from a smart watch.

1:40 This gives the ring a slight signal advantage,

1:43 but it's one that unfortunately gets dragged down

1:46 by the constraints that arise from its size [music] and from its battery.

1:51 The amount of light you need in order

1:53 to measure an appreciable pulse is proportional

1:56 also to the active area of the photodiode that you're using to sense it with.

1:59 So, a bigger photodiode, a bigger sensor, means you can use less light,

2:03 less current, in order to perform the same relative measurement.

2:07 If you only just shrink the photodiode itself,

2:09 you're going to be decreasing your your signal-to-noise

2:12 ratio and have to, you know,

2:14 boost your LED light in order to accommodate for that.

2:17 You can think of it kind of like listening to a conversation in a crowded bar.

2:20 A smartwatch stands next to a person with a large, directional microphone.

2:24 A smart ring, on the other hand, is trying to hear them from across the room

2:27 using a tiny little spy mic taped to their collar,

2:31 which brings us to the battery.

2:32 The Samsung Galaxy Watch Ultra packs 590 mA hours.

2:38 [music] The Galaxy Ring, 17 to 22 and 1/2.

2:42 And yet, the ring claims it lasts for up to 7 days.

2:46 How?

2:47 Well, because it doesn't.

2:49 Both devices use duty cycling to extend their usable charge time,

2:53 but because the watch has a larger battery,

2:55 one intense session isn't going to drain it.

2:57 The ring, on the other hand, has about the battery of a Cheerio.

3:00 [music] They have also large dormant periods,

3:04 where the ring simply is not measuring at all.

3:07 And and so you can kind of think

3:08 of it as like interlacing on top of interlacing,

3:10 you know, in order to [music] make it so maybe you only measure the pulsations,

3:13 measure the heart rate, pulse ox, you know, for a minute, you know,

3:16 every 10, 15, 20 minutes, for example, in order to to keep conserving uh power.

3:21 [music] To put this in perspective,

3:22 when the Franklin Research Lab built their own DIY smart ring,

3:26 the battery gave out after about 4 to 6 hours of continuous use.

3:30 So, if we take that as a baseline for true continuous tracking,

3:34 to survive a full 168-hour week of nonstop data collection,

3:38 you'd need to wear 34 rings and swap them out like ammunition magazines.

3:43 By Wednesday, you'd look more like the Divination

3:46 instructor at Hogwarts than like a fitness enthusiast.

3:50 [music] The reason that smart ring bands can claim a week while Dr.

3:52 Franklin gets only 5 hours [music] is the sampling rate.

3:56 When a researcher says continuous, they usually mean high-frequency sampling.

4:01 When a consumer wearable says continuous, they often mean,

4:05 "Ah, we check in every now and then,

4:07 and then we interpolate a bit so the graph looks like

4:09 a solid line." And if you have a darker skin tone,

4:12 the LEDs actually have to work harder to get a readable signal,

4:15 which can cut that already slim battery life nearly in half.

4:18 [music] Now, that isn't to say that the form factor has no benefits.

4:22 There is a reason that hospitals clip sensors onto your fingers.

4:25 They are densely packed with arteries that are

4:28 right close to the surface of the skin.

4:30 But for smart rings, that's kind of a double-edged sword.

4:32 Besides being so darn cute and, [music] you know,

4:35 fingery, fingers are your body's temperature control meat sticks.

4:39 When you have vasoconstriction or dilation uh due to being

4:43 cold or due to, say, being in a sauna,

4:45 that will, again, change the distribution of blood

4:48 within the skin and impact your photoplethysmography signal.

4:51 Typically, vasoconstriction or having that blood

4:56 redistributed away from the surface

4:57 of the skin can have a detrimental effect on the PPG.

5:00 So, it'll decrease your pulses, it will decrease your signal-to-noise ratio,

5:04 and then the opposite is true when you have perfusion.

5:07 Cold hands equals a weaker signal,

5:10 which equals your ring struggling to find your pulse.

5:13 Lock outside in the winter or sit next to an aggressive air conditioner,

5:16 and your ring huh might think you're a spooky vampire with no pulse.

5:21 Or Sam Altman.

5:22 Unlike the finger, the wrist sits closer to the body's core,

5:26 [music] staying warmer and maintaining more consistent blood flow,

5:28 giving the watch a natural edge in colder environments.

5:31 But hey, I mean, if sitting still in a freezing room isn't your only hobby,

5:35 what about if you actually exercise?

5:37 Even a simple jog is going to slosh

5:39 your blood around enough to give the sensors a headache.

5:42 But for hand heavy lifting, the ring becomes basically decorative.

5:46 See, the problem isn't just motion, it's grip.

5:48 Odoo, your business doesn't need different apps for signing agreements,

5:53 running a POS, doing inventory management, or running email marketing.

5:57 All you need is Odoo.

5:59 It's a business management software with just about anything

6:02 you could imagine needing integrated into a single platform.

6:05 It's customizable and built to be easy to navigate since

6:08 you pick and choose only what you need, no extra fluff.

6:12 And if you only need one app, it's free.

6:15 Use our link for a free 15-day trial with no credit card required.

6:20 When someone tightly grips a steering wheel or a barbell,

6:23 we typically call this motion artifacts.

6:25 At the uppermost level of your skin, you have a capillary bed,

6:28 and when you apply pressure to this, that's

6:29 the first layer that actually gets compressed.

6:31 And so you end up with a a large spike in your photoplethysmography signal.

6:34 That's not attributed to your pulsations or your heart rate,

6:37 which is the thing we want to measure.

6:38 These motion artifacts and pressure-induced

6:40 [music] signal loss hit simultaneously, and they compound fast.

6:43 White-knuckling a barbell squeezes blood out of the very

6:46 capillaries the ring is trying to read.

6:48 So, by the time your heart rate has climbed to 160 beats per minute mid-set,

6:52 the ring might be reporting a breezy 80.

6:54 A quick TechQuickie safety tip, by the way,

6:56 wearing a metal ring during heavy lifts is not a good idea.

6:59 Beyond scratching the finish of your ring,

7:01 you're actually risking a degloving accident.

7:03 Seriously though, do not Google that.

7:04 The standard way to deal with motion

7:06 artifacts is to cross-reference the accelerometer data.

7:09 If motion exceeds a specific threshold,

7:11 then the PPG data just gets discarded entirely

7:14 because a gap is better than a wrong number.

7:17 A 2024 study even found that you might need to discard up

7:20 to 70% of heart rate variability readings due to poor signal quality.

7:25 To combat this data trashing, researchers are working on transformer-based AI

7:29 models to reconstruct corrupted PPG signals.

7:32 So, the sensor's tiny, the battery prevents continuous tracking,

7:36 cold hands kill the signal, gripping blinds it,

7:38 and most of these rings aren't even FDA-regulated medical devices,

7:42 making them pretty much high-end accessories with an asterisk.

7:46 So, are they an overpriced piece of e-waste?

7:49 Actually, no.

7:50 There is one time of day where almost all of these problems disappear, [music]

7:54 and that's during sleep.

7:55 Unless you're David Goggins,

7:57 you're probably aren't gripping a barbell in your sleep,

7:59 so your hands are relaxed and your arteries are right at the surface.

8:02 Smart rings can get incredibly accurate resting data while being

8:06 far more comfortable to sleep in than this darn thing.

8:10 Aura even boasts about their .999 correlation with medical-grade

8:15 ECGs for resting heart rate, which checks out.

8:17 Even with a few misbeats, basic average is easy.

8:20 Their .984 HRV correlation claim, however, has a massive catch.

8:24 Heart rate variability measures

8:25 millisecond differences between individual beats,

8:28 and it needs [music] beat-to-beat continuity.

8:30 So, a gap in the data doesn't just reduce accuracy,

8:33 it invalidates the entire measurement window,

8:35 which is why smart rings filter out low-confidence data before reporting.

8:39 To legally be able to claim this HRV accuracy,

8:42 many validation studies use sedentary participants.

8:45 In one, no subject's heart rate broke 90 beats [music] per minute,

8:49 so it's kind of easy to be accurate when your user's just vibing on the couch.

8:53 As for the high-level scores, like sleep score, readiness,

8:56 and stress, those are proprietary algorithms that arguably don't mean much.

9:01 It's kind of a way of gamifying your sleep

9:03 to see who [music] can have the most readiness,

9:06 I suppose, and to help you justify your $5.99

9:08 monthly subscription for the rest of your life.

9:10 If you want to see the testing facilities that smart wearables, like rings,

9:13 are designed in, check out this video where I visited one in China.

9:16 It's actually, for all of the dunking on these devices that we just did,

9:20 a lot more elaborate and way cooler than you might think.