The bracelet Putin saw: Healbe Gobe is still the only gadget that counts incoming calories

I hope you will excuse us for this “Kolesnikovsky” approach. I wanted to talk about Healbe a very long time for many reasons. This will not be a review in the classical sense of the word, but perhaps you will be curious.



At one time, we were the first to have the first version of the device on hand. It was in 2014, we joyfully reported this both at home and at your place . The first version was extremely imperfect in the first place as “hardware”: the bracelets literally crumbled in the hands, user expectations were not met, the company received a bunch of negative karma, but did not stop. A few years ago, the second version of the bracelet appeared : it received a slightly more interesting design and good assembly.

I must say that the second version turned out to be, which is logical, more successful than the first. Now, if you go upstairs, you will probably find dozens, if not hundreds, of reviews, some of which are in the Western press, such as the Daily Mail. Healbe is still an international company. And, of course, this is one of the few devices with domestic roots, which reached the top officials of the state and the first broadcast channels.

It seems that last year the foreign leader in Yandex.News was a demonstration of the bracelet to the president:



After which the team visited the “Evening Urgant”.



As far as I remember, Vladimir Vladimirovich twisted the device in his hands, made a couple of ironic remarks, and that was it. Nevertheless, thanks to this, the picture was dragged along the top press and interest in Healbe seemed to have revived.

In this connection, we probably need to share our observations with us.

What can he do?

In general, strictly speaking, a lot of characteristics are stated in the bracelet, for example, he must determine:

  • Incoming calories automatically
  • Pulse
  • Pressure
  • Sleep quality
  • Hydration
  • Stress level

Let's start simple. What he definitely doesn’t do : the bracelet does not measure pressure. This item can simply be thrown out, and it’s rather strange that it has not yet been sawn out of software. The bottom line is that the algorithm tries to assume pressure based on the pulse, which is very primitive and unreliable mathematics. This is definitely not suitable for hyper hypotension. Thus, we have:

  • Incoming calories automatically
  • Pulse
  • Pressure
  • Sleep quality
  • Hydration
  • Stress level

Pulse



In principle, there were no problems. In our humble opinion, this is not a sports heart rate monitor, a bit slow, but which may well be useful for monitoring heart rate at rest. Compared with an ECG flash drive , and if you give Healbe a little time, then the data is relevant.

Sleep



There are no problems, everything is decent. There is a built-in smart alarm clock where you can set your preferred range from 5 to 60 minutes.

Stress Level



A fairly simple metric based on heart rate data. We have not been closely monitored.

Hydration



This is a curious phenomenon for a bracelet. In general, when we did densitometry, the doctor said that with the help of bioimpedance it is quite difficult to determine the exact composition of the water, but it seems like everyone is using it in the form of weights now, why not. The bracelet authors say the following:

Independent validation studies conducted at the University of California, Davis (USA) and the Red Cross Hospital (Guangzhou, China), confirmed that GoBe2 accurately monitors dehydration and rehydration of the human body.

Since, honestly, this parameter was not directly very interesting, we did not dig deep here either. In our assessment, we settled on the only unique and really interesting characteristic - the ability to calculate calories consumed in automatic mode .

What thoughts on this subject

Before we take up any comments, we read what other people write. I got a lot of enthusiastic reviews about Healbe, which confirmed a nearly 100% hit when counting. How this was confirmed is not obvious. Because there are a couple of small problems:

  • We do not know how and under what instrument the device was calibrated from this point of view.
  • We do not know what objectively this could be compared from improvised methods.

The thing is this. Even if we drive data from labels with our hands, it will still have a certain margin of error. And there is not a single similar device in the world at the moment: that is, we literally cannot even compare the gadget with the gadget, not to mention some procedures more accessible to the wide user.

The manufacturer, referring to the same sources as above, notes:

The accuracy of calorie intake (calorie intake) is 89%.

By the way, this figure has increased since the appearance of the first bracelet. As far as I remember, the error was defined as 15-20%. Which in general is also quite satisfactory for a home household gadget.

We decided to assume how he calculates.

Generally, to be honest, the degree of distrust reached insanity. One of the hypotheses was that the accelerometer works stupidly in terms of calorie counting. For example, if you wear a bracelet on your right hand, at lunchtime, he will understand from your movements that you bring a spoon to your mouth and sprinkle you + calories.

In everyday conditions, it was not possible to confirm this: firstly, Healbe all the same counted when transferring to the left static arm of Healbe. Secondly, in case of fraud, the bracelet should have been taken literally in real time - and it counts with a delay, but at the same time it falls into peaks.

How this is explained It

seems that Healbe is based on a bio-impedance analysis of body composition, which the team calls Flow technology. Further, it probably makes sense to think this way. The developers say that they determine the load by incoming glucose.

This is logical. And hypothetically, this is even quite real, because it works on other devices. True, they determine glucose not by blood, where it would be faster, but by interstitial fluidwhere everything goes with some delay.

According to this principle, for example, semi-invasive glucometers, such as Freestyle Libre, work. If we compare the indicators with a puncture from a finger and "Libra", then the latter catches up with blood in 15-20 minutes. Here, since we have a completely non-invasive technology, the delay may be even greater.

Given that fluid analysis is also one of the key features in positioning, the accuracy of which is confirmed by institutes, it is logical to assume that this is the way to evaluate. And this is generally scientific enough to be true.

To see how it works, we took on a pair of the most popular applications where you need to drive calories manually.


Healbe - left

Or:


Healbe - bottom

The numbers are not mirrored, of course. However, if we take into account that we are dealing with an error of 10-15%, then, for example:

1738 + 15% = almost 2000, which already generally gives not such a large spread.

On the other hand, manual input and mobile applications too - is it worth believing? In different applications for the same products there may be a calculation, which we, in general, also encountered. Well, from the simplest:



In one case, 200 grams of coffee is 30+ Kcal, in the second - more than 100. That is, we have a device that is quite difficult to confirm or refute, but which generally gives fairly close numbers to what you can score manually . To argue, of course, that this is more accurate is probably a completely ungrateful story, but with these examples it turns out that Healbe looks quite convincing.

How statistics are displayed.

Healbe statistics in the application are carried out per week. Data for the entire period can be viewed in the web version. Before your eyes you will have a day by the clock with the arrival and consumption of calories.



Data is displayed with a slight delay, but then substituted into the desired interval. Let's say yesterday about 20/00 there was a dinner, calories began to be absorbed, plus or minus a little later - this can be seen in the peak on the chart.

Or, this morning - coffee, and immediately after it the energetic:



For example, if by eye: according to the application, a cup of coffee will be about 120 Kcal ±. While the device did not calculate the total figure, it was quite similar:



Then we subtract from 356 - 156 = 200. We still have 200. The calorie content of 335 mm of Redbull will give us 150 Kcal. In general, about 50 Kcal hangs, however, it is likely that something is throwing us morning coffee in the office, and taking into account the small error that is stated, we get a rather folding and plausible picture of calorie counting.

What are the findings?

Healbe is no longer a startup, but still, perhaps, an experimental device. In today's market, this is the only gadget that analyzes the calories coming from food through the skin. In our opinion, he does this very, very believable, but you need to consider what guided us in our conclusions: we took other consumer-friendly tools available, primarily software, and compared them. This, of course, is not RCT, but it is quite honest: we compared one household tool with another household tool.

In any case, Healbe has its own unique competitive advantage, which is complemented by some other features: nevertheless, there is a heart rate monitor, the gadget analyzes sleep, the level of hydration - in general, a kind of healthy lifestyle bracelet.

I believe that, despite the given university names, the bracelet is still quite controversial. Not all parameters are interesting, some objectively do not withstand criticism, but from the point of view of the main positioning, he is undoubtedly very curious. As far as we know, work is underway on the third version, and therefore the second has significantly fallen in price. It still does not cost as MiBand, but the price is already below 10 thousand rubles .

All Articles