Hacker News new | past | comments | ask | show | jobs | submit login
My toothbrush streams gyroscope data (johannes-mittendorfer.com)
158 points by zdw on Oct 24, 2020 | hide | past | favorite | 99 comments



Gyro/accelerometer measurements alone can't tell you how well you cleaned your teeth, only how you moved the toothbrush. A simpler and more effective solution is plaque-staining dye:

https://en.wikipedia.org/wiki/Disclosing_tablets


This comment is like the one person in the room at <startup company> trying to say an idea is bad who is drowned out by others foaming at the mouth with reasons to build complexity for the sake of it


For the record I’m not putting down the author, I love a good hack for the sake of hacking :)


Or me any time I say anything negative here about cryptocurrency haha


You know what would be awesome? A toothbrush that collects gyroscopic data, sells it off to gaming companies that want to study user hand and arm behavior, so that they can optimize their games for it. Added bonus: if the toothbrush company decides to mine crypto currency for the 23hrs and 50 mins you won't be using it each day.

/s if it isn't obvious lol


The Pokémon Company recently released a tooth brushing game.


The future is tokenized brushing data


Can we make some other assumptions, then the gyroscope data is useful?


There's also a great complementary trick. If your floss gives a funny odor when examined closely after going through a single interdental gap, you have too many anaerobic bacteria there. Those are the nasty ones that destroy your gums.

This is the poorman's version of bacterial cultures and BANA tests, which IMHO were great ideas but never took off.

Dyes are good for tooth plaque, which is a different kind of biofilm. You can have good dental hygiene, but bad gum hygiene and vice versa.


I am wondering how such situation develops as flossing your teeth should in theory avoid such issue, or not?


Well, infrequent flossing would do that. I floss once a week and occasionally it leaves a slight smell, I know I should do it more often but that's a habit I've had to learn as a grown up.


> You can have good dental hygiene, but bad gum hygiene and vice versa.

Quote from Dr Zoidberg - “why not both?!”


Thanks so much for the reminder about disclosing tablets!

We had them when I was a kid in the 1970's. I've wanted a way to check the quality of my brushing, and my kids' brushing as well. I totally forgot about these.


Blue tooth > Bluetooth.


Bluetooth the tech is named after this guy https://en.wikipedia.org/wiki/Harald_Bluetooth


Dental hygienist use these here. Though n=1 (snapshot), its a useful way to figure out where you are lacking with regards to brushing, as long as you are honest to yourself when making the snapshot.


Yes. The positional data is a great way to determine that something was missed. If you missed a spot then you definitely didn't do a good job of cleaning your teeth.


Some childrens mouthwash has this built in


As with every amazing invention, the people who made the breakthrough of miniaturizing accelerometers into an integrated solid state silicon package would be proud to see it used to power mankind-saving tooth brushing advances.


Are they solid state? I thought the accelerometers were, in fact, essentially tiny boxes with a ball in with some springs? See https://en.m.wikipedia.org/wiki/Microelectromechanical_syste... ; I can't actually find a decent reference specifically for MEMS accelerometers.


Here's an actually entertaining video on it: https://www.youtube.com/watch?v=KZVgKu6v808

(the rest of the channel's videos are good too -- the aluminum can video is great)


So, how well would a smartphone work on the Moon or Mars? (ignoring the fact it is not shielded for cosmic radiation). If the spring&ball accelerators are measuring the effects of gravity, would they notice the rotation when a != 9.81m/s/s?


The way the asics were written in 2000 it should work. Should work in free-fall also.


That was really great and explained it really well, thanks!


Technically yeah they are literally a tiny weight on springs. But they are in a tiny silicon package that is effectively "solid state"


Solid state doesn’t mean “tiny”, nor “packaged in silicon”. A thing with moving parts, no matter how small or hidden, is not solid state.


You can model it as a spring, but it's not a spring in the slinky sense - it's just a thin piece of material whose electrical properties are affected by tiny changes in stress/strain.

Its operation fundamentally involves (near infinitesimal) movements, but then so does a cell phone pole. I don't see what we gain with debating its "solid stateness" (especially when hedged with "effectively"). Maybe you could point to some effect that this distinction implies?


A transistor has moving parts: electrons.

More importantly, https://en.wikipedia.org/wiki/Solid-state_electronics:

”The term "solid state" became popular in the beginning of the semiconductor era in the 1960s to distinguish this new technology based on the transistor, in which the electronic action of devices occurred in a solid state, from previous electronic equipment that used vacuum tubes, in which the electronic action occurred in a gaseous state.”

⇒ the term doesn’t reference the hardware being solid.


From the second sentence in your linked source: "The term is also used for devices in which semiconductor electronics which have no moving parts replace devices with moving parts"


That gets to be a nitpicking distinction when you get to the level of micron moving parts. If a practical definition of solid state is that something can survive, say, 20G of abuse, then an accelerometer in silicon, even with moving parts, is pretty indistinguishable from true "solid state".


Solid state as a term is a discipline of materials, practically speaking it's a bad idea to overload it with your imprecise generalization. Solid state as mentioned above specifically refers to a lack of moving parts (even at a micron level). In this space without multibody dynamics some wave properties are determinable in a much more deterministic sense.


Quartz crystal oscillators make use of the piezoelectric effect. Do they not qualify as solid state? For that matter, capacitor whine is due to electrically induced mechanical vibration as well; I've never heard anyone claim they don't qualify as solid state components though.

To me, moving parts generally refers to physically coupled but distinct pieces that move relative to one another. Otherwise heating anything that contained a piece of metal by a few degrees would disqualify it.


Electrolyte caps are not solid state and they leak - likely the most common electronic failure. I have not heard of ceramic or polypropylene caps whine. OTOH coils are solid state and they are a lot more common to cause 'whine' than any type of capacitors.


Your semantic prescription does not seem to be the common semantic prescription in electrical engineering. MEMS seems to be typically called solid state for the reasons mentioned by OP. Example in practice (second paragraph): https://www.analog.com/en/technical-articles/mems-accelerome...


Solid state really refers to not using gas plasma (vacuum tubes). Personally I would call MEMS devices like this solid state, even though it has moving parts. Everything in it is in the "solid" (physics solid) state, there is no liquid or gas involved.


Then why is my 5400 RPM HDD not an SSD?

Wikipedia:

> The term is also used for devices in which semiconductor electronics which have no moving parts replace devices with moving parts, such as the solid-state relay in which transistor switches are used in place of a moving-arm electromechanical relay, or the solid-state drive (SSD) a type of semiconductor memory used in computers to replace hard disk drives, which store data on a rotating disk.[6]


Hard drives probably count as solid state, except the interior is filled with an inert gas, so they're really a mixed phase system.

In short: neither the term definition nor usage is entirely consistent.


It's not a ball with looping-wire springs in any sense. That's just a model/illustration of the measurement principle.


My dental hygienist is always stressing the important of getting under the gums, mainly by getting the right angle on the brush. People have a tendency to just brush with the bristles perpendicular to the surface of the tooth which misses under the gum.

This could probably be useful for a fair number of people. With all the changing directions as you move to different spots in your mouth during brushing, it is not that hard to have a section where you miss doing under the gum. If you always brush using the same route around your mouth, you could miss the gums in the same section every time.

My understanding is that the point of these fancy position sensing toothbrushes is to help you avoid doing that.

I certainly would not go out and buy one, though, unless my hygienist told me that I'm missing particular spots and I failed to fix the problem by being more careful during brushing and by using something like this [1].

[1] https://www.amazon.com/GUM-Stimulator-Convenient-Refills-Rep...


I can't help but wonder whether the app also streams that information to a cloud service somewhere.


Let me give you something else to wonder about:

Given that things ranging from toothbrushes through shoes to headsets are broadcasting or reflecting radio waves, how many receivers hidden in plants, furniture, and garden gnomes does it nowadays take to construct with entirely passive surveillance a model of the goings-on within an average house or office building? (-:


Wait til you hear about the information that a single WiFi emitter's signal strength gives you:

http://rfpose.csail.mit.edu/


Pose estimation gives you a body standing in the bathroom. Recognizing the Bluetooth signal of the toothbrush gives you what that body is doing. (-:


I also looked into that and found that it extracts some metrics from it that will get stored in a cloud service called "Aurora".


>It connects to an mobile App using Bluetooth that tells you how well you brushed your teeth, shows the orientation in real-time, and also notifies you when to change the brush for a new one.

Of all the pointless internet-of-shit ideas. My toothbrush does not need an app.


My partner once received a toothbrush like this. I enjoyed the instruction manual's image showing someone using their phone instead of the mirror: https://imgur.com/a/VfT2WEl


Alternative interpretation: person is looking at the data in the app while brushing.

Of course, it could also be a replacement mirror, but given that there are tiles drawn next to it and the phone holder is a suction cup on what vaguely looks like it might be a mirror, I doubt the phone is meant to be a mirror. Or am I missing the point here and is the message simply that one should be looking in the mirror rather than their phone while brushing teeth?


I own one of these toothbrushes. It’s a total gimmick and I never used the smart features.

Here’s why: You need to look into the phone while brushing because it uses the front camera to do its thing. Pairing bluetooth and opening an app every morning is way way too much work for a 2min task. Screw that


Why does it unpair every day?


I like the Snapchat text box on top of it.


10 years ago, a Bluetoothbrush would be a joke, but now it's reality.


I remember a good 20 years ago a joke being a remotely controlled vibrator. I believe it was called iBrator, or something akin to that. Fast forward, I saw it in a video the other day..


Not only that but they’re hugely popular in camming. They hook them up to the tipping system to provide a form of viewer interaction.


It's called ohmibod, for those doing scientific research. Interesting at which point in time will youtube, twitch, pornhub and livejasmin merge into one big ...ing experience, that we yet have to invent adjective for.

Worse, I remember a joke about buying a bottled water. Now they seriously shoot bottled water reviews on youtube and tap water is a health danger.



It still is a joke (of a product), it’s just that it’s now also a reality.


It doesn't say you need the app to use the toothbrush either? It's just a gimmick to charge a premium over a regular electrical toothbrush.


It’s as if we hadn’t discovered mirrors yet.


Mirrors? You mean those webcam-connected monitors that livestream your image back to you after running image recognition to figure out how well you brushed your teeth?


I just had a weird realization that we are probably actually only 10-20 years away from people starting to use a webcam monitor combination instead of mirrors as a mainstream (albeit high-end) option.


It adds 100 milliseconds of latency for no benefit, but it only costs $1000!


And it’s sensitive to moisture! And requires power! And probably software updates. And at some point, it’ll start selling your data and showing you ads. And then one day the manufacturer will brick it remotely because they don’t want to support it anymore, even though it was still working.


.. and when inexorably it gets hacked it will get you your 15 Mb of fame.


Sibling comments might want to search for "smart mirror" ;) though they're usually an actual mirror with an embedded display.


... that's weird... I've been using my smartphone like that since my extremely late 2015 adoption of the technology.


No, I mean a scenario where instead of hanging a mirror in your high-end designed bathroom you are actually using a monitor. I can already imagine the advertisements in a catalogue where you are checking out the weather or watching CNN while brushing your teeth and it's all integrated.


And I can imagine the advertisements on the mirror you have to look at while shaving. Or before.

A quick message from our sponsors before your mirror sessions starts


"Sorry I'm looking a bit out of sorts, my mirror had a update so I wasn't able to do my makeup this morning"


For you, it was the day of your job interview for your first C-suite position.

For us at Microsoft...it was Tuesday.


well benefit is that the fancy mirror can add filters to make you feel better about yourself :)


One could use social media as their distributed mirror.


Some people use their cellphone as a pocket mirror to apply make-up in a pinch. So we're there already.


How does this stuff get funding?

Why do humans use their brains for this?


Because other humans pay money for this.

And we're probably not that far away from Health Insurance Companies using your toothbrush's data to calculate your rates.


Am I playing poker with my insurance companies?


In that insurance is a game of hidden information, yes.


The problem being solved isn’t how to have cleaner teeth, it’s how to sell people more expensive toothbrushes when the existing options do the job perfectly adequately.


I have one of the Sonicare toothbrushes. They're pricey but quite good. My dentist even asked me what toothbrush I had because my teeth were so clean. You can charge it wirelessly through a shaver socket, or with USB in the case.

I can only get replacement heads on Amazon though, I've never seen them in any UK shop.


What does get_brush_payload() do?


see toothpaste.c, line 2817


Can’t tell if I’m getting brilliantly trolled, but I can’t find such a file.


Not positive but my toothbrush has interchangeable heads and applies different settings based on what head is being used: ie some heads are gentle, some are whitening, some are gum care, etc.


I'm the author of the articles and did just now see the comments here. The brush contains a NFC tag and will open phillips.com once you hold it against your phone's NFC reader. That is the payload that comes back from this call.


There truly is a lib for everything now.


Well, at least it was fun to write it.


This is fun can't believe all the grumps here!

Here's the thing though: I've been brushing with a high tech brush for years but lately I've been developing an ache under an old crown. I delayed the dentist visit due to Covid but finally got it checked out. High tech CT scan revealed a big bubble of infection and bone loss right between the roots! Now I have to get another root canal or an implant, we'll see. I was super impressed with the imaging technology the dentist/endoguy had, much more useful than a toothbrush with gyro.


A guy I know uses 5-axis milling machines and special forges to make replica teeth from different substrates. According to him the cubic zirconia is the better option. I have seen many of these scans, and the instruments he uses are pretty impressive.

The software that runs the scans allows him view them in 3D with full rotation, correct for scanning errors and make adjustments in real-time, and has a library of cadaver teeth which can be used to automate creating a suitable replacement, which can then be modified to suit. It’s really amazing.

Then the milling machine picks its tools and does the carving out of a circular puck of the desired substrate. It’s a closed box and talks to the software over the LAN or USB.

After milling the tooth, he color matches it with powders and color swatches and then cures it in a desktop forge. Really quite amazing technology, and it fits in a couple small rooms in his home lab. He makes good money and employs his wife and another tech. Not bad for being your own boss. He gets orders from local and distant dentists, and he loans out his own scanners, as many dentists still use older or lower resolution units.

Apparently the milling machine costs around $150K, and the software is also pretty pricey. Not sure about the forge, but it’s also expensive. Not really sure how you would get into this field, but many of the skills seem to be something you can learn on the job. No idea about qualifications or licensing requirements, etc.


My dentist used this setup last time I got a crown. I got to watch as she adjusted the shape and then watch the milling. I didn't get to see it come out of the kiln, which I'm told is quite a sight since it's still glowing red hot when they take it out.

It's impressive tech and the crown fit better on the first try than my previous crowns that were done the old fashioned way.


I forgot to mention the fit is better due to the software; it can simulate occlusion from adjacent teeth and also from opposing teeth, and marks in red areas of overlap before you even mill the part, which helps get consistency and minimizes repeat visits and rework. Thanks for mentioning the bit about the forge/kiln, as I haven’t seen that part either.

The major pain point in his workflow is working with scans from older scanners and their file format quirks. He usually opens those scans in a less-capable but more compatible software, then exports to the format his main software uses, which is made by the vendor of his milling machine. Definitely risks of vendor lock-in, but he is aware of the trade-offs and is always looking for better technology. The pricing of the hardware and software seems pretty steep and maybe ripe for disruption, but for his workflow, he’s still pretty happy with how much he makes for how hard he works.


My dentist (in Norway) has had a machine for doing this on demand while you wait for years. Both of my crowns were done this way:

- take X-rays to diagnose the problem

- take photographs of the existing tooth

- set the machine running to make the crown

- grind away the surface of the old tooth, drill out whatever needs removing as usual

- apply adhesive

- fit new tooth

- adjust tooth slightly to fit by grinding as usual for any filling.

From arriving at the dentist to paying the bill is less than an hour.


One of the major differences is the scanning tools involved, which also eliminates the need for redundant xrays, as the scanner is an oral device that scans the mouth digitally using a 3D scanner inserted in the mouth and passed along the gumline and on all sides of the teeth.


What redundant X-rays?


The person I replied to said they needed xray(s) to get their replacement tooth made, and worded their reply to imply that Norway has had this tech for some time, implying or assuming that the tech in Norway they experienced is the same as what I described. It is not the same in this key way: the system I described doesn’t use or require xrays, surely not additional ones over and above the ones many dentists already do with new patients.

Xrays are potentially harmful and exposure to them should be limited to those medically necessary. The system I described does not require “extra“ xrays or any additional xrays at all, thus my characterization of any xrays the person I replied to received as redundant compared to the treatment plan I was describing.

Is that clear? Hard to assume what the person in Norway got from their care but they didn’t get the same thing I described.

Edit: I see now you’re the same person I was replying to; I don’t mean to be pendantic, but the absence of xrays in the workflow I described is kind of a big difference, as what I described uses visual 3D scanning with an orally-inserted imaging device with no need for more xrays if the dentist doesn’t otherwise need them.


The X-ray was nothing to do with the crown it was simply the normal X-ray taken to check the condition of my teeth and it was that together with visual inspection that made my dentist decide that a crown was necessary. The rest of the process was essentially as you described except that it took place immediately after diagnosis. I just had to wait a few minutes for it to be completed the order did not have to be sent to a separate organization merely to a machine in another room.


It's such a shame for profit companies abuse computers and computer users this way. Some of these ideas are really cool and could make people's lives better but they get given a terrible name. We're quickly reaching the point where the general public is becoming very afraid of technology.


I understand how technology works, but this understanding, and understanding of the incentives and inner workings of most IT companies, just makes me trust it less. Nothing is reliable, almost nothing caters to the needs of the users, everything actively discourages any kind of education unless it's for monetary benefit. I miss the times when it was expected that people actually learn how stuff works to make most of it. Now we have someone else's trusted public keys burned into OTP ROMs in our personal devices, and not many people are even aware of that.


Not all software is written and controlled by a corporation.


Is there any mainstream operating system that runs on end users' devices that's not written by an American corporation?


GNU/Linux runs on most things. The contributions come from many people, some are American or German corporations. The FSF is technically an American company, they don't behave like one though.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: