Tangential, but I have sleep apnea. Fortunately I have it mild enough to get away with a mouthpiece instead of a CPAP (which is good anyway because I also have bruxism).
The mouthpiece works great and I would recommend everyone get tested for sleep apnea if your insurance covers it, but I have to admit that paying for it bothered me. Even with insurance covering some of it, it cost me about $600.
I know that there's a deceptively high amount of engineering required for these kinds of things, but it was very hard to wrap my head around paying $600 for what amounted to a couple pieces of clear plastic. I actually got them to send me the STL of the scan of my teeth, and some back of napkin math indicated that it would have cost me about fifteen cents of resin to print it out myself. Instead I'm paying about 4000x that price.
Obviously this is not apples to apples, I'm sure they're using different and/or better resin that what I have, and as I said there's probably engineering and fine-tuning for this, but even still it was not fun to pay for.
All the same, I sleep like 10x better, so I suppose that considering that $600 is a cheap price to pay.
I think your comment would have been more insightful if you had at least pretended to try to account for labor before saying "They charged me 4000x the cost of goods!".
To be clear I'm not even asking for you to account for the cost of your printer, the 3d scanner, and software licenses in your math. Let's assume that all those are free. How many hours of specialized human time was spent on consultations, scans, design, reviews, to produce working guards for you?
The next question is then, of course, how much do they charge for subsequent guards now that the scan has been done and validated? Is it still 4000x the cost of raw resin?
I suspect the introduction of the infinite-money pits of insurance doesn't help either.
Since I do have the 3D scan of my teeth, I've debated designing my own, but I'm not sure which resins to buy that I could safely put in my mouth every night.
Might there be microscopic layer lines? Or other unknowns you're not familiar with? Making 3d prints that can be cleaned is non-trivial, maybe there is a surface finish involved, etc.
Also how do you know your design is correct? Won't cause your teeth to move? A 3d scan doesn't mean you know what a mouth guard should look like.
All of a sudden, having a product that's made with a vetted process is pretty attractive -- and 600 USD seems like a bargain.
What's the cost of having your teeth fixed, if they accidentally move? (Not to mention the discomfort, which can be considerable)
> I know that there's a deceptively high amount of engineering required for these kinds of things
I think there's a deceptively low amount of engineering required for most medical and medical-adjacent tech. The high costs are rooted in pervasive industry-wide centuries-long FUD campaigns.
I’ve tried custom molded mouthpieces for grinding but couldn’t adjust to sleep with them. I notice them and feel the discomfort of it not feeling natural despite it being a pricey professionally manufactured piece. It just feels too artificial. I also wonder about the long term effects of polymers in the mouth for long periods of time daily.
It took me a couple weeks to get used to it, but get used to it I did. I would still prefer a universe where I didn't have to wear it, but at this point it doesn't bother me too much.
I agree the plastics might be bad for me, but I justify it because I suspect that just continuing with the apnea is almost certainly worse for me, considering how many diseases appear to be caused by it.
So shouldn't this really be something that could be opened sourced. I think I've seen a few write ups of people that did their own, but seems like a highly functional implementation could be democratized.
It is definitely not that simple for a number of reasons.
Yes, aligners and retainers in theory may be printed on some commercially available hardware. At your own risk, because you will be printing a medical device and you will need:
1. A treatment plan: simulated movement of teeth at every step, taking into account all forces. That’s specialized software or external lab service.
2. Precision. You put too much pressure at the wrong angle and you will need a surgery to fix the damage, because the tooth root moved in wrong direction.
3. Plastic. You cannot use ordinary 3D printer ink. You need a plastic that can survive the chemical environment in your mouth, maintain the pressure, and you probably want it to look good (no discoloration etc).
4. Finish: Align Tech, Straumann etc do not stop after 3D printing, there are few other steps involved to make sure there’s no sharp edges etc.
5. Maybe you will need attachments (to focus pressure in the right direction on certain teeth) or wires.
Align Tech is Apple of clear aligners, but now competition exists, producing aligners at scale is commercially more efficient, considering all the risks and required qualifications, and of course the best materials for aligners are patented and not sold OTC to everyone.
Disclosure: I worked at Align 10 years ago and later was CTO of European DTC competitor.
How many of these risks and problems are exaggerated in scope and potential due to both a desire for a regulatory moat and a general fear of litigation in the medical space?
That is to say, how good is “good enough” when done at small-scale in developing nations or medically underserved communities?
1. There‘s always a reason or two for the treatment. The problems with bite may affect your health in various ways. There’s aesthetic component in it, sometimes even cultural preferences for how your teeth should look like. But let’s say patient cannot afford to care about final position when signing up for the treatment. They just trust the doctor saying they need it. Failure mode: patient is unhappy with outcome and will ruin your business from marketing perspective.
2. Let‘s say the practice does it in old way, with impressions - no intraoral scanner. The scanner of impressions still needed, but it can be cheaper. Someone needs to build it and achieve required accuracy (let’s say, 50 μm). Who? Why? Failure mode: bad scan leads to aligners not fitting your teeth from day 1. Oops.
3. Let‘s say someone builds a good OSS alternative to OrthoCAD (Who? Why?), so that orthodontist on site could build a treatment plan and export it into series of 3D models for printers. Failure mode: good treatment plans are rarely possible or output is garbage (aligners do not fit, cause pain etc)
4. Maybe some company develops good plastic or patent expires, so that it is possible to produce it in India, China or other inexpensive location with strong industrial base. That would be cool, otherwise: non-compliant plastic breaks in patient’s mouth, decomposes with patient ingesting some toxic chemicals or is simply not strong enough to move teeth in desired position, so you have problem with 2nd aligner.
5. Maybe you get to this point, but you still need a printer that can maintain the same precision in printing. And you need a good cutting and finishing process. Someone needs to build such device. Failure modes are similar to the mentioned above.
6. The ortho supervision sounds easy, but how many patients in developing countries do even have a possibility to see orthodontist? They are definitely not in position to treat themselves.
So, in this process, what is good enough exactly? Who and why would drive the costs down while building an on-site solution?
These kinds of things can slowly move your bone structure over time. After all, that is their entire point. You don't want to accidentally mess up your teeth and jaw even more.
IIRC, yes. It’s been some time ago, I don’t know how manufacturing looks now. It’s different process compared to 3D printing at home. It doesn’t mean it should be different, it just has to maintain certain properties. I’m not chemical or bioengineer to go into detail of it :)
I did Invisalign a few years ago. Manufacturing the retainers is surely only a small part of the puzzle.
They used a specialized sort of 3D camera on a stick to get an incredibly accurate model of my mouth, any open source solution would need an equivalent. And you’d also need open source code from somewhere to work out which teeth need to move where and at what stage in the treatment.
They also use this camera system when creating implants. After the implant post was installed, they scan your mouth to determine the optimum shape for your crown (that goes on the post).
Dry heaving would have been great. I would regularly vomit from impressions. My orthodontist would just prepare two sets if impression trays, cause the first one was going to go in the medical waste bin.
Impressions for invisilign (when I did it, about a million years ago) weren't so bad though. Unfortunately invisilign resulted in an open bite for my molars, which I really should go back to an orthodontist to address, but I'd rather not.
While both were originally companies based in Israel, the technology behind the Kinect is different.
iTero scanners (owned by Align Technology) use parallel confocal imaging via red light lasers. Their newer models also use Multi-Direct Capture techniques.
Kinect used a Light Coding technique, an infrared projector and camera. It was developed by a company called PrimeSense, which was later purchased by Apple.
>The difficult part is not the manufacturing, but knowing how to do it properly so you don’t harm the patient.
And yet I read plenty of horror stories of bad orthodontic results. Ask me how I know.
Went to 3 different orthodontist to fix what a bad orthodontist did to me when I was a kid, and each gave me a completely different treatment plan. I feel like being an orthodontist is just eyeballing and patching your way as you go to an acceptable resolution.
Certainly it's not impossible to DIY, but it's more difficult than just popping some aligners on your 3d printer.
Manufacturing them requires a resin printer and a vacuforming setup, but that's still the easy part. It's a whole system with a dental 3D scanner, software for rearranging your mouth, and attachment points that have to be epoxied onto (and later removed from) your teeth by a dentist.
They have to have at least 2 different materials as well. The temporary trays were much softer and I had almost ground through them in my sleep by the time I had to switch to the next one but the final set is much more robust.
A point I didn’t see sibling comments make is that the dentist often has to file between teeth for them to sit and align correctly. They did so several times in my case. I would not want to do that to myself!
It’s been tried, with some success. Pretty sure I’ve seen a post here on HN from someone that DIy’d it end to end.
But it’s also something that’s not responsible to shortcut. Shifting teeth around too fast can result in permanent root damage and even loss of teeth. There was a whole cottage industry in the US for a while focused on under cutting Invisalign with a reverse-engineered product, but they often moved on accelerated treatment timelines that caused a not-insignificant amount of harm to patients, and cut corners on intake (DIY at home mold kits) that also contributed to problems. Pretty sure all of the companies doing this are basically dead now.
Need expensive printers and you need CAD software that can correctly move the teeth. Also not all it can be done by software, sometimes you need to blank out certain teeth that dentist will make the call.
If you keep them on soft food young, the jaw muscles don't get stimulated enough to grow so that the teeth that are coming out sit properly. Misalignment and opportunity for decay ensues.
If you give them decently hard food early on in life you solve most problems.
After that it can still happen but there's less chance.
Tangential, but I have sleep apnea. Fortunately I have it mild enough to get away with a mouthpiece instead of a CPAP (which is good anyway because I also have bruxism).
The mouthpiece works great and I would recommend everyone get tested for sleep apnea if your insurance covers it, but I have to admit that paying for it bothered me. Even with insurance covering some of it, it cost me about $600.
I know that there's a deceptively high amount of engineering required for these kinds of things, but it was very hard to wrap my head around paying $600 for what amounted to a couple pieces of clear plastic. I actually got them to send me the STL of the scan of my teeth, and some back of napkin math indicated that it would have cost me about fifteen cents of resin to print it out myself. Instead I'm paying about 4000x that price.
Obviously this is not apples to apples, I'm sure they're using different and/or better resin that what I have, and as I said there's probably engineering and fine-tuning for this, but even still it was not fun to pay for.
All the same, I sleep like 10x better, so I suppose that considering that $600 is a cheap price to pay.
I think your comment would have been more insightful if you had at least pretended to try to account for labor before saying "They charged me 4000x the cost of goods!".
To be clear I'm not even asking for you to account for the cost of your printer, the 3d scanner, and software licenses in your math. Let's assume that all those are free. How many hours of specialized human time was spent on consultations, scans, design, reviews, to produce working guards for you?
The next question is then, of course, how much do they charge for subsequent guards now that the scan has been done and validated? Is it still 4000x the cost of raw resin?
>Even with insurance covering some of it, it cost me about $600.
Well that's pretty close to how much the entire CPAP system would cost depending on the sale.
> All the same, I sleep like 10x better, so I suppose that considering that $600 is a cheap price to pay.
First time?
Price is not determined by cost to produce.
It’s determined by the price a customer is willing to pay.
Every company operating on that basis is a busniess waiting to be disrupted by a cost-plus competitor.
The company competing on cost plus is wasting resources by solving problems that arent very profitable to solve
As is every individual earning more than median income.
Every business still operating on that basis has a moat somewhere else.
An important corollary to this rule is that that customer need not be you.
And the material cost also isn't the main thing you're paying for, it's time and expertise involved in scanning and designing it.
I suspect the introduction of the infinite-money pits of insurance doesn't help either.
Since I do have the 3D scan of my teeth, I've debated designing my own, but I'm not sure which resins to buy that I could safely put in my mouth every night.
Even if you find a safe resin.
Might there be microscopic layer lines? Or other unknowns you're not familiar with? Making 3d prints that can be cleaned is non-trivial, maybe there is a surface finish involved, etc.
Also how do you know your design is correct? Won't cause your teeth to move? A 3d scan doesn't mean you know what a mouth guard should look like.
All of a sudden, having a product that's made with a vetted process is pretty attractive -- and 600 USD seems like a bargain.
What's the cost of having your teeth fixed, if they accidentally move? (Not to mention the discomfort, which can be considerable)
> I know that there's a deceptively high amount of engineering required for these kinds of things
I think there's a deceptively low amount of engineering required for most medical and medical-adjacent tech. The high costs are rooted in pervasive industry-wide centuries-long FUD campaigns.
> centuries-long FUD campaigns
That dastardly Ben Franklin with his bifocals..
I’ve tried custom molded mouthpieces for grinding but couldn’t adjust to sleep with them. I notice them and feel the discomfort of it not feeling natural despite it being a pricey professionally manufactured piece. It just feels too artificial. I also wonder about the long term effects of polymers in the mouth for long periods of time daily.
It took me a couple weeks to get used to it, but get used to it I did. I would still prefer a universe where I didn't have to wear it, but at this point it doesn't bother me too much.
I agree the plastics might be bad for me, but I justify it because I suspect that just continuing with the apnea is almost certainly worse for me, considering how many diseases appear to be caused by it.
So shouldn't this really be something that could be opened sourced. I think I've seen a few write ups of people that did their own, but seems like a highly functional implementation could be democratized.
It is definitely not that simple for a number of reasons. Yes, aligners and retainers in theory may be printed on some commercially available hardware. At your own risk, because you will be printing a medical device and you will need:
1. A treatment plan: simulated movement of teeth at every step, taking into account all forces. That’s specialized software or external lab service.
2. Precision. You put too much pressure at the wrong angle and you will need a surgery to fix the damage, because the tooth root moved in wrong direction.
3. Plastic. You cannot use ordinary 3D printer ink. You need a plastic that can survive the chemical environment in your mouth, maintain the pressure, and you probably want it to look good (no discoloration etc).
4. Finish: Align Tech, Straumann etc do not stop after 3D printing, there are few other steps involved to make sure there’s no sharp edges etc.
5. Maybe you will need attachments (to focus pressure in the right direction on certain teeth) or wires.
Align Tech is Apple of clear aligners, but now competition exists, producing aligners at scale is commercially more efficient, considering all the risks and required qualifications, and of course the best materials for aligners are patented and not sold OTC to everyone.
Disclosure: I worked at Align 10 years ago and later was CTO of European DTC competitor.
How many of these risks and problems are exaggerated in scope and potential due to both a desire for a regulatory moat and a general fear of litigation in the medical space?
That is to say, how good is “good enough” when done at small-scale in developing nations or medically underserved communities?
1. There‘s always a reason or two for the treatment. The problems with bite may affect your health in various ways. There’s aesthetic component in it, sometimes even cultural preferences for how your teeth should look like. But let’s say patient cannot afford to care about final position when signing up for the treatment. They just trust the doctor saying they need it. Failure mode: patient is unhappy with outcome and will ruin your business from marketing perspective.
2. Let‘s say the practice does it in old way, with impressions - no intraoral scanner. The scanner of impressions still needed, but it can be cheaper. Someone needs to build it and achieve required accuracy (let’s say, 50 μm). Who? Why? Failure mode: bad scan leads to aligners not fitting your teeth from day 1. Oops.
3. Let‘s say someone builds a good OSS alternative to OrthoCAD (Who? Why?), so that orthodontist on site could build a treatment plan and export it into series of 3D models for printers. Failure mode: good treatment plans are rarely possible or output is garbage (aligners do not fit, cause pain etc)
4. Maybe some company develops good plastic or patent expires, so that it is possible to produce it in India, China or other inexpensive location with strong industrial base. That would be cool, otherwise: non-compliant plastic breaks in patient’s mouth, decomposes with patient ingesting some toxic chemicals or is simply not strong enough to move teeth in desired position, so you have problem with 2nd aligner.
5. Maybe you get to this point, but you still need a printer that can maintain the same precision in printing. And you need a good cutting and finishing process. Someone needs to build such device. Failure modes are similar to the mentioned above.
6. The ortho supervision sounds easy, but how many patients in developing countries do even have a possibility to see orthodontist? They are definitely not in position to treat themselves.
So, in this process, what is good enough exactly? Who and why would drive the costs down while building an on-site solution?
These kinds of things can slowly move your bone structure over time. After all, that is their entire point. You don't want to accidentally mess up your teeth and jaw even more.
Smile Direct Club and co have shown its founded 100% in real risk of harm.
They were a company theoretically doing the same thing with still more resources than an average individual has, and ruined people's bites and teeth.
I don't think there's a good enough here
Per point #3, aren't the liners thermo formed around at 3d printed model of your teeth?
IIRC, yes. It’s been some time ago, I don’t know how manufacturing looks now. It’s different process compared to 3D printing at home. It doesn’t mean it should be different, it just has to maintain certain properties. I’m not chemical or bioengineer to go into detail of it :)
I did Invisalign a few years ago. Manufacturing the retainers is surely only a small part of the puzzle.
They used a specialized sort of 3D camera on a stick to get an incredibly accurate model of my mouth, any open source solution would need an equivalent. And you’d also need open source code from somewhere to work out which teeth need to move where and at what stage in the treatment.
They also use this camera system when creating implants. After the implant post was installed, they scan your mouth to determine the optimum shape for your crown (that goes on the post).
That’s pretty wild. I got a mould taken with some sort of a putty .
I wear a night guard and have had them made both ways.
The 3D camera was really neat. A little faster, and I didn’t once dry heave.
I could watch the software and a 3D model slowly form of my mouth. Looked surprisingly user friendly. Missed areas were highlighted, for example.
> A little faster, and I didn’t once dry heave.
Dry heaving would have been great. I would regularly vomit from impressions. My orthodontist would just prepare two sets if impression trays, cause the first one was going to go in the medical waste bin.
Impressions for invisilign (when I did it, about a million years ago) weren't so bad though. Unfortunately invisilign resulted in an open bite for my molars, which I really should go back to an orthodontist to address, but I'd rather not.
>They used a specialized sort of 3D camera on a stick to get an incredibly accurate model of my mouth
AFAIK Align's 3D scanning system is more or less branched from the same Israeli tech that went into the Xbox 360 kinect camera and the iPhone face-ID.
While both were originally companies based in Israel, the technology behind the Kinect is different.
iTero scanners (owned by Align Technology) use parallel confocal imaging via red light lasers. Their newer models also use Multi-Direct Capture techniques.
Kinect used a Light Coding technique, an infrared projector and camera. It was developed by a company called PrimeSense, which was later purchased by Apple.
Orthodontics is simply “making a retainer” the same way orthopedics is simply “putting screws in a leg”.
The difficult part is not the manufacturing, but knowing how to do it properly so you don’t harm the patient.
As the joke goes, $10 to tighten the bolt, $90 to know which bolt to tighten.
>The difficult part is not the manufacturing, but knowing how to do it properly so you don’t harm the patient.
And yet I read plenty of horror stories of bad orthodontic results. Ask me how I know.
Went to 3 different orthodontist to fix what a bad orthodontist did to me when I was a kid, and each gave me a completely different treatment plan. I feel like being an orthodontist is just eyeballing and patching your way as you go to an acceptable resolution.
Given that, shouldn't you be even more concerned about people YOLOing it, if even highly trained orthodontists are regularly screwing this up?
Certainly it's not impossible to DIY, but it's more difficult than just popping some aligners on your 3d printer.
Manufacturing them requires a resin printer and a vacuforming setup, but that's still the easy part. It's a whole system with a dental 3D scanner, software for rearranging your mouth, and attachment points that have to be epoxied onto (and later removed from) your teeth by a dentist.
They have to have at least 2 different materials as well. The temporary trays were much softer and I had almost ground through them in my sleep by the time I had to switch to the next one but the final set is much more robust.
Yeah it's also not unreasonably expensive. At least when I had them it was only a few thousand pounds. I think they do offers regularly.
A point I didn’t see sibling comments make is that the dentist often has to file between teeth for them to sit and align correctly. They did so several times in my case. I would not want to do that to myself!
It’s been tried, with some success. Pretty sure I’ve seen a post here on HN from someone that DIy’d it end to end.
But it’s also something that’s not responsible to shortcut. Shifting teeth around too fast can result in permanent root damage and even loss of teeth. There was a whole cottage industry in the US for a while focused on under cutting Invisalign with a reverse-engineered product, but they often moved on accelerated treatment timelines that caused a not-insignificant amount of harm to patients, and cut corners on intake (DIY at home mold kits) that also contributed to problems. Pretty sure all of the companies doing this are basically dead now.
Also no one has mentioned many mouths needed some teeth shaving to re-align. You are NOT doing that at home.
Also who’s attaching the attachments (I had 13 at first) to the teeth to help the aligners grab hold?
https://amosdudley.com/weblog/Ortho
Need expensive printers and you need CAD software that can correctly move the teeth. Also not all it can be done by software, sometimes you need to blank out certain teeth that dentist will make the call.
There’s a dental procedure in Latin America that is cheaper but only a few places in the US do it.
I don’t like this interviewer’s tone, at all. It was like… confrontational but in a teenage way.
Most teeth align themselves as kids age. This is almost never necessary.
Yes, the 12 billion dollar company must be doing something that no one needs.
Until wisdom teeth come in later on and cause crowding or visual issues
There are folks working on solutions for when teeth do not align themselves.
Depends on diet.
If you keep them on soft food young, the jaw muscles don't get stimulated enough to grow so that the teeth that are coming out sit properly. Misalignment and opportunity for decay ensues.
If you give them decently hard food early on in life you solve most problems.
After that it can still happen but there's less chance.
Also, breastfeeding and not weaning them off on to "baby foods" too soon is better in the same ways you describe.
How early on in life?
Ok, sure, it's usually cosmetic. So what?