UKC

Is it discrimination if people born with a genetic defect are charged to buy spectacles so that they can see well enough to do their job and function in society?

 

 

In reply to Jon Greengrass:

If you're an obese, short sighted, chain smoking alcoholic you're pretty much screwed these days.

In reply to Jon Greengrass:

Deaf people get hearing aids free on the NHS.

(Ok, they're pretty crappy ones, but still....)

In reply to Jon Greengrass:

https://www.theguardian.com/news/2018/may/10/the-invisible-power-of-big-gla...

An interesting read if you have 20 mins spare. Glasses cross over from medical assistance into fashion , and the companies involved in manufacture have somehow managed to keep hidden the real cost of lens manufacture and therefore have managed to keep costs to consumer ridiculously high, making it tough for the NHS to warrant spending the money I would imagine.

In reply to Jon Greengrass:

They don't have to. The NHS will provide spectacles for free that will enable them to see well enough to function.

 

They will just have the unfortunate side effect of looking like a 1950's trainspotting paedo.

In reply to Jon Greengrass:

I also have a hereditary genetic condition that means that my hair has fallen out.

I now need to wear a hat whenever I'm outside; obviously I need this in order to do my job and function in society etc. at the moment I'm stuck indoors until deliver of my NHS bobblehat, flatcap or other.

I'm actually after a woven straw pork-pie hat with hatband - I think I can wangle un upgrade by claiming lack of self esteem if forced to go out of the house looking like a Yorkshireman

In reply to Just Another Dave:

Not true unless

• you are under 16
• are 16, 17 or 18 and in full-time education
• are a prisoner on leave from prison
• are eligible for an NHS complex lens voucher – your optician can advise you about your entitlement

 

You also qualify for an optical voucher if you:

 

• get Income Support
• get Income-based Jobseeker’s Allowance
• get Income-based Employment and Support Allowance
• get Pension Credit Guarantee Credit
• get tax credits and meet the criteria
• get Universal Credit and meet the criteria
• have a low income and are named on a valid NHS HC2 certificate for full help with health costs

 

 

In reply to Big Ger:

My Audiologist other half would disagree quite strongly that the NHS hearing aids are crappy. Although she would acknowledge that localised procurement means some variation in the product.

If you get your NHS hearing aid through a high street supplier such as Specsavers, it is likely to be an inferior product, compared to the one supplied directly by the NHS Trust and of course Specsavers have an incentive to sell you a more expensive product.

I appreciate that is a long response to a throwaway remark.

In reply to Jon Greengrass:

I think the prices for glasses and hearing aids are totally crazy.

My wife paid about 3x as much for a pair of glasses as I paid for an intel i7 processor and more than 4x what Amazon charged me for one of the fancy new Alexa Show devices with the largish LCD touchscreen.   The manufacturing tolerances in something like an i7 processor and complexity of the electronic devices is hundreds of times that of a couple of polished bits of glass. The signal processing in something like Alexa is in a different league to that in a hearing aid.  Yet the consumer electronics guys can make a profit selling it at a fraction of the price.

I don't think the price difference between consumer and medical devices which don't need surgery to implant can last.   Cameras in cellphones will get good enough so eye exams can be done almost as effectively with an app allowing a whole new business model completely cutting out the 'medical professionals' that act as a gatekeeper and create the quasi-monopoly that prevents competition lowering prices.  It'll be a case of downloading the app checking your eyes yourself to get a prescription and buying the glasses from Amazon at about 1/10 of what they cost now.  Hearing aids are probably even easier to do without any high-price human experts.  Chances are that after a few years the machine learning algorithms will do a better job than the people at diagnosing any medical conditions as well.

In reply to Jon Greengrass:

But what if you have bad eyesight from not paying heed to mothers warnings.

In reply to tom_in_edinburgh: >  Cameras in cellphones will get good enough so eye exams can be done almost as effectively with an app allowing a whole new business model completely cutting out the 'medical professionals' that act as a gatekeeper and create the quasi-monopoly that prevents competition lowering prices.  It'll be a case of downloading the app checking your eyes yourself to get a prescription and buying the glasses from Amazon at about 1/10 of what they cost now.  



I look forward to Jon Stewart's reply to this one

[edit] er....it IS Jon Stewart who posts occasionally about his job as an optician/optometrist, isn't it?

 

In reply to tom_in_edinburgh:

You can already buy online at a small fraction of the cost, all you need is your prescription.

In reply to Jon Greengrass:

Thanks, I thought this was the case, and was being careless and facetious.

And given that it is the case, I do agree it's wrong for the NHS not to provide some form of free eye correction that is at least good enough to function in society .

And whether it is down to congenital defect, or, as someone above just asked, to lack of self-care, ---well the same applies to other long-term conditions .T2DM etc.

In reply to The New NickB:

After getting tested for free at Tescos

In reply to Blue Straggler:

I am not he, but I have looked into this somewhat tangentially in a professional capacity in the past.  

You don’t need a smart phone app to measure your own focal errors.  I’ve worked with Joshua Silver’s adjustable liquid lens eye glasses.  Most people can rapidly adjust these to improve their vision in a minute or so - effectively and implicitly measuring their focal error.  I couldn’t as I don’t have any vision problems but watching them in action it’s clearly trivial - and evidence suggests most individuals self-measure more accurately than an optician can measure them...

A phone app can measure the quality of someone’s eyesight, but without additional optics it can’t meausre the far limit of their focous, and has no accurate measure of display/eye distance to measure their near focous, so it can’t deliver a proscription.  There are ways to infer the proscription - including astigmatism - through tests where the human evaluates a series of carefully designed images, but it’s not simple in any of the ways that matter.

The real benefit of a human optician - missed by app or optical self measurement - is that a well trained human looking down an opthalmascope can spot the early signs of a whole bunch of health problems visible on the retina.  

Really that health screening should by orthogonal to eye glasses I think - the danger of new technology gutting the ophthalmology business is that in reality these things aren’t orthogonal.

In reply to wintertree:

Thanks, good comments, I had not heard of Joshua Silver, sounds interesting. 

In reply to Jon Greengrass:

I got 5x glasses from a well known online retailer for 10:99 these cover my reading needs. This was fine according to the optician who did my eye test (independant).

In reply to The New NickB:

 

But a useful one in what is an otherwise rubbish thread. Thank you and I shall file this in my memory banks.

In reply to wintertree:

The husband of my wife's boss had an ambulance called for him straight away during an eye exam as the optician spotted the onset of (I think) a stroke which was serious enough to warrant immediate medical attention.

Like others have said above, there's a crossover between medial need and fashion and its perfectly possible to get cheap glasses if you don't care what they look like. 

In reply to alx:

Can you type that in larger font please? 

In reply to The New NickB:

'You can already buy online at a small fraction of the cost, all you need is your prescription.'

This ain't necessarily so. Yes you can buy a cheap pair of glasses for less than you will pay at a bricks and mortar opticians; but if you start going to quality branded lenses, with branded coatings, and genuine designer frames (the internet is of course full of fakes) then you won't be saving much, if at all.

In reply to tom_in_edinburgh:

There's already a few companies selling apps although still with clip on lenses and still trying to work through the health care providers rather than end-run round them:

https://qz.com/629270/a-kenyan-smartphone-app-is-preventing-blindness-in-ki...

In reply to Big Ger:

Not true, the standard NHS aid is pretty good and makes you realise how much people who pay thousands are being ripped off. They are certainly good enough that my sick MIL's hearing aids would vanish from her sick room in the care home whenever agency staff were on duty for the night. The same used to happen to two other residents on the same nights so I don't think we can be accused of irrationality.

In reply to The New NickB:

It's a good and informative response Nick, my thanks.

However, I've gone private in any case.

In reply to tom_in_edinburgh:

That app just diagnoses very bad eyesight, it doesn’t - and can’t - give a proscription to fix it.  It can’t even identify if the problem is in a person’s optics, or in cataracts, or their eye fluids, or the brain stuff.

What it does do I suspect is overcome the standard “letter chart” paradigm for measuring eye sight, which is surprisingly useful thing to do, as bad eye sight and illiteracy (which voids using the letter chart stuff) go hand in hand in the developing world.  The exact same thing could be achieved with paper with arrows or USAF test charts on.

 

In reply to wintertree:

 

Agreed, my optician noted some damage "white spots" on my retina, and I was advised to wear sunglasses when out in the Aussie sunshine.

In reply to Rigid Raider:

My experience is different.

 

In reply to tom_in_edinburgh:

Perhaps if people only bought goods from companies with credible track records of paying taxes, there would be more revenue for governments to spend on healthcare?

In reply to wintertree:

There's at least 4 or 5 companies in the space.   This one has a clip on retinal imaging camera:

https://www.d-eyecare.com/en_GB/shop

They're still selling it as something for professionals but I don't see why it would be a problem for image processing software to spot anything an optician can spot in the image and it's just a matter of time before someone does to opticians what Uber did to licensed taxi drivers.

 

In reply to summo:

Perhaps if people weren't continuously getting ripped off by government sponsored anti-competitive industries they'd have money left over at the end of the month to spend on products they actually want.

In reply to tom_in_edinburgh:

The counter argument to that, surely, is that you get what you pay for - no doubt a cabbie with the Knowledge would say they're 'better' than the Uber driver.

It's hard not to get steamed by the alleged mark-ups in the article linked above, but that aside if a pair of glasses costs me, say, £300 over the expected lifespan (several years wearing every day - and being almost immeasurably more functional with them), it's not so much of a cost. Obviously getting the same benefit for less would be nice, but quite a lot of the cost pays - I hope - for the time and skill of the opticians and technicians

In reply to Big Ger:

ActuallyI have recently got mine on the NHS and I am delighted with them. Comfortable, digital, discreet and compact, and they WORK

In reply to nufkin:

They can say what they like and if they persuade the customer of their argument then fair enough.

The problem is when the cabbie goes to Transport for London or the optician goes to the Department of Health and tries to get competition based on new technology banned so the customer doesn't have an option and they don't have to cut their prices.

 

In reply to Jon Greengrass:

Well, me being quite "blind" these days (since about age 25: -6 roughly) have often wondered about this.  However, I am quite happy to have great contact lenses which don't cause me any bother and cost only £25 a months including solutions and checkups. I can see very well with them and yes, my eyes are watering a bit after 20 years of usage, but really, I leave them in most times and it's just like normal vision. Without it, with glasses only, it's a nightmare in an outdoor environment. Steamed up glasses, the glasses break, etc. Yes, ideally , somebody would pay for them, but I am quite happy to fork this out for good eyesight and care.

In reply to tom_in_edinburgh:

As someone whom develops medical devices for a living I can with some authority say you have got it wrong.

In short a medical device regulation (outgoing MDD or the new MDR May 2015) is incredibly tight. Unlike a high street computer which has no detailed intended purpose, and only needs to meet the CE or 510k FDA equivalents a medical device needs an “intended purpose” in which all the precautions, safety measures, specificity, false-positive, false-negative, and precision tolerances are tested to the nth degree. You also need to pay for market surveillance for the lifetime of your device (a bit like reporting side affects to drugs).

This is actually very tough and depending on what class of medical device can be incredibly expensive. All these things plus the higher manufacturing costs due to the quality assurance means cost of production and cost of device to the purchaser is higher.

In short if your i7 falls over, it’s no big issue, you troubleshoot the problem, wipe the hard drive, reinstall and move on with a days delay. The circuitry in life support system, gamma knife, oxygen delivery device or even a hearing aid will be infinitely simpler (decades even behind the available current high street tech) but meeting the requirements to ensure it doesn’t break at a critical time or break in a critical way (e.g deafen the user, oversupply the oxygen to a premature baby or irradiate a healthy piece of tissue) are significant.

In reply to alx:

Forgot to add prescription glasses are class 1 medical device which is the lowest risk as the user can always take them off, manufactures self certify and unlike a hearing aid which is most likely a class 2a active device due to its semi in dwelling position, probably needs to be water ingress tested, has an energy source, and has the added risk of a battery which could leak in an ear before you could stop it.

In reply to alx:

Thanks for the interesting post.  

Hearing aids are a great example of why medical devices cost so much more than consumer ones.  A mobile phone battery explosion is one thing...   That being said a quick look at the USA shows the ugly extremes price gouging can go to.

I thought more along the “computers analysing medical imagery” side of Tom’s post which I think counts as a less regulated area than medical devices, depending on what you call the outputs (indicative vs diagnostic type of distinction?).

I’m sure that one day computers will trump the best radiologists and ophthalmologists etc when it comes to interpreting medical imagery, but it’s a long way off.

Bioimageinformatics (or whatever) is littered with failed projects that start of gushing enthusiasm about replacing trained humans in some area of medical image analysis...

It’s particularly difficult with ophthalmology because the first pass human analysis is done live looking down the microscope and deciding where to look next whilst the victim is in the machine, it’s not all offline post-processing like with CT, MRI, PET etc.  A really interesting area with handheld decides capable of resolving individual foveal cones being just about here...  Although the devices for this are a loooong way off being replaceable with an iPhone app!  I mean you could strap an iPhone to it as a marketing gimmick...  Which is a bit fashionable in medical devices research for the developing world at the moment...

 

In reply to Blue Straggler:

Found myself needing glasses about 12 months ago for reading fine print. I too thought one eye test would be as good as another so took the cheapskate option of a well known high street chain. Wasn’t happy with the prescription so went to a more expensive independent opticians on recommendation.

Not only is the prescription spot on but they also spotted cataracts developing in one eye.

I need my eyes. In the future I will always seek out a well regarded professional to look after them.

In reply to Stichtplate:

That is very interesting to hear, thanks. May I ask what it was that made you unhappy about the prescription?

My reason for asking is that I have been using high street cheapskate chain opticians for years and my prescription has changed just once in around 18 years, and just in the past year or so I've been feeling a bit more eyestrain and have wondered about a second opinion. But it's taken a while for me to get around to wondering about that. You seem to be describing unhappiness with the first ever prescription you got, so I am curious as to why. No worries if you prefer not to give details though.

In reply to Blue Straggler:

My eyesight is fine for most things just had problems with very fine print (recent work requirement). The high st prescription provided only a very marginal improvement.

In reply to tom_in_edinburgh:

I haven't got time to go through all the points, but just to give an example of what an optometrist does in pretty standard eye test that technology is nowhere near replacing:

 - patient comes in saying their vision is getting worse and they're no longer happy driving

 - optom asks some questions and measures their vision to see how bad the problem is

 - optom does an examination and finds that there's a change in the spectacle prescription, which has been caused by cataracts; there's also the start of macular degeneration which has progressed a bit since the last test

 - optom advises patient on their options (e.g. whether to go for cataract surgery) and helps them come to an informed decision

 - optom then helps implements that decision (e.g. refers the patient for surgery) and gives some advice on how they can keep driving, etc.

You're right that certain elements of the sight test can be automated, but at present this will only work for a fairly small subset of patients (basically healthy myopic working age people). It'll certainly be interesting to see how things change in 50 years, but not so much in 20.

In reply to Stichtplate:

Thanks, OK so there was a time gap between the high street prescription and the independent one, during which you'd actually tried prescribed glasses or contacts, and been disappointed with the improvement....is that a correct assumption?

I was only asking the question in order to clarify that, as a first-time testee, you hadn't looked at the numbers and immediately said "this ain't right"!

 

In reply to Blue Straggler:

I was only looking for a fairly marginal improvement, what I got was an improvement so marginal as to be almost imperceptible. The new prescription was much better.

More worrying was that the high st company hadn’t spotted the developing cataract which was probably the result of a decades old blow to the eye.

In reply to Stichtplate:

Thanks. I'm going to look into this, I've been putting increased eye strain down to workload etc but it could equally be a standard muscular deterioration that has not been picked up by Sp*cs*vers. To be honest I always felt that their test seem quite thorough and adequate at least for a relatively mild prescription like mine, and I don't have any real practical problems with sight (a large part of my job is spotting small defects in monochromatic images on VDUs and I'm pretty good at it which is why I wonder if I have nothing worse that eye strain sometimes)

But for the price of a single consultation it's got to be a win-win - either be assured that the current prescription is fine, or learn that it's not. 

In reply to alx:

I understand all that and I don't think things will change for actual safety-critical equipment used in hospitals.   However I don't think this system is going to survive for spectacles and hearing aids.  Hearing aids are too close to accepted consumer electronics like Bluetooth headsets and spectacles are just bits of carefully polished glass.    There's no mystery or fear like there would be with surgical equipment.   

It's another example of free market and new technology against regulations and professional guilds:  "All of this has happened before and will happen again." 

 

 

 

In reply to Jon Stewart:

That isn't how disruptive technology competes.   A cellphone running the Uber app couldn't 'pass the knowledge' and get a taxi driving licence so Uber don't even try to play by the old rules.  Similarly an aggressive and well funded company who wants to grab the market for glasses isn't going to try and deliver 'the sight test'  they're going to define their own service and sell it straight to the consumer.   

The pitch will be you can use their app to get a prescription in 5 minutes without leaving your house and get a nice pair of glasses sent out next day delivery and the whole thing costs far less and is far more pleasant than a trip to an optician.  

 

In reply to Jon Greengrass:

Don't get this buying glasses online thing. Last couple of times I bought frames a member of staff spend a good half hour with me trying on different styles to find what suited and fitted my face, including above on lense shape to minimise neck strain when using them at a computer.

Yes maybe I pay more for an NHS eyetest, but I get quality service and a full eye healthcheck. Also they don't push expensive frames on me every time and are more than happy to re lense the frames they do sell for a fraction of the cost of new glasses. 

In reply to tom_in_edinburgh:

As mentioned above. The processing power is not the problem, its the tolerances of the camera lens and the ability of the user to keep their device up to date and clean. 

There are already too many people driving with uncorrected eyesight. The ramifications of people checking their eyesight themselves or relying on an app are far reaching. 

What happens when someone tries to sue the creators of the app for giving them a defective prescription. 

Ultimately the opticians make money for selling glasses, not for giving out prescriptions. 

In reply to Jenny C:

I tried online once for some safety glasses, the try out service was great they sent me a bunch of sample frames to try on to choose the ones that fitted best, postage was only £3.75 with free returns, less than the cost of driving in and parking in town.  Unfortunately with my very high prescription ( -10 ) the PD measurement is critical and despite sending in a photo it was slightly off, which left me feeling slightly sea sick when wearing them. I've since bought another pair of safety glasses from a high street optician, the frames aren't as nice because they were unwilling to order in the frames I wanted but the optics are perfect thanks to having been fitted in person.

In reply to Big Ger:

What!?

In reply to wintertree:

Software as a Device is now part of the new medical device regulation (MDR) which specifically addresses the analytical side of devices. If your software is part of an actual device (like a handheld scanner) you register the whole package under the part that has the highest risk classification. This being said many people are developing either a;  “black box” which is software coded into a piece of kit that plugs into an existing medical imaging technology that then provides the radiographer or equivalent with a piece of software to do the extra work; or a cloud base system in which the image is analysed remotely by trained specialists in the new software. The former approach gives the radiographer the help and can be done right there and then, the latter takes up to 24hrs and the radiographer just gets a value.

In both instances it is still the “expert medical opinion” of the attending consultant or physician which combines all the other factors as to driving treatment. If the technology drove treatment or intervention without this safety net then it would be the highest risk category like a pacemaker.

Up until last year there was no approved medical imaging specialist technology using virtual or computational software. HeartFlow obtained the first ever approval for their technology from the FDA which was considered “de novo”, completely new device of its type. As it was de novo tech, the regulators had to co-develop the requirements, as such HeartFlow bore the brunt of all possible questions and costs whilst the FDA figured things out.

Since approval and forming the regulatory requirements a significant amount of new technology in this area is coming forward utilising the new roadmap.

In reply to tom_in_edinburgh:

Hi Tom,

Its based on risk and not perception of fear, and based on modifying and existing state of a persons health.

For instance if you needed to design a wall bracket for a large TV to hang over a bed for purchase by the general public then you would just go the normal route. If you specifically built the wall bracket for the intended purpose of hanging a large TV over a very sick persons bed then it would need to go down the medical device route (most likely class 1 self certification) purely based on the risk and it’s management. A healthy person could most likely either get out of the way or suffer recoverable injuries if th tv fell on them, a sick person may suffer badly and possibly may not be able to shift the tv off them compounding the problem into a more serious issue.

The same applies for hearing aids, if the person is hard of hearing or has loss of sensation in the ear then lithium battery leaking into their ear may potentially not be identied until too late.

In reply to alx:

I think we are talking past each other.  You are talking about the medical device regulations and I accept that's what they are.  My view is it doesn't matter because when market forces collide with regulations and the end customer needs to save money the regulations lose.   If someone offers consumers glasses that an optician would sell for £800 for £200 or a better hearing aid than the professional would prescribe for 1/3 the money then people will buy it.  Customers will treat these things as consumer products and buy on price/performance no matter what the rules say.  

Of course the established suppliers won't like the competition and expect the regulators to enforce the rules.  Some people will get prosecuted but if there's a massive price differential it will be futile.   For example Uber just ignores any rules about taxi's that it doesn't like.  It has a ton of money and doesn't mind spending some of it on lawyers and lobbyists.   Most cities can't afford to fight it.   A rich US startup with political connections can use international trade treaties to thwart national regulations that frustrate it's business model: Qualcomm got a lot of help with its entry into the cellphone chipset business.  Chinese companies sell counterfeit products through websites like AliBaba which are outside the control of regulators in the EU or US: they can take the same approach with cheap glasses or hearing aids.  There's no effective way to stop them: it's pointless to carry out any kind of complex enforcement against a company in China that will just change its name and reappear.  They can ship unregulated 'reading glasses' which just happen to be real glasses or unregulated 'bluetooth headsets' that just happen to work as hearing aids.

 

In reply to krikoman:

"I don't turn them on, it runs the battery down.."

youtube.com/watch?v=tcliR8kAbzc&

In reply to tom_in_edinburgh:

The trouble they'll have is that the market is strictly regulated - as things stand, you can't legally sell spectacles unless there is a prescription signed by a GOC registered optometrist. There are good reasons for this, people's eyesight is important so selling people dodgy glasses to dodgy prescriptions is something we, as a society, do not want and our elected representatives have legislated against it.

I don't see that regulatory system changing any time soon - the arguments that it protects us from harm that an open deregulated market would cause are very strong.

While you don't need an optom to get a spectacle prescription, you do need more than an app. The best way to automate it would be a machine that you look into with 3 dials that you turn while looking at different targets. Many people could easily get their best possible prescription from such a self-operated phoropter system like this, but it's not possible with software alone, it needs moving parts.

The technology you predict is not going to happen any time soon: it requires both expensive hardware and a change in the law.

 

In reply to alx:

Very interesting; thanks for taking the time to write and post that.

In reply to daWalt:

Crap trolling effort- not funny...........D minus.........

In reply to Jon Greengrass:

Is it discrimination if people born with a genetic defect are charged to have their teeth fixed so they can eat and therefore not die?

As someone who is both four-eyed and has weak teeth, there are worse problems in the world.  Chain opticians do free eye tests and spectacles online (better looking than NHS ones) cost next to nothing and last ages.

So does it bother me?  No, not really.

In reply to Jon Stewart:

I see spectacles on sale all over the place without prescription. Supermarkets, filling stations etc. 

In reply to aln:

Yes, ready made reading specs are legal, but their power and what they can be used for is regulated. Of course you can't stop someone driving in a pair of reading specs, but they must be:

For a person not under the age of 16
Of spectacles with two single-vision lenses of the same positive spherical power not exceeding 4 dioptres
Where the sale is wholly for the purpose of correcting, remedying or relieving presbyopia.

Above I was simplifying by assuming we were talking about glasses for people who need them all the time, or for driving/TV etc.

In reply to Jon Stewart:

I see your point and no doubt these two factors i.e. the regulations surrounding prescriptions and the difficulty of doing the measurements with just an app are why this has not happened already.

However, my experience in the tech industry is that:

a. New technology and free trade are unstoppable forces. 

b. Electronics and software displaces mechanical devices and it can get really innovative to do it.  Just because it seems impossible for some particular function to be done without expensive mechanics doesn't mean it is impossible.

c. These things happen a lot faster than you think.

I remember the same kind of argument about Netflix against the TV/movie industry, e-books against paper books and before that about Amazon vs bricks and mortar.    There's been a ton of industries with their regulations and anti-competitive practices and established structures and reasons why those structures won't change.  In the end e-commerce and tech just steamrollers them because people want better for less money and tech companies either spend money to get laws changed or use the internet to operate from where they don't apply.    

In reply to Jon Stewart:

So you can legally sell spectacles without  a prescription. 

In reply to tom_in_edinburgh:

To go back to your uber analogy: the technology can process information differently to change the process by which you get a cab. But you still can't get from A to B with an app, you need a vehicle.

Similarly, you can't do a refraction without a lot of lenses. That's what a refraction is. What you're suggesting is a bit like an app that makes tea or does the dishes. You could use software to help get these things done with less human intervention, but there's got to be some physical interaction between the boiling water and the tea leaves in order for a cup of tea to exist. Similarly, light needs to have passed through a lens into the eye to determine the power that will focus the rays onto the retina.

I very much disagree with the idea that "free trade is an unstoppable force". It's stopped all the bloody time. We don't have free trade in healthcare, and for very good reasons. There is no appetite to change that, and companies that sell unregulated medications on the cheap are marginalised as dodgy grey-market internet wheeler-dealers. They're not popular because they're not trusted and they don't comply with the law. Anyone can self-diagnose and get meds off the internet if they want to, but this "disruptive technology" is just a dodgy grey-market area of the internet that has little impact on how we do things.

In reply to aln:

Do you think you've scored a point there?

In reply to Jon Stewart:

No. 

In reply to krikoman:

A cool guy who came to service my boiler told me he had hearing aids in, and then said 'What?' very shortly after.  

He reminded me of the two grey haired guys who repair the roundabout in The Magic Roundabout book The Adventure Of Dougal, pointing out things to me as if I knew as much as him, and laughing knowingly about whatever it was...  

In reply to Jon Stewart:

Yes, Uber need a vehicle.  But it isn't a Black Cab it's a Toyota.  There were rules that said taxi drivers needed to buy Black Cabs with specific features like disabled access.   I got a lecture about this once from a cab driver: in Edinburgh there was just one dealer that sold 'legal' taxis so they had a monopoly and the prices were far higher than for a normal car.  That got reflected in the average age of the taxis in use.  Uber got around the rule about what vehicles were allowed and ended up with cheaper and newer vehicles.

But you can play image processing games.  For example 'digital zoom' where instead of having a movable lens you have a higher resolution sensor and throw away some of the image.  You can replace a good lens with a poor lens and fancy software to make corrections.  You can use multiple small/cheap cameras and combine the images to make a better quality single image.

When you have a 500 million unit a year market like cellphones you can also think about using exotic materials and really novel technologies:

http://appleinsider.com/articles/14/06/19/apple-invents-iphone-camera-lens-...

I'm not saying that particular invention solves the problem.  I'm saying I think the cellphone industry are going to be able to keep making their cameras better and they've already come an awful long way without bulky lenses.    There's also the 'get out of jail' option of a clip-on device hat fits over the camera.

I think psychologically drugs are different from glasses and hearing aids.  Nobody is scared of being injured by glasses and hearing aids the way they are scared of getting poisoned by dodgy drugs.

The situation in the UK is also unusual because the NHS provides things for free.  In countries where people have to pay for medicine there's a lot more incentive to buy it in the grey market.  US citizens are much more likely to get drugs from Canada or Mexico to save money.

Speaking personally, if I was asked for £800 for a pair of glasses (like my wife was) I would buy them online in a second and I wouldn't care if the website was in the US or China.   In fact if I could get a prescription from an app and buy online I'd do that in preference to going to an optician even if the price was the same because it's more convenient and avoids the whole medical consultation aspect. 

 

In reply to tom_in_edinburgh:

I agree with you that you can work wonders with crap optics and “computational photography” but you can’t work miracles.

The lenses in a computational camera do something very different (forming images as input to the algorithms) to the lenses used in measuring a patient’s proscription (modifying the phase of the light reaching the patient’s eyes to alter the images formed on their retinas).  

A display that controls both the phase and intensity of light would be required for a phone to refract in this way and these are barely even a thing in research labs let alone on mobile devices.  There’re all manner of very crude approximations out there (stereoscopic displays etc) but these aren’t good enough to use for measuring proscriptions.

You can actually generate blurred images, the human perception of which helps to estimate refractive error but I don’t think this could easily be productionised into a robust app-based measuring system.

In reply to Jon Greengrass:

Anyone know where you can buy hearing aids at sub- boots or sub-specsavers prices safely?

In reply to wintertree:

Yeah, after writing the last post I thought of doing it that way.  I don't think it needs to be easy, there's enough money at stake to put a lot of effort into it.

It would also be fun to write code that distorted the display on a monitor  or iPad so that the distortion cancelled out the distortion of my crappy eyes and resulted in me perceiving a crisp image.   Basically do the maths to calculate what a camera with perfect optics would see if it looked at the screen through the corrective lenses you need then put that pre-distorted image on the screen so you see what you would see if you had your glasses on without wearing them.

Kind of like Billy Connolly's 'prescription windscreen' for cars.

 

In reply to tom_in_edinburgh:

Unfortunately the physics just doesn’t work like that.  Image formation is not a reversible process in humans or CCD/CMOS sensors as the phase component of the complex EM field is discarded.  This is compounded by refractive errors modulating only the field phase (in what amounts to a Fourier transform of what the eye sees), and displays modulating only field intensity.

Another way to think of it is that different aberrations (+1 or -1 diopter of focus for example) produce the exact same blurred image on the eye - information is lost or there is degeneracy - so there isn’t and can’t be a pre-computable solution.

You can design fonts and images that remain clear under a specific person’s eye distortions - but this isn’t pre-compensation so much as making sure features that will be most affected by their distortion are not present.

In reply to wintertree:

I'm no expert on this stuff, you've obviously got a lot more background on optics than I do!

The situation with trying to 'sharpen' someone with bad eyesight's view of the image on a computer monitor by pre-distorting the screen image appears to me to be simpler than the general problem of correcting someone's view of the 3D world as they move about in it.  The non-distorted computer screen is already a flat 2D image and the viewer's head is at a fairly fixed position relative to it.  I can see that a 2D camera image of the 3D real world inherently loses information but I'm not sure that a 2D camera image of the 2D image on a screen does and I'm not sure that in this specific setup more parameters won't be able to be pre-computed than in the general case.

I see that one of the commercial systems which uses a cellphone for sight tests mounts the phone in a VR headset like setup so the phone screen is physically pretty much where the lenses in the glasses would be.  Maybe this is addressing the problem.

In reply to tom_in_edinburgh:

It really does loose information though. It’s hard to understand intuitively because it’s not information you “see”. 

If you get a chance to cycloplege your eye to paralyse the focous muscles, look at your phone screen somewhere it’s in focous.  Display a small bright dot. Then put a +4 diopter trial lens infringement of the eye - you see a blur.  Now put a -4 diopter trial lens in - you see an identical blur.  In this case information on the sign of the focous has been lost.  To visualise consider two light rays crossing at a focal point.  Intersecting two planes equidistant either side of the crossing, they form identical intensity images on the two planes.  You can only tell which side of the focous they are by the angles they make with the planes - their phase data.  This is discarded by eyes and uncontrolled by phone displays and that’s what prevents you pre-correcting for focous with a display.

SLRs, microscopes and binoculours use movable lenses to compensate for viewer focal errors...  electronic view finder cameras use a lens in front of the EVF...

When considering an intensity only source image, a purely refractive error in the lens, and a pure intensity sensor, one can treat the refractive error as having the effect of convolving the “perfect” image with some blur function determined by the error.  There is no image in general that can be computed that is sharper *after* convolution with the blur.  The best computational imaging can to is to attempt to estimate the blur and perform a post processing deconvolution - this is a powerful technique in microscopy and astronomy - but it only works with detector data, not source images, due to the information loss.

In reply to wintertree:

Thanks for taking the time to reply.  Like you say it is hard to understand this intuitively.

I understand your point about not being able to determine the sign of the focus just from an image in that set up.  But it seems like in the situation with a person sitting in front of a monitor they want to use without wearing glasses there is a lot of extra information available beyond an image.  For example I can measure the distance between the user's eyes and the screen.  Also I can do trial and error and incrementally change the transform I'm applying until the user tells me it is sharp or ask the user to move relative to the screen and tell the software how things change.   In the situation where the user has glasses but just doesn't want to wear them when working on their computer the software could even use their prescription as an input.

If you can model the error in the users eyes as a convolution with a blur function then why can't you calculate the inverse of that convolution then apply it in advance to the image on the screen so that you cancel out the error?  

In reply to tom_in_edinburgh:

Everything in your reasoning in your post is spot on about trial and error and changes and what is known, except...

Because light is a complex field with phase and amplitude, and a display only modulates amplitude.  Sure, you can compute the image you need to display but it will have negative - or complex - values in it.  They translate to changing the phase of the light.  With an amplitude only display you can’t do it.  

Think of it another way - what image containing only positive values (ie magnitude of the light field) when convolved with a Gaussian blur gives a sharp point?  Whatever input image you convolve will be less sharp by the blur.

This is why proper holographic displays aren’t really a thing yet - devices that can arbitrarily modulate optical phase with any spatial resolution are (a) a bit shite and (b) very expensive.

This is also part of why “predator” style camouflage isn’t a thing.  

It’s easily done at radio frequencies where we can build electronics that are higher frequency than the EM waves but much harder with light, where our electronics is orders of magnitude slower.

This is why in astronomy radio interferometery can be done in post processing (amplitude and phase are recorded) but optical interferometery has to be done optically - limiting optical interferometers to a few hundred meters vs planetary sized radio interferometers.

In reply to wintertree:

Got it now!   Thanks!

I wonder if a custom 3d printed transparent plastic 'screen protector' could correct an iPad or computer screen for one users vision.   If it has variable thickness it could modulate the phase of the light.  

In reply to Big Ger:

Try a listener. Mine's been a revalation.

In reply to Dom Connaway:

Link mate?

In reply to Lusk:

Thankfully I don't smoke.

In reply to tom_in_edinburgh:

Very interesting way to think of the problem from a completely different angle. I'm not confident that dioptric blur is just a result of phase change - a lens bends the angle of a 'ray' because the speed of light is different in the refracting material compared to air. So yes, there is phase change once the light has passed through different thickness of lens, but it's the change in angle of the 'ray' which determines whether an image is focused (e.g. on the retina) or whether there is blur. I'm not convinced changing the phase in the way you describe would result in a focused image on the retina - maybe wintertree can help, I'm not big into physical optics (we do geometric optics and don't go into a great deal of detail about the physics of it).

 

 

In reply to Jon Stewart:

You’re spot on for geometric optics.  The direction of a geometric ray is perpendicular to the wavefront of a physical optics wave - with the relative phase of different bits of the wave giving the angle of the wavefront.  So for a simple wave like light coming from a point source there is a definite relationship between physical optics phase and geometric optics ray angles.

I’ve been thinking about Tom’s suggestion - it’s clear to me it won’t work because the screen is in a conjugate plane to the person’s aberrations (light from a single point on the screen goes through every part of their eyes lens, as does light from every other part of the screen - so there isn’t a physical mapping that puts the correction needed to match the distortion in the plane of their eye lens to the plane of the screen).  

But I kept quiet as I’m thinking - it’s also obvious to me that full control of phase and amplitude on the screen allows for an arbitrary wave to be generated allowing an image to be displayed perfectly through an imperfect eye lens.  

The answer I believe to resolving these incompatible statements is that (1) the correct phase must modifying element is totally specific to an individual image, and (2) you probably need more display resolution.  For (1) a trivial example are two different sparse patterns of dots, each with its own matching array of lenses to compensate for the eye.  Clearly one pattern of dots won’t work with lenses for another pattern.

 

In reply to wintertree:

The key word missing from that is “not”.  The screen is not in a conjugate plane to the lens...  oh to be able to spot mistakes when I make them.  

In reply to wintertree:

Have you considered glasses?

In reply to wintertree:

I don't quite understand why the filter on the screen is not in a conjugate plane to the persons eye but glasses on their face would be.  

Your comment about needing more display resolution interests me because display resolution is one thing we have got in abundance on something like an iPad Retina display and will only improve.   When someone with crap eyesight is reading on an iPad or computer they often push up the font size so they are reading huge letters and most of the resolution is wasted.    Also, a useful product doesn't need to be as good as someone with perfect vision looking at a high res display.  The state of play it is competing with is much worse than that.   Even with multiple pairs of glasses my wife doesn't have a pair which works perfectly at the distance of a computer screen: the goal for a correction product could be less blurred rather than not blurred at all.

So I think it may not be a problem to trade quite a lot of resolution for the ability to correct vision: the user would still come out ahead.  For example we could take a 2x2 array of pixels on the iPad and treat that 2x2 block as a single pixel of a lower resolution display.   Above the screen the 3D printed 'screen protector' could put a different depth of optical material above each of the 4 pixels.  Then during operation the screen driver software selects one of the 4 pixels in the block to illuminate.   So the phase of the light can be modified dynamically during operation to one of 4 predefined values.  An added benefit is this printed pattern isn't specific to one particular user.

 

In reply to tom_in_edinburgh:

Glases are close to the lens of the eye and far from the screen.  Light from one point on the screen goes through all parts of the lens of the eye, and all parts of the corresponding glasses lens.  Light from another part of the screen goes through all points of the eye and lens.  You can move the phase change of the glasses to one point of the screen or another but not both - at least not in a general sense - you can for well separated sparse points.    

Yup - that works - but there are limits/problems.

 

(1)  You need the phase of the light to be controlled before you can meaningfully modulate it - which means the backlight would in practice have to be a laser somehow delivered to all pixels, not incoherent LEDs.  You don’t just need to change the phase but to control it deterministically.  Image formation occours through constructive and destructive interference.

Light from a single point is in phase regardless of laser or LED or thermal lamp, so it’s not normally a problem.  But what you propose needs to make light from separate points in space appear to come from one point - and to do this, modifying the angle of the geometric rays is not sufficient, the absolute phase of all rays must match when they meet on the image sensor, for them to interfere and produce a bright spot.  This can only happen if you know or control the phase of all the separate points emitting light.

 

(2) The obvious structure for such a “light field display” is that each output pixel is a single micro lens with a higher resolution pixelated display behind it as you describe.  The problem is you need microlenses the size of individual pixels - and this is so small that the diffraction limit of the lenses renders them almost useless.  The inverse of this is a light field camera and is easier to realise - eg Lytro.

 

Production of a high quality video rate lightfield display is something of a holy grail.  There are rumours of gigapixel phase modulators loitering in secret basement labs of a car manufacturer and a couple of defence firms, and a fair few people have multi faceted mirrors spinning ludicrously fast, but at a guess I’d say we are 10-30 years away from a consumer grade system.  

In reply to wintertree:

Isn't there any locality?  It seems like the blur we are trying to correct is only across a relatively small distance so isn't it enough that the light within a certain range of pixels is pretty much in phase as opposed to the much tougher criterion that light from the whole display is exactly in phase.   I don't see the mechanism which causes a pixel at the bottom right corner of a 20 inch monitor to affect blurring of a letter displayed on the top left.   

It's all sounding really difficult and maybe not possible with an aftermarket 'screen protector'.  Which is probably why you can't buy one already

How about with a projective display?  That could have a single laser light source much more easily, the 'screen' the image was formed on could have the optical structure and because it's projective the size of the screen could be quite large so the physical size of the lenses could be larger than if they were mounted directly above pixels (or maybe no screen is needed and something in the projector could select the phase for a particular pixel).

It seems like the people playing in this space are mostly trying to build holographic displays and I'm still wondering if the special case of correcting for vision defects of someone reading an iPad or computer monitor where the customer is willing to tolerate something far removed from perfection may allow for some shortcuts compared to solving the difficult general problem.