Eye scans detect signs of Parkinson’s disease up to seven years before diagnosis
moorfields.nhs.uk
moorfields.nhs.uk
There is significant overlap in the distributions of prodromal and control Parkinson’s patients even for the measure with the strongest signal. While some values for that measure seem to be almost conclusive markets of Parkinson’s, there are ranges in which the overlap is enough to make it not useful as a sole measure without having a high false positive rate.
Seems promising at catching a decent proportion of Parkinson’s super early, but I don’t think by itself this could yield both a good false positive and false negative rate. Keep in mind this is an ebullient press release summarizing a more conservative publication
What does early intervention look like for Parkinson’s (besides lifestyle changes)?
As far as I know the treatment for Parkinson’s is basically lifelong dopamine pills, which of course you build a tolerance to and which lose effectiveness over time, requiring ever higher doses. Meaning once you start on them, the ticking countdown of your brain’s lifespan starts as well.
"There is evidence to support the use of DBS in less advanced PD and that it may be appropriate for earlier stages of the disease than for which it is currently used. The improving short and long-term safety profile of DBS makes early application a realistic possibility. "
Lifestyle changes, on the other hand, mostly meaning exercise, significantly and reliably slow the onset and development of symptoms.
We have talked to many different Parkinson's experts around the world, and I think at this point the only thing she seems to think really makes a difference is lots of exercise. She completed halting any evidence of the progression of the disease for about 5 years with exercise, but it will not last forever.
So completely serious, if you get flagged for Parkinsons, and start getting in amazing shape, you have a good chance of really helping yourself get more good years.
Fingers crossed for ongoing research to finally find some way to stop or reverse the damage in another way! (not medical advice)
In terms of reversing the damage, look into photobiomodulation (PBM) aka red light therapy. In "Improvements in clinical signs of Parkinson's disease", Liebert et al in 2021, she shows improvement in all symptoms including cognition, which is one of preciously few such results I've found in my extensive search of the literature[1]. Caveats are that this was a proof-of-concept study with n=6 only and that the red light helmet is somewhat expensive if you want to try it. There's a Canadian company that makes one for above $2000 (modern, very sci-fi thing), and a more hackerish version from an Australian company for ~$700 (pairs of diodes on aluminum bands you have to finish assembling yourself).
[1]: At least which are actionable for the public. There are tons of trials but good luck getting in early unless you can donate a new library to the university! Gene therapy, stem cell therapy, GDNF, drugs that target alpha-synuclein are all promising but not yet accessible. PBM is something you can do today and since mitochondrial dysfunction is a leading hypothesis in the pathogenesis of PD, the treatment fits.
Looking around a bit it seems that individual GCIPL thickness varies quite a bit even in healthy young adults, with the range exceeding the effect detected in the study. Practically this might mean you'd have to have lifelong GCIPL monitoring to detect the early onset of disease (and GCIPL thinning doesn't seem sufficient to diagnose Parkinson's, it occurs in other diseases like Alzheimer's and perhaps even as a natural outcome of aging).
https://techcrunch.com/2018/02/17/righteyes-portable-eye-tra...
https://techcrunch.com/2022/05/24/neuralight-aims-to-track-a...
Probably there are more by now. As I recall this data is great for essentially saying "you should go get a real test." No one is diagnosing - just saying that a certain, say, eye movement slowness is correlated and you can detect it much earlier than a hand tremor or other more prominent symptoms.
It’s not just movement. Retinal tissue develops from the same embryonic tissue as the brain. They probably looked at structure, vasculature.
The best anecdote in this space- nobody with congenital blindness has developed schizophrenia.
https://www.lesswrong.com/posts/z9Syf3pGffpvHwfr4/i-m-mildly...
That is a much more risky idea in the first place, and your link provides quite a few such anecdotes.
What’s the rationale for why this might happen?
But to me, very low incidence rate for both seems like a sufficient rationale. Given that there's both viral and genetic causes for congenital blindness, it just doesn't make sense to imply the protection is genetic: even if true, there would still be cases of schizophrenia among the people that were blinded by a viral infection in-utero.
Parkinsons affects motor control, not mental functioning. The point still stands -- if I got a diagnosis I would be less likely to embark on learning to play the piano, and I'd start educating myself on the disease.
EDIT: I'll correct myself. It primarily affects motor control, but late in the disease it can affect cognition.
Perhaps counterintuitively, doing something like learning to play the piano could potentially slow the progression of fine motor skill impairment. Same with things like Tai Chi, dancing, exercise in general...
But hopefully if you could get a Parkinson's diagnosis seven years prior to symptoms, you'd be able to save some money to keep from becoming a burden on your family later on.
And yes I'm well aware this applies inductively to all of life, it's just somehow better if you don't know the value of N.
I tried that and found it absolutely depressing and futile. Quick experiences never did anything for me or my family. But just doing what I like/want to do only (which not everyone can; some people seem to like/want things that cost a lot of money) is what I should've done for the past 40, not only the past 20 years is what I mean. Maybe dying next year is stressful for most, so make it 5 years; is that better? Or 7 as the article says?
But euthanasia as a non traumatic form of death unlike suicide, is still prohibitively expensive.
Crowdfunding steals your privacy.
Science steals your privacy, and can keep you alive for longer.
So are you not over rating life somewhat?
Parkinson's is funny in that sometimes what initially presents as Parkinsonian symptoms could develop into something like MSA over a span of 5 or more years. If this person were in a clinical trial for pure Parkinson's, it would sort of mess things up...
There are some candidates advertised as promising but I'm not sure how much stock I have in this. This isn't my area of expertise but disease modifying therapies for neurodegenerative diseases have been touted as being "around the corner" since I was a first year medical student.
I expect people will look into why eye scans work and how to make them work better.
I also expect them to use this to catch potential Parkinson's patients early and see if they can find out more about early stages of the disease. Maybe they'll find something that works even earlier than eye scans.
It's also a little amusing to take the idea to its logical extreme: What if it becomes a societal norm to know? Would public venues allow soon-to-be dementia patients line cutting priority etc? Think amusement parks, transit reservations, hotel booking, bars, movie theaters -- all of it. There could be a whole industry built overnight of catering to this specific demographic.
It's not fun, but being able to do long-term planning around challenging medical conditions is without a doubt better than not knowing. This includes everything from financial planning to finding a house that is ideal for your condition (no stairs, single level, etc.)
Ideally we'll have a better understanding of disease-modifying drugs in the near future that can slow progression of Parkinson's. If we get confirmation that drugs like Exenatide actually do slow the progression of Parkinson's (trials ongoing) then you'd definitely want to know as early as possible.
Do we do this for people with cancer?
This situation is unique because of the time span. I'm sure there would be an increased sustained and consistent demand for nostalgia-focused apps. And probably for many other products I haven't considered.
Pre-existing conditions can't be used to deny health insurance coverage in the United States.
This has been true since 2014.
Please don't perpetuate the myth that people should avoid screenings for fear of losing health coverage.
How much opportunity is drained form everyone who relies on insulin to survive every month?
Most developed countries (and many less developed countries) have some government coverage for cancer or dementia.
In New Zealand we can also pay for premium private insurance - although I don't think that helps much for chronic conditions.
Dementia will often still drain your bank account in New Zealand because the government here will take all your equity to pay for nursing home fees. But you do get looked after a bit when you have nothing, so there is that. You can avoid having all your equity taken but it does take some planning, some costs and some effort (e.g. give equity to children more than 7 years before you lose your marbles, e.g. use a family trust, e.g. if married it is very important to ensure your will limits equity damage to one if the other falls ill).
Insurance is a business built on exploiting information asymmetries to make money from groups who they try to show empirically can't actually cost more than they pay. If they can find a way to exploit extra information in order to show you're in a riskier group and hence exclude you from coverage, they will. They have absolutely no reason not to.
Insurance companies and you absolutely have an oppositional relationship; they love when we collectively ignore this.
Something to think about.
For two or three generations, that is, with very few notable exceptions.
So now computers use AI to do something, interesting...
I always liked how it's always "AI did X" or "AI discovered Y", but never a "crane built a building".
E.g.1: Galileo saw Jupiter, not The telescope saw Jupiter.
E.g.2: Joe built the wall, not The crane built the wall.
In this case, AI is the thinker, not a human, hence the phrasing.
A human creating the AI does not remove the attribution either, just as we never say Joe's mother built the wall just because Joe was created by his mother.
It is a tool.
the average code monkey cranking out web microservices is not exactly plato either here, after all. look at the existing codebase, try to mimic the style, can you have the PR by friday?
As an AI specialist, I strongly object that statement. There's usually very little surprise in what "AI has discovered", usually we spend months of hard work to make it discover that one thing. But all the credit goes to "AI" :)
PS: I'm yet to see an AI (in any area) that's the thinker.
My father had it, and he's a definite data point for it being caused by trichloroethylene, as an engineer that worked with metal and degreasers, he was exposed to it for years.
[1] https://www.science.org/content/article/widely-used-chemical...
How long before we can leverage a smart phone to become the tricorder Star Trek had always promised?
Apple Watch looks to be promising, and I imagine a world where humans won’t have to visit a doctor for a diagnosis, but rather to confirm, and walk a patient through a successful treatment plan
> High-resolution images of the retina are now a routine part of eye care – in particular, a type of 3D scan known as ‘optical coherence tomography’ (OCT), which is widely used in eye clinics and high-street opticians. In less than a minute, an OCT scan produces a cross-section of the retina (the back of the eye) in incredible detail – down to a thousandth of a millimetre.
[1] https://www.cureus.com/articles/106024-effective-low-cost-op...
I think the real issue with OCT cost is machine utilization. An OCT machine is cheap as medical devices go, but, in the usual US model, it gets extremely low utilization. It takes a minute or so to get a scan, it’s completely painless and requires neither dilation or any particular expertise, but the machines only get used a couple times a day.
A venue like Costco or a comparably busy place in a different country could probably offer walk-up OCT scanning for a couple dollars or maybe even free and encourage people to do it once per year. If the FDA played along, software could likely analyze the images and detect the progression of quite a few diseases, several of which would be treatable.
(A brand new machine seems to cost something like $20k. If it gets used 60 times per day (credible for a large scale user but not for a small shop), $1 of revenue per scan gives an adequate payback.)
the star trek tricorder had attachments, for example the medical tricorder had an additional module and a detachable hand scanner.
https://memory-alpha.fandom.com/wiki/Medical_tricorder#24th_...
the cellphone as the basis for the processing of various sensors, delivering software and connectivity, is essentially similar to this idea. Like when you have a thermal camera that connects to your cellphone.
In both cases, they're correct - since no intervention is risk free, even if you have a 0% false positive rate in the sense of "I found a mass in your arm", for example, you might still have a worse outcome rate if the rate of "mass in your arm causes harm" is lower than "complications from removing mass in your arm".
That's not to say that their suggestions are right and yours are wrong, just that they're right about the reasoning but tend to dismiss concerns that they didn't think of and justify ones they did, and we need a more complicated basis for justifying this than "I think".
To be fair, with complex diseases, the experts don't even understand them fully...
Trying to explain something super complex with incomplete knowledge to most humans, who often don't have even a basic understanding of how biological systems work, seems like a bad idea...
I suppose if you tried to get certain types of long-term care insurance or life insurance and failed to disclose this you'd be at risk of losing the policy.
But nobody should be avoiding screenings or care for fear of losing health coverage for pre-existing conditions. That stopped being true almost a decade ago.
They don't explain this to you when you get the check.
Other things such as life insurance is a very different story.
No denying you or charging you more because of a pre-existing condition.
I was reading this paper yesterday: https://www.frontiersin.org/articles/10.3389/fchem.2023.1106... . Wouldn't it be good if we could scale those tools to human organisms and to individual patients? Use ML to compensate for the fact that our CPUs and GPUs can't really simulate something that big? What if we could see into the private lives of individual diseases with more resolution than the world's best spying agencies? [^1]
[^1] I dream I could work in something like that, send me an email if you are hiring.
Matthew 6:22
I find it oddly reductive, unsubstantiated, and meaningless.
Perhaps I'm not the intended reader :P
[0] http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S...!
Not to nitpick, but it's my understanding that we all lose dopaminergic neurons in the Substantia Nigra as we get older. Some people lose them a a lot faster, and they develop what we call "Parkinsons". There is also presence of Alpha Synuclein in the dysfunctional dopaminergic neurons, which are not generally present in the "non-parkinsons" population...
I am skeptical for three reasons.
First, the article goes on about the eyes being “windows” to the body and introduces us to the new “science” of “oculomics”. That might be journalist license but it smells of pseudoscience; phrenology anyone?
Second, the idea that any image of an eye can be used to diagnose lots of different neurological conditions like schizophrenia again smells like pseudoscience; reminds me of excessive claims of what chiropractic can do.
Finally, the study seems to be “let’s dump medical records and retina images into a neural network and see what correlates”. I didn’t read the actual study but I would be extremely concerned about accuracy rates.
So I remain cautiously optimistic but still very skeptical; I would be happy to be wrong in this case.
I understand why this seems like pseudoscience, but I'd like to explain the assumption better. As the article states,
> a scan of the retina is the only non-intrusive way to view layers of cells below the skin’s surface.
As an extension, you could also make the case for being able to view the nervous system via such imagery. As stated in [1],
> Despite its peripheral location, the retina or neural portion of the eye, is actually part of the central nervous system.
The theory is that such a non-invasive of visualizing crucial parts of the inner body opens up a "window" into noticing such biomarkers well before the common-but-late-stage symptoms are noticeable.
As for your gripe about:
> let’s dump medical records and retina images into a neural network and see what correlates
, I agree that this requires more careful analysis, but I assume that will happen as part of clinical trials of any such technology. The first foray, which is more experimental than anything concrete, is what this article seems to suggest.
[1]: https://www.ncbi.nlm.nih.gov/books/NBK10885/#:~:text=Despite...).
I was left with exactly the same impression. This article was about a data set and the predictions being made from it without any mention of accuracy.
I would hate to get a wrong diagnosis of Parkinson's.
The answer invariably seems to be you can’t (yet).
Realistic: Doesn't earlier "treatment" of Parkinson's Disease yield better outcomes? I.e. dopamine supplements, etc.