[HN Gopher] Machine learning finds an improved way to match dono...
___________________________________________________________________
Machine learning finds an improved way to match donor organs with
patients
Author : abrax3141
Score : 50 points
Date : 2022-08-07 15:01 UTC (8 hours ago)
(HTM) web link (ccaim.cam.ac.uk)
(TXT) w3m dump (ccaim.cam.ac.uk)
| t_mann wrote:
| needs a [2020] tag. Mihaela van der Schaar's research group is
| highly productive, I'd be surprised if they hadn't published some
| update to that methodology by now
| carvking wrote:
| In leu of current events - the first thing that crossed my mind
| is what the CCP will do with this technology.
|
| But it's good to see more real world application of machine
| learning - feels like the distance from research to application
| is very slow.
|
| "Machine learning radiology" is something I used to monitor as a
| sign of improvement - since it's a "simple" use case. Too "busy"
| to google about now - any news on this ?
| VHRanger wrote:
| It's just a more efficient way to do a form of graph
| matching[1] on a dynamic graph (one that evolves over time).
|
| There wasn't really a pure closed form mathematic solution to
| the problem like we can on static graphs (Gale-Shapely
| algorithm for one). You find papers like this one [2] with
| heavy math and some sort of proposed solution.
|
| So this type of problem is a really a good fit for machine
| learning to propose solutions.
|
| 1. https://en.wikipedia.org/wiki/Graph_matching 2.
| https://www.semanticscholar.org/paper/Thickness-and-Informat...
| carvking wrote:
| For more context:
|
| https://en.wikipedia.org/wiki/Organ_harvesting_from_Falun_Go...
| lr1970 wrote:
| As someone doing ML for a living this development scares me.
| Trying policy to de-bias dataset against prior policy (which is
| changing with time by the way) is very difficult. And any new
| policy will effect the outcomes and new datasets going forward.
| Positive feedback loops in Reinforcement Learning is a huge
| problem.
| jjcon wrote:
| If it can give better outcomes wouldn't it be more unethical
| not to use it? We should certainly be aware of feedback loops
| but we can't let perfection be the enemy of good
| trident5000 wrote:
| "Sorry, our internal black box system determined you are not a
| candidate at this time. Please try again later."
|
| Aside from reinforcement loops being an issue, ML is only as good
| as the programmer makes it. And we will have no idea how it was
| constructed and neither will non-tech health professionals.
| Theres no way a system like this can be described in its entirety
| without mistake to surgeons when one line of code changes
| everything.
|
| So maybe it can help out human decision but I would not want this
| to have final say for quite some time.
| abrax3141 wrote:
| The FDA is trying hard to figure out how to validate ML:
| https://www.fda.gov/medical-devices/software-medical-device-...
| echelon wrote:
| We could revolutionize not only organ transplants, but all of
| human health though advances in human cloning.
|
| The biggest increase to human health span will be when we start
| replacing our bodies wholesale as they age.
|
| Head transplants onto monoclonal, HLA-non-expressing (antigen
| free) bodies at age 50 will renew cardiovascular, pulmonary, and
| immune health (eg new thymuses). It resets almost all of the
| genetic cancer clocks in our bodies. Coupled with drugs to combat
| brain tissue waste aggregation and plaque formation, it could see
| humans through to multiple hundred years life spans.
|
| As for ethics, the monoclonal bodies will be grown in a lab,
| without brains, born out of chimeric pig uteruses and
| artificially innervated for endocrine regulation.
|
| I talk about this a lot in my HN post history.
|
| https://news.ycombinator.com/item?id=26673243
|
| https://news.ycombinator.com/item?id=28534681
|
| https://news.ycombinator.com/item?id=30622820
|
| https://news.ycombinator.com/item?id=30407908
| abrax3141 wrote:
| But why?
| echelon wrote:
| To not die at 80.
|
| Do be young as long as you live.
|
| To cure people of cancer, heart, lung, liver, kidney, and
| other diseases. Intractable diseases.
|
| Once it becomes normalized, you'll even see people swapping
| their bodies for ones they prefer. Gender. Increased height,
| musculature, or VO2max. Transgenic craziness.
|
| Why not?
|
| Why accept what we have now?
| abrax3141 wrote:
| Why not: Permitting our children to breathe and eat, maybe?
| You had your 80 years. Get out of the way.
| echelon wrote:
| You're free to do you.
|
| If you haven't checked, we're below replacement rate in
| the West. Apparently most don't find kids to be a
| priority either.
|
| I'm going to spend every effort to maximize my one in
| infinity chance of experiencing the cosmos. Because after
| this, there's nothing.
|
| Make it long enough, and we might get to see it through
| to the end. An extreme long shot that demands a near
| perfect play through. But hey, what else is there to
| pursue?
|
| Making widgets and drinking whiskey while growing old?
| Everyone else seems to be exploring these well traveled
| paths. I'm not content to hope it ends there.
| tsimionescu wrote:
| By what possible moral judgment would you be entitled to the
| body or organs of a clone of yourself, any more than the clone
| would be to you?
|
| I do believe that growing clone organs individually, if it
| turns out to be possible, would be a solution. But otherwise,
| slaughtering a living human to save another is never going to
| be acceptable.
| echelon wrote:
| > slaughtering a living human
|
| A human body grown without a head or brain?
|
| That's the moral equivalent of a houseplant or a pile of
| lumber.
|
| Nobody cries over HeLa cells dying millions of times over. Or
| any other cell line.
|
| This is far more moral than killing lab rats. And it could
| end most diseases forever.
| tsimionescu wrote:
| Your said cloning, which is a technology we have today, but
| which implies growing a fully functioning human being.
|
| It may be possible in the future to grow parts of human
| bodies without the conscious bits, but we definitely don't
| know how to do it today beyond style very simplistic
| tissues.
|
| It's also important to note that there is no reason to
| believe that it is possible to grow a human body without a
| head or brain. It may be, but it may well not.
|
| Also, the brain and head age like the rest of the body, and
| a full body transplant would be just about the most extreme
| kind of surgery you could undorgo. I see no reason to
| believe that such extreme surgery would succeed in
| prolonging life, except in the most dire of circumstances
| (imminent vascular failure, terminal metastasized cancers
| that happen not to affect the brain). In most cases, for
| relatively healthy aging humans, it would certainly be a
| significant reduction in lifespan with anything similar to
| today's medical technology, even if you had a young healthy
| donor body.
| drc500free wrote:
| What about growing three or four organs at a time that
| mutually depend on each other?
| zzbn00 wrote:
| Around 90% of liver disease deaths is from preventable risk
| factors (alcohol, viral hepatitis and obesity,
| https://ukhsa.blog.gov.uk/2014/09/29/liver-disease-a-prevent...
| ). We have the tools to fix to these (the main being more even
| distribution of resources in society) yet we don't.
| ryan93 wrote:
| How would giving people more money decrease obesity and
| alcoholism? The richer our country has gotten the fatter it has
| gotten!
| fredoliveira wrote:
| I'm sure that if you think about it, you'll realize it is not
| the very wealthy that are getting more obese or drinking more
| alcohol.
| ryan93 wrote:
| Because that's who they are. The lottery doesn't change
| your character it reveals.
| monocasa wrote:
| Or perhaps it's not really a character thing at all, as
| the cheapest foods these days are high-calorie and
| engineered not to satiate.
| tyingq wrote:
| A fair amount of the lottery is family wealth at birth.
| And while it's just opinion, I believe that alcoholism
| has more to do with circumstance than character. There's
| certainly more access to treatment if you're wealthy too.
| conviencefee999 wrote:
| Not really, when Black Americans and other minority groups
| that grew up in poverty rise up in economic status they
| actually get fatter. Not sure where the statistic on it is
| but it was a really interesting find.
| SV_BubbleTime wrote:
| There are more calories in cheap foods. This has gotten
| worse with processing, but from where I can see, no
| obvious solution exists.
| zzbn00 wrote:
| "Poverty and childhood obesity: a 21st century role
| reversal" : https://foodactive.org.uk/poverty-and-
| childhood-obesity-a-21...
| cuteboy19 wrote:
| Seems like a difficult problem. Do you let 15 people have a
| subpar match or give 14 a perfect match and 1 person dies.
| dmckeon wrote:
| Improvements in organ donation/procurement networks in the USA
| are needed. "In 2020, 21.3 percent of procured kidneys were not
| transplanted," per
| https://www.washingtonpost.com/health/2022/08/03/transplant-...
| and https://onlinelibrary.wiley.com/doi/epdf/10.1111/ajt.16982 We
| should not let an unknowable and unachievable "perfect" be the
| enemy of better.
___________________________________________________________________
(page generated 2022-08-07 23:01 UTC)