Results 1 to 16 of 16

Thread: Columbia Uni researchers demo injectable single-chip system

  1. #1
    HEXUS.admin
    Join Date
    Apr 2005
    Posts
    31,206
    Thanks
    0
    Thanked
    2,003 times in 709 posts

    Columbia Uni researchers demo injectable single-chip system

    Tiny implantable medical device is a temperature monitor with ultrasound antenna.
    Read more.

  2. #2
    RIP Peterb ik9000's Avatar
    Join Date
    Nov 2009
    Posts
    7,184
    Thanks
    1,645
    Thanked
    1,232 times in 926 posts
    • ik9000's system
      • Motherboard:
      • Asus P7H55-M/USB3
      • CPU:
      • i7-870, Prolimatech Megahalems, 2x Akasa Apache 120mm
      • Memory:
      • 4x4GB Corsair Vengeance 2133 11-11-11-27
      • Storage:
      • 2x256GB Samsung 840-Pro, 1TB Seagate 7200.12, 1TB Seagate ES.2
      • Graphics card(s):
      • Gigabyte GTX 460 1GB SuperOverClocked
      • PSU:
      • NZXT Hale 90 750w
      • Case:
      • BitFenix Survivor + Bitfenix spectre LED fans, LG BluRay R/W optical drive
      • Operating System:
      • Windows 7 Professional
      • Monitor(s):
      • Dell U2414h, U2311h 1920x1080
      • Internet:
      • 200Mb/s Fibre and 4G wifi

    Re: Columbia Uni researchers demo injectable single-chip system

    And people wonder where the conspiracy theorists get their worries from about state surveilance via vaccines...

    Like all tech this has good applications, but can potentially be exploited for nefarious reasons if not carefully controlled. For my relative with diabetes it could help her monitor it without having to have that weird wire thing in the side of her abdomen so there's definitely good uses this can to.

  3. Received thanks from:

    kompukare (15-05-2021)

  4. #3
    Registered User
    Join Date
    Jan 2021
    Posts
    14
    Thanks
    0
    Thanked
    1 time in 1 post

    Re: Columbia Uni researchers demo injectable single-chip system

    How will they keep this microcomputer from reaching and getting stuck in some small blood vessel?

  5. #4
    Member
    Join Date
    Nov 2018
    Posts
    104
    Thanks
    0
    Thanked
    1 time in 1 post

    Re: Columbia Uni researchers demo injectable single-chip system

    Scientific ethics should impose the R&D team to test this microcomputer firstly into their anus.

  6. #5
    Registered+
    Join Date
    Dec 2016
    Location
    Buckinghamshire
    Posts
    73
    Thanks
    2
    Thanked
    3 times in 3 posts
    • Ballantin's system
      • Motherboard:
      • Asus Rampage Extreme Omega
      • CPU:
      • Intel i9 9900X
      • Memory:
      • 64Gb Gskill Trident Z 3600
      • Graphics card(s):
      • 2x RTX 2080Ti
      • PSU:
      • Corsair AX1600i
      • Case:
      • Corsair 800D
      • Operating System:
      • Windows 10 Pro
      • Internet:
      • Vodafone 900Mb

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by Wirko View Post
    How will they keep this microcomputer from reaching and getting stuck in some small blood vessel?
    You usually encapsulate it in a glass capsule, like the ones used in NFC implants or the ones used to identify pets.

  7. #6
    Member
    Join Date
    Mar 2014
    Location
    Helion Prime
    Posts
    146
    Thanks
    2
    Thanked
    6 times in 6 posts
    • fend_oblivion's system
      • Motherboard:
      • ASUS M5A78L-M/USB3
      • CPU:
      • AMD FX 4100 3.6 GHz Black Edition
      • Memory:
      • G.Skill RipJawsX 8 GB DDR3 @ 1600 MHz
      • Storage:
      • Western Digital Blue 1 TB
      • PSU:
      • Seasonic S1211 520 W
      • Case:
      • Cooler Master K380
      • Operating System:
      • Windows 7 Ultimate Edition 64-bit
      • Monitor(s):
      • Acer P166HQL

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by john_amstrad View Post
    scientific ethics should impose the r&d team to test this microcomputer firstly into their anus.
    rofl

  8. #7
    Senior Member
    Join Date
    Aug 2016
    Posts
    2,305
    Thanks
    557
    Thanked
    530 times in 397 posts

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by ik9000 View Post
    And people wonder where the conspiracy theorists get their worries from about state surveilance via vaccines...

    Like all tech this has good applications, but can potentially be exploited for nefarious reasons if not carefully controlled. For my relative with diabetes it could help her monitor it without having to have that weird wire thing in the side of her abdomen so there's definitely good uses this can to.
    Well, true enough I suppose BUT the fact remains that we could take that basic logic and apply it to just about every medical advance in, oh, pretty much forever.

    Some such advances will, yes, need regulation and, yes, contain a potential for abuse but some others (including, I would suspect, this one) also have a potentiaaly HUGE benefit in either providing similar treatments to those currently available but in a much less invasive (and probably vastly less expensive) way, or to provide treatment options that just either aren't available any other way, or are but at much greater risk to the patient.

    For instance, delivering a drug payload right to where it's needed. Chemotherapy might be a good example, whereby injecting this into the right blood vessel, monitoring where it is then triggering drug deployment right at the target tumour might even provide a cure for some cancers, or at least, a way to poison the tumour more effectively than poisoning the whole patient in the process.

    So like any medical procedure, I'm all for developing this but with the necessary oversight. Then, I guess, like vaccines, the overly suspicious can always decline such treatment .... but they then can't really moan and whine if they don't benefit from the resulting treatment benefits or vaccines.

    For that reason, anybody going down with Covid that was offered vaccination and turned it down is not going to get much sympathy from me, and if my vaccination jabs turn out to be some government/CIA plot for mind control, I won't expect sympathy from the vaccine sceptics.

    Meantime, there's billions of people wanting vaccinations taht aren't (yet) getting the opportunity so I'm not wasting much effort worrying about the Darwin Award self-selection effect on those lucky enough to live where they get the offer, and yet stupid enough to turn it down. Fine, move on with the program and shift the doses to those billions that do want it.

    And ditto with treatments derived from this.
    A lesson learned from PeterB about dignity in adversity, so Peter, In Memorium, "Onwards and Upwards".

  9. Received thanks from:

    Ttaskmaster (17-05-2021)

  10. #8
    Long member
    Join Date
    Apr 2008
    Posts
    2,237
    Thanks
    67
    Thanked
    376 times in 269 posts
    • philehidiot's system
      • Motherboard:
      • Father's bored
      • CPU:
      • Cockroach brain V0.1
      • Memory:
      • Innebriated, unwritten
      • Storage:
      • Big Yellow Self Storage
      • Graphics card(s):
      • Semi chewed Crayola Mega Pack
      • PSU:
      • 20KW single phase direct grid supply
      • Case:
      • Closed, Open, Cold
      • Operating System:
      • Cockroach
      • Monitor(s):
      • The mental health nurses
      • Internet:
      • Please.

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by Wirko View Post
    How will they keep this microcomputer from reaching and getting stuck in some small blood vessel?
    There are a couple of considerations. Firstly whether it will be bioinert as is or need encapsulating and how it's injected. If it's a simple temperature or tissue glucose monitor, subcutaneously would be fine. We do this all the time with contraceptive implants, miniature heart monitors, etc. I inject myself regularly subcutaneously and it's not hard (although the more viscous injections are often quite painful). I'd doubt you'd want to go into a muscle with something like this.

    If you did send it into a blood vessel, then it'll likely get wedged somewhere. The risk is more if it's thrombogenic and becomes a source of clots or if it's put in a silly place and goes somewhere important. Most of the time small venous clots just get dumped in the pulmonary capillary bed and the arterial ones go as far as they can down and just wedge, usually harmlessly. If the technology gets smaller, you might be able to mount this on an arterial stent and deploy it using existing, minimally invasive technology. You might also be able to place it on a lead so it's in the venous system but retained outside the vessel, implanted like a pacemaker lead or certain types of implantable line for drugs (which are fully implantable and allow administration of drugs without an exterior line).

    We already have widespread implantable technology in pacemakers, defibs, nerve stimulators, cochlea implants, various diabetes management systems, etc. Yet for some reason people seem to look at the tiny chip and get paranoid. Trust me, as someone who routinely injects himself and is an user of various medical implants - you are going to notice if someone injects you with this, no matter how small.

    There's real potential for a system like this. Nefarious uses would be few and far between, given we already have similarly sized RFID chips which are powered by RF, rather than ultrasound. The technology for powering medical implants with ultrasound isn't new and is currently used in a generation of heart failure pacemakers.

  11. Received thanks from:

    ik9000 (17-05-2021)

  12. #9
    RIP Peterb ik9000's Avatar
    Join Date
    Nov 2009
    Posts
    7,184
    Thanks
    1,645
    Thanked
    1,232 times in 926 posts
    • ik9000's system
      • Motherboard:
      • Asus P7H55-M/USB3
      • CPU:
      • i7-870, Prolimatech Megahalems, 2x Akasa Apache 120mm
      • Memory:
      • 4x4GB Corsair Vengeance 2133 11-11-11-27
      • Storage:
      • 2x256GB Samsung 840-Pro, 1TB Seagate 7200.12, 1TB Seagate ES.2
      • Graphics card(s):
      • Gigabyte GTX 460 1GB SuperOverClocked
      • PSU:
      • NZXT Hale 90 750w
      • Case:
      • BitFenix Survivor + Bitfenix spectre LED fans, LG BluRay R/W optical drive
      • Operating System:
      • Windows 7 Professional
      • Monitor(s):
      • Dell U2414h, U2311h 1920x1080
      • Internet:
      • 200Mb/s Fibre and 4G wifi

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by philehidiot View Post
    There's real potential for a system like this.
    Agreed, like I said, there are good uses this can be put to.

    Quote Originally Posted by philehidiot View Post
    Nefarious uses would be few and far between, given we already have similarly sized RFID chips which are powered by RF, rather than ultrasound. The technology for powering medical implants with ultrasound isn't new and is currently used in a generation of heart failure pacemakers.
    Nefarious uses could be as broad as someone's imagination - but it's not to say that's how it will be used. Everything has a potential misapplication for no-good. Even a postage stamp can be used to send hate-email, junk mail, etc. Just because that isn't the intent doesn't mean it won't be used in other ways. TNT for example. This is SFAIK how conspiracy theories gain traction, by pointing out such examples and saying "so you can't rule it out because... look at this..."

    Now, likelihood, motive and opportunity etc all come into consideration, and I would hope you are correct, but it is usually best to consider how something might be abused and build-in safeguards or apply controls from the outset if they are merited. Even if just to stop the CS snowball over the fraction % outlier chance of a problem.

  13. #10
    Long member
    Join Date
    Apr 2008
    Posts
    2,237
    Thanks
    67
    Thanked
    376 times in 269 posts
    • philehidiot's system
      • Motherboard:
      • Father's bored
      • CPU:
      • Cockroach brain V0.1
      • Memory:
      • Innebriated, unwritten
      • Storage:
      • Big Yellow Self Storage
      • Graphics card(s):
      • Semi chewed Crayola Mega Pack
      • PSU:
      • 20KW single phase direct grid supply
      • Case:
      • Closed, Open, Cold
      • Operating System:
      • Cockroach
      • Monitor(s):
      • The mental health nurses
      • Internet:
      • Please.

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by ik9000 View Post
    Agreed, like I said, there are good uses this can be put to.



    Nefarious uses could be as broad as someone's imagination - but it's not to say that's how it will be used. Everything has a potential misapplication for no-good. Even a postage stamp can be used to send hate-email, junk mail, etc. Just because that isn't the intent doesn't mean it won't be used in other ways. TNT for example. This is SFAIK how conspiracy theories gain traction, by pointing out such examples and saying "so you can't rule it out because... look at this..."

    Now, likelihood, motive and opportunity etc all come into consideration, and I would hope you are correct, but it is usually best to consider how something might be abused and build-in safeguards or apply controls from the outset if they are merited. Even if just to stop the CS snowball over the fraction % outlier chance of a problem.
    Random thoughts whilst I'm in withdrawal and my brain is dead...

    The controls for medical implants are pretty sound. There's a load of testing and approval and a massive barrier to entry in terms of technology required to create something implantable that's safe and lasts. Attempts to do this with RFID chips and magnets by the biohacking community are mostly sucessful but degredation of the coating happens and explantation is occasionally required.

    Security of existing implants is mostly by obscurity and we have been known to have to apply cybersecurity updates to devices. Companies using established technologies like Bluetooth usually have a secondary bespoke chip to handle security. But when someone in the US government got a defib, the manufactuer produced a bespoke one with the telemetry disabled. Security by obscurity is a dangerous thing.

    We could pop them in bullets and Judge Dredd style encode them with the shooter's DNA.

  14. #11
    RIP Peterb ik9000's Avatar
    Join Date
    Nov 2009
    Posts
    7,184
    Thanks
    1,645
    Thanked
    1,232 times in 926 posts
    • ik9000's system
      • Motherboard:
      • Asus P7H55-M/USB3
      • CPU:
      • i7-870, Prolimatech Megahalems, 2x Akasa Apache 120mm
      • Memory:
      • 4x4GB Corsair Vengeance 2133 11-11-11-27
      • Storage:
      • 2x256GB Samsung 840-Pro, 1TB Seagate 7200.12, 1TB Seagate ES.2
      • Graphics card(s):
      • Gigabyte GTX 460 1GB SuperOverClocked
      • PSU:
      • NZXT Hale 90 750w
      • Case:
      • BitFenix Survivor + Bitfenix spectre LED fans, LG BluRay R/W optical drive
      • Operating System:
      • Windows 7 Professional
      • Monitor(s):
      • Dell U2414h, U2311h 1920x1080
      • Internet:
      • 200Mb/s Fibre and 4G wifi

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by philehidiot View Post
    Random thoughts whilst I'm in withdrawal and my brain is dead...

    The controls for medical implants are pretty sound. There's a load of testing and approval and a massive barrier to entry in terms of technology required to create something implantable that's safe and lasts. Attempts to do this with RFID chips and magnets by the biohacking community are mostly sucessful but degredation of the coating happens and explantation is occasionally required.

    Security of existing implants is mostly by obscurity and we have been known to have to apply cybersecurity updates to devices. Companies using established technologies like Bluetooth usually have a secondary bespoke chip to handle security. But when someone in the US government got a defib, the manufactuer produced a bespoke one with the telemetry disabled. Security by obscurity is a dangerous thing.

    We could pop them in bullets and Judge Dredd style encode them with the shooter's DNA.
    The CS crowd have theories that go along the lines of the uber government / elite/ lizard people / alien overlords (take your pick) want to chip everyone with nano chips that let them monitor/control/selectively terminate you at will (take your pick). They still buy the latest iphone mind.

  15. #12
    Long member
    Join Date
    Apr 2008
    Posts
    2,237
    Thanks
    67
    Thanked
    376 times in 269 posts
    • philehidiot's system
      • Motherboard:
      • Father's bored
      • CPU:
      • Cockroach brain V0.1
      • Memory:
      • Innebriated, unwritten
      • Storage:
      • Big Yellow Self Storage
      • Graphics card(s):
      • Semi chewed Crayola Mega Pack
      • PSU:
      • 20KW single phase direct grid supply
      • Case:
      • Closed, Open, Cold
      • Operating System:
      • Cockroach
      • Monitor(s):
      • The mental health nurses
      • Internet:
      • Please.

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by ik9000 View Post
    The CS crowd have theories that go along the lines of the uber government / elite/ lizard people / alien overlords (take your pick) want to chip everyone with nano chips that let them monitor/control/selectively terminate you at will (take your pick). They still buy the latest iphone mind.
    Oh I love that irony. "WE'RE ALL GOING TO BE CHIPPED WITH TRACKING DEVICES WHICH LISTEN TO US AND TRACK OUR MOVES AND CAN PREDICT OUR THOUGHTS!"

    Uh huh... so you're typing this on an Android phone that sends all your data to Google who predict your thoughts and movements?

    I do love how Google is curently managing to help diagnose adults with undiagnosed autism by recognising patterns and recommending videos entitled "Do you have autism?"

  16. #13
    Intelvidia Vanguard CAT-THE-FIFTH's Avatar
    Join Date
    Aug 2006
    Location
    Not here
    Posts
    31,016
    Thanks
    3,611
    Thanked
    4,954 times in 3,827 posts
    • CAT-THE-FIFTH's system
      • Motherboard:
      • Less E-PEEN
      • CPU:
      • Massive E-PEEN
      • Memory:
      • RGB E-PEEN
      • Storage:
      • Not in any order
      • Graphics card(s):
      • EVEN BIGGER E-PEEN
      • PSU:
      • OVERSIZED
      • Case:
      • UNDERSIZED
      • Operating System:
      • DOS 6.22
      • Monitor(s):
      • NOT USUALLY ON....WHEN I POST
      • Internet:
      • FUNCTIONAL

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by philehidiot View Post
    Oh I love that irony. "WE'RE ALL GOING TO BE CHIPPED WITH TRACKING DEVICES WHICH LISTEN TO US AND TRACK OUR MOVES AND CAN PREDICT OUR THOUGHTS!"

    Uh huh... so you're typing this on an Android phone that sends all your data to Google who predict your thoughts and movements?

    I do love how Google is curently managing to help diagnose adults with undiagnosed autism by recognising patterns and recommending videos entitled "Do you have autism?"
    Whilst our actual medical data is up for sale(or analysis) by huge big data firms such as Palantir technologies. I mean even the big centralisation of all NHS data,or the fact the government were hosting Google,etc all to see how much they can charge for the access,is all swept under the carpet.

    But,but injectable 5G microchips in our vaccines - yes lets go the convoluted way!!
    https://t6.rbxcdn.com/8d3e07528aae38af2cb81bdfa2a712e3

    I might not be serious.

  17. #14
    Long member
    Join Date
    Apr 2008
    Posts
    2,237
    Thanks
    67
    Thanked
    376 times in 269 posts
    • philehidiot's system
      • Motherboard:
      • Father's bored
      • CPU:
      • Cockroach brain V0.1
      • Memory:
      • Innebriated, unwritten
      • Storage:
      • Big Yellow Self Storage
      • Graphics card(s):
      • Semi chewed Crayola Mega Pack
      • PSU:
      • 20KW single phase direct grid supply
      • Case:
      • Closed, Open, Cold
      • Operating System:
      • Cockroach
      • Monitor(s):
      • The mental health nurses
      • Internet:
      • Please.

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by CAT-THE-FIFTH View Post
    Whilst our actual medical data is up for sale(or analysis) by huge big data firms such as Palantir technologies. I mean even the big centralisation of all NHS data,or the fact the government were hosting Google,etc all to see how much they can charge for the access,is all swept under the carpet.

    But,but injectable 5G microchips in our vaccines - yes lets go the convoluted way!!
    I'm torn on the NHS selling data for analysis to big data firms. I can really see the benefits of analysing the data and patterns being identified that we simply don't spot. The potential for products that can be applied to GP systems or even the central patient record system, to identify and manage at-risk patients way earlier than we ever could is huge. The potential for cost savings are also huge.

    Even more so, we might find that a cluster of certain signs and symptoms mean that certain procedures have a really poor outcome, or that stuff we have been doing for decades is downright harmful.

    We're good at seeing patterns within our specialties (a perfect example is a guy spotting on MRI that a certain, weird motion of a bit of cardiac muscle was happening but being unable to explain it) but it requires discussion with people outside of our areas and chance to understand or broaden a perspective (he happened to be stuck in a room with me and I mentioned I'd analysed the same thing using a different modality and was equally clueless, but together we got an explanation that went into a paper). We do some studies trying to predict adverse events by looking at bloods and other test results but it's educated guessing. Big data could simply be asked to look for this outcome and work backwards, seeing what factors play into it from across medicine, not just in our own little areas.

    I can see real potential but, equally, there's a moral issue with selling healthcare data, anonymised or not, to companies who want to monetise it as best they can.

    But then you could imagine an awesome kind of price war. Big data companies promise to prevent severe disease by identifing it early and offering effective lifestyle mods or supplements for the price of the analysis package for your patient group. Drugs companies currently charging huge amounts for the treatment would be forced to look more into preventative medicine as well, essentially making expensive treatments a thing of the past.

  18. #15
    Intelvidia Vanguard CAT-THE-FIFTH's Avatar
    Join Date
    Aug 2006
    Location
    Not here
    Posts
    31,016
    Thanks
    3,611
    Thanked
    4,954 times in 3,827 posts
    • CAT-THE-FIFTH's system
      • Motherboard:
      • Less E-PEEN
      • CPU:
      • Massive E-PEEN
      • Memory:
      • RGB E-PEEN
      • Storage:
      • Not in any order
      • Graphics card(s):
      • EVEN BIGGER E-PEEN
      • PSU:
      • OVERSIZED
      • Case:
      • UNDERSIZED
      • Operating System:
      • DOS 6.22
      • Monitor(s):
      • NOT USUALLY ON....WHEN I POST
      • Internet:
      • FUNCTIONAL

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by philehidiot View Post
    I'm torn on the NHS selling data for analysis to big data firms. I can really see the benefits of analysing the data and patterns being identified that we simply don't spot. The potential for products that can be applied to GP systems or even the central patient record system, to identify and manage at-risk patients way earlier than we ever could is huge. The potential for cost savings are also huge.

    Even more so, we might find that a cluster of certain signs and symptoms mean that certain procedures have a really poor outcome, or that stuff we have been doing for decades is downright harmful.

    We're good at seeing patterns within our specialties (a perfect example is a guy spotting on MRI that a certain, weird motion of a bit of cardiac muscle was happening but being unable to explain it) but it requires discussion with people outside of our areas and chance to understand or broaden a perspective (he happened to be stuck in a room with me and I mentioned I'd analysed the same thing using a different modality and was equally clueless, but together we got an explanation that went into a paper). We do some studies trying to predict adverse events by looking at bloods and other test results but it's educated guessing. Big data could simply be asked to look for this outcome and work backwards, seeing what factors play into it from across medicine, not just in our own little areas.

    I can see real potential but, equally, there's a moral issue with selling healthcare data, anonymised or not, to companies who want to monetise it as best they can.

    But then you could imagine an awesome kind of price war. Big data companies promise to prevent severe disease by identifing it early and offering effective lifestyle mods or supplements for the price of the analysis package for your patient group. Drugs companies currently charging huge amounts for the treatment would be forced to look more into preventative medicine as well, essentially making expensive treatments a thing of the past.
    The issue when big data companies like Palantir are very opaquely run,and seem to have everything from access to NHS data,to intelligence data from MI5/MI6. OCP and Weylan-Yutani comes to mind with these companies - fingers in every pie. Its why its good companies like Google are under more scrutiny - these companies don't know national boundaries so can shirk their responsibilities.

    We can't even get companies like Amazon to pay proper taxes,as we then get threatened by the US,etc. So how are we going to regulate these foreign big data companies,when any sort of action could lead to a trade war?? Because of their wealth they can game the system very easily though lobbying.

    Also as shown in the US it won't drop prices. The big data companies will work with pharma companies - this is why they are entering the medical arena. Its worth trillions worldwide - why would they want to make it less profitable? Its more likely they will use the data to recommend earlier medication - lots of people are addicted to various prescription drugs in the US:
    https://www.nature.com/articles/d41586-019-02686-2

    The only way a system will have price reductions is if people can abstain from it,if it does not deliver and wait for a better deal. That is not going to happen with medical stuff as its time limited - either you get treatment or you die. This is why in the US everything is so overpriced - they have you by the privates. Most people will think about saving their life first,and then worry about the cost. Nearly half of personal bankruptcies in the US are due to medical costs.

    The problem here is also the track record of the last 30 years of Tory/Labour governments are incredibly poor. If this was a job and your customer had to repeatedly keep a track of whether anything you were doing was done properly,that you were not stealing,etc you would be soon out of a job. IMHO,democracy is being made a joke off. A normal person is trying to just barely survive instead of trying to keep our politicians on the straight and narrow!
    Last edited by CAT-THE-FIFTH; 18-05-2021 at 01:52 PM.
    https://t6.rbxcdn.com/8d3e07528aae38af2cb81bdfa2a712e3

    I might not be serious.

  19. #16
    Long member
    Join Date
    Apr 2008
    Posts
    2,237
    Thanks
    67
    Thanked
    376 times in 269 posts
    • philehidiot's system
      • Motherboard:
      • Father's bored
      • CPU:
      • Cockroach brain V0.1
      • Memory:
      • Innebriated, unwritten
      • Storage:
      • Big Yellow Self Storage
      • Graphics card(s):
      • Semi chewed Crayola Mega Pack
      • PSU:
      • 20KW single phase direct grid supply
      • Case:
      • Closed, Open, Cold
      • Operating System:
      • Cockroach
      • Monitor(s):
      • The mental health nurses
      • Internet:
      • Please.

    Re: Columbia Uni researchers demo injectable single-chip system

    Quote Originally Posted by CAT-THE-FIFTH View Post
    The issue when big data companies like Palantir are very opaquely run,and seem to have everything from access to NHS data,to intelligence data from MI5/MI6. OCP and Weylan-Yutani comes to mind with these companies - fingers in every pie. Its why its good companies like Google are under more scrutiny - these companies don't know national boundaries so can shirk their responsibilities.

    We can't even get companies like Amazon to pay proper taxes,as we then get threatened by the US,etc. So how are we going to regulate these foreign big data companies,when any sort of action could lead to a trade war?? Because of their wealth they can game the system very easily though lobbying.

    Also as shown in the US it won't drop prices. The big data companies will work with pharma companies - this is why they are entering the medical arena. Its worth trillions worldwide - why would they want to make it less profitable? Its more likely they will use the data to recommend earlier medication - lots of people are addicted to various prescription drugs in the US:
    https://www.nature.com/articles/d41586-019-02686-2

    The only way a system will have price reductions is if people can abstain from it,if it does not deliver and wait for a better deal. That is not going to happen with medical stuff as its time limited - either you get treatment or you die. This is why in the US everything is so overpriced - they have you by the privates. Most people will think about saving their life first,and then worry about the cost. Nearly half of personal bankruptcies in the US are due to medical costs.

    The problem here is also the track record of the last 30 years of Tory/Labour governments are incredibly poor. If this was a job and your customer had to repeatedly keep a track of whether anything you were doing was done properly,that you were not stealing,etc you would be soon out of a job. IMHO,democracy is being made a joke off. A normal person is trying to just barely survive instead of trying to keep our politicians on the straight and narrow!
    Earlier treatment saves lives and reduces cost. If you start someone on a statin (who is actually at risk) before an event or before significant vascular disease sets in, you make a huge difference.

    As for prescription drug dependence, it's not just the US. I'm dependent on opiates and steroids as a result of long term medication. But that medication allowed me to stay in work and not end up on disability. What would have happened with big data (in my world) would be me not being bounced around from pain specialist to physio to blah blah. But, instead the AI seeing the pattern of signs and symptoms which meant I'd respond to a specific treatment. It took finding a doctor who specialised in that treatment (by chance) to get anywhere because not everyone can know everything. They certainly can't review all your records, scans, meds and process that data in a ten-twenty minute clinic appointment. The correct big data analysis would have happened likely at the GP, leading to referral to the correct specialist.

    Bear in mind, drugs aren't started willy nilly. Just because the AI says to give a drug doesn't mean it'll happen. Many a time our protocols tell us to refer a patient for a specific treatment (because large studies have given nice simple thresholds for referral) but we look at the person in front of us and go "nah".

    In the US, the system is very different. Here a lot of the motivation is on reducing costs, especially with an ageing population. That means the emphasis is on prophylactic treatment. It's a lot cheaper to prevent a stroke than to treat it. We have simple scoring systems to predict stroke risk, but imagine an AI that could see the subtle changes in blood results that just go unnoticed and flag that risk. Or analysis of 20 million chest X rays shows that before getting diagnosed with cancer Y, there's a consistent, measureable but not really visible tiny shift in radio-opacity of a certain bit of body? Yes that patient may well end up on drugs, but drugs to treat a cancer earlier than would otherwise be detectable.

    We're already getting systems like this - a frailty score is provided to us on accessing a patient record and it's using loads of markers. Not as useful when a patient is in front of you, but very useful when you're writing a letter or on the phone to them. Also great for deciding on management as a severely frail patient has a prognosis of around 6 months or less. This simple score uses data that would take 20 minutes or more to acquire by looking through loads of letters and notes. We don't take it on face value, but it's a great guide and as a result of big data.

    Also bear in mind that the scandal around opiates being used for chronic pain was mainly due to a company lying to prescribers. You'd hope an AI wouldn't lie but you'd have that human input at the end making the decision.

    I absolutely agree with you on the relatively opaque nature of the companies. There's going to be a big problem validating the results. For example if your AI tells us to medicate 800,000 people in the UK with drug X, our regular statisticians and so on would have no way of verifying that data and there's a huge GIGO risk. Neural networks having weightings "tweaked" like Intel has benchmarks tweaked is a real risk and creating an oversight mechanism is going to be a challenge.

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •