1. Patients launch $1.27 million crowdfunding campaign for ME/CFS gut microbiome study.
    Check out the website, Facebook and Twitter. Join in donate and spread the word!
ME/CFS: A disease at war with itself
We can all agree that ME/CFS is a nasty disease, particularly in its severe form, but there are abundant nasty diseases in the world. What is unique and particularly confounding about our disease is that so much controversy surrounds it, and not only surrounds it, but invades it too.
Discuss the article on the Forums.

If nothing does, Quantum Computing Will Cure CFS

Discussion in 'Other Health News and Research' started by Waverunner, Nov 10, 2012.

  1. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    CFS is a tough nut to crack but it will be cured. It's just a question of when. Within the next 5 to 10 years, the first efficient quantum computers will become available. The power of quantum computers is measured in qubits (quantum bits). A quantum computer with only 30 qubits will be as powerful as a current super-computer. A quantum computer with 300 qubits would be as powerful as all the computers in the world connected together. What does this mean for the medical field? A quantum computer will enable us to simulate the whole human body up to its DNA. Of course we need to supply the computers with the input they need and this could take some years but it will be worth it. The creation of drugs will no longer be a series of trial and error but a well aimed simulation, that takes into account nearly all possibilities, interactions and dangers. So if everything fails and world war III is cancelled, quantum computers will solve the riddle of CFS.



    TedxTalks

    GhostGum, MishMash and Bob like this.
  2. MishMash

    MishMash *****

    Messages:
    445
    Likes:
    479
    Georgia
    I hope so. I've been following the application of computing to medical innovation for many years now. It has been quite a disappointment. I remember all the promises that were made back in the 90s. Most pharmaceutical companies got rid of many bench scientists and started buying super computers to the do the work. The medical breakthroughs in the last thirty years have been mostly patchy. Knowledge of the human organism is still pretty scant.

    However, history shows that it usually take several decades for a technology to come into optimum usage. For example, the Dept of Energy funded the first hydraulic fracturing trials back in 1973. It took more than thirty years of trial and error, and now natural gas production is the highest it has ever been. So I agree that after 30 or 40 years, computing will finally come around to completely understanding the human illness. That should be soon, god willing.
    taniaaust1 likes this.
  3. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    In my eyes natural gas production is a perfect example for trial and error. My fear in the health system however is, that not even trial and error are used efficiently. Most new scientific findings cannot be transformed into a practical state, because over-regulation creates huge upfront costs that keep many smart scientists or business men from entering the field. As soon as you don't have to pay for phase 3 trials anymore, medical development might get a huge boost. Quantum computers could circumvent the whole process of in vivo testing.
  4. xchocoholic

    xchocoholic Senior Member

    Messages:
    2,605
    Likes:
    1,231
    Florida
    My concern is that a computer is only as good as the data it's given and it's code. And so far much of the data
    is flawed by anyone wanting to make a profit off healthcare.

    Did you see Watson on Jeopardy ? His answers were purely based on his data banks and code / programs. Anything he got wrong on the first try
    could be corrected by coding for it or adding to his data banks so he got it right the second time. That's doable.

    That's not true in bio chemistry. There are simply too many possible combinations that must be present
    at the exact same time in just the right quantity in our bodies to code for.

    And then there's the fact that many of the chemicals in and outside of our bodies can't be identified or measured. And some can't be recreated outside of our bodies. And who's to say how much of each chemical each of us needs. Genes get turned on by "x" so what good is
    knowing our genes ?

    I see this as a tool for identifying diseases easier based on labs but that's all. Tc .. X
    Lou likes this.
  5. alex3619

    alex3619 Senior Member

    Messages:
    7,006
    Likes:
    10,757
    Logan, Queensland, Australia
    The complexity of the software needed is the killer for such simulations. Its not only not easy, the potential for error is extreme. Its a beginning, but its still a long road.
    taniaaust1 likes this.
  6. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    I get your point but Watson and its transistors cannot be compared to a quantum computer. Watson is more like a modern answering machine, it has nothing to do with artificial intelligence or the possibilities of quantum computers. Quantum computing and nanotechnology go hand in hand. In the future, MRI scans will be able to look at single molecules and they will tell us where atoms are located. This will lift diagnostics to a whole new level. Breathing upon your mirror will give you more sophisticated information about your health than a whole laboratory will give you today.

    Quantum computers could help to program quantum computers. If there will ever be something like AI or singularity, quantum computers will lay their basis.
  7. alex3619

    alex3619 Senior Member

    Messages:
    7,006
    Likes:
    10,757
    Logan, Queensland, Australia
    Yes Waverunner, I agree on the AI/singularity aspect. The potential impact of AI and quantum or nanoscale computing is why I studied AI in the first place. However while I am a fan of AI, and think most of the horror stories depicted about runaway AI is nonsense, its not entirely nonsense. When machines design machines we will gradually lose understanding of what is happening, particularly as the pace of that design cycle will probably increase. Bye, Alex
  8. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,258
    Likes:
    1,648
    Australia
    Quantum computing right now is mostly hype. Not only does serious quantum computing technology not exist, but there are so few quantum algorithms developed that are more effective than classical computer algorithms.

    Not saying it is impossible, but very much remains to be seen.
    Lou likes this.
  9. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    Well, we'll see.

    http://www.nature.com/srep/2012/120813/srep00571/full/srep00571.html

    "...This first implementation of a biophysical problem paves the way towards studying optimization problems in biophysics and statistical mechanics using quantum devices."
  10. Snow Leopard

    Snow Leopard Senior Member

    Messages:
    2,258
    Likes:
    1,648
    Australia
    The theoretical/algorithmic approach does seem quite impressive, but only getting the answer correct 13 out of 10000 times (and that was still a fairly simple example), shows that reliable QC technology is still a major barrier.
  11. xchocoholic

    xchocoholic Senior Member

    Messages:
    2,605
    Likes:
    1,231
    Florida
    I wouldn't say watson is like an answering machine but I'm an ex computer programmer. Tha algorithms necessary
    for Watson to answer the questions he did accurately are extremely complex and require him to go from
    one data base to another looking for clues based on what he found in each data base. Some of his answers were totally off track but that's what his
    data and programs led him to.

    Without these data bases of info tho, he's clueless. Which is the problem with using a computer to determine
    a cure or some causes for diseases. The data / chemicals aren't always something tangible that can be measured. And
    unless you go to a specialty clinic, the medical
    community doesn't have the capability or knowledge to accurately measure these chemicals. They can't measure
    everything either. We
    see this discussed here all the time. Finding xmrv is a good example of this.

    Another ex is tryptase. It has to be measured when someone is having an attack. That's
    not the way anaphylaxis should be dealt with tho. The drugs to stop the anaphylaxis and lower tryptase need to
    to be given.

    Fwiw, as a programmer I could tell you every single component of every airplane that came out of our plant. Down
    to the composition of each screws and where they were used and if they were the cause of any problems. That's just from reading data. I couldn't pull data that wasn't available tho. Data banks of all the chemicals in our bodies aren't available.
    They never will be ...

    Just my 2 cents ... X
    taniaaust1 and GracieJ like this.
  12. snowathlete

    snowathlete

    Messages:
    2,038
    Likes:
    2,116
    UK
    Computers are still very, very, dumb, depite the improvements in power that we have already seen and will continue to see. The raw power is helpful, but it only takes you so far.

    Of course there may be indirect benefits from quantum computing that will benefit medical discoveries. But until we teach computers to sucessfully go and collect and process data on their own (which is no small task) they wont be able to do incredibly complex stuff like solve illnesses. One day we will be able to inject nanobots to go and look at the human body and collect data on their own then come back and report and work out what it all means. That will be powerful. But i doubt we are really anywhere near that yet.
  13. alex3619

    alex3619 Senior Member

    Messages:
    7,006
    Likes:
    10,757
    Logan, Queensland, Australia
    The point of very powerful computing is not to solve such complex problems in absense of data. The point is to process complex and massive data in an effort to find solutions. If we can improve diagnostic technologies over time, particularly imaging technologies, we will need more and more computing power to process it. Much of the medical revolution has come from such computing. Modern MRI and similar technology relies on it. I don't think we could do coherent spectrum EEG in ME or CFS without it.

    Quantum computing like optical computing is still in experimentation. We have been on the "verge" of solving fusion power for decades. Nobody can predict when technology will suddenly cross the divide from experimental to commercial until it happens. We can only guesstimate.
  14. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    I agree that without diagnostics, the progress will be very slow but I'm quite confident that nanotechnology delivers the new means to test and scan for disease.
  15. MR AIDAN G WALSH

    MR AIDAN G WALSH

    Messages:
    25
    Likes:
    7
    Impossible to correct broken and translocated cromosones as a result of Alpha or Beta internal ionization radiation 'injuries' and that's a fact...CFS no longer exists period... the cause has been identified a 'misdiagnosed injury' not a syndrome, illness, disease, infection or all in one's head...fact...Leukemic cromosone experts are what is needed not imagination theories but replicted scientific evidence forthcoming...Sincerely and always the truth MR AIDAN G WALSH
  16. GracieJ

    GracieJ Senior Member

    Messages:
    432
    Likes:
    694
    Utah
    It's a nice dream, isn't it... a computer that can copy the body, DNA and all, and compute answers. I think it will facilitate the process along the way, but do not think humans will ever know enough about the body to program a machine like this.

    I am reminded of the AI experiments where the machine was asked to define the phrase "out of sight" (as in cool!) and the answer given was "blind."
    Little Bluestem likes this.
  17. Waverunner

    Waverunner Senior Member

    Messages:
    992
    Likes:
    835
    Seems the "Alpha and Beta internal ionization radiation" already took its toll on you. If you want to spam your weird theories, do it in your own thread and stop quoting a huge post when you have nothing to say about it. Thanks.

    "Sincerely and always the truth" :rolleyes:
  18. alex3619

    alex3619 Senior Member

    Messages:
    7,006
    Likes:
    10,757
    Logan, Queensland, Australia
    When I was teaching basic AI and discussing robotics, I gave an experience I had with two different type of robots. The first was a traditional AI. On board processor radio linked to big computer. It planned a move, started across room .... I stepped in front of it. Major meltdown! A minute later it figured a way around, started moving, so I stepped in front again. Another halt.

    An autonomous robot using cybernetic principles rather than symbolic logic just moves around you instantly. No planning. No thinking. No pausing.

    Getting the technology right is critical in these things.

    In the brain "symbolic" systems lie on a layer of non-symbolic neural systems. Symbols are not real. They are a tool. Think of symbols, including words, as a mathematical tool.
  19. user9876

    user9876 Senior Member

    Messages:
    707
    Likes:
    1,621

    There were ideas a while ago where people talked about subsymbolic representations which involved representing a symbol or a composition of symbols as a point in a high dimensional vector space.The idea was that the spacial relationships would capture the meaning and relationships between concepts. This allowed for associative inference where mappings could be learned (using neural networks) between different sets of concepts. There was a lot of work in natural language understanding in this area but also I think someone produced a planning system. However, there seemed to be a trade off between the ease of representing complex compositional structures in traditional AI but with slow speeds of algorothms that perform inference vs difficulties in representing concepts but with fast inference using neural network approaches. I've not followed the area since the mid 90s though so I don't know where it has gone.

    Although AI never lived up to its hype techniques developed are used in various systems and as data analysis techniques. Some of the formal logic based approaches have been used to model systems either proving properties (e.g. security protocols) or as a means of abstracting, simulating and exploring properties of a system.
  20. alex3619

    alex3619 Senior Member

    Messages:
    7,006
    Likes:
    10,757
    Logan, Queensland, Australia
    Hi user9876, my PhD proiject was a composite neural network system, the target being to develop methods for better real world categorization by neural networks. I am a big fan of neural networks, but have no problems with composite neural/symbolic systems for engineering applications. Neural networks are indeed fast to operate, but as the complexity of the system and the training set increases the time to train increases. I am not sure but I think that increase can be exponential unless it can be optimized, though once you have one trained you can mass produce more. My project was based upon neural nets that were modelled after the brain, and actually used in real time robotics where the robots learned to perform their tasks, they were not programmed. Learning systems are critical if AI robotics is to advance, but much less important if the robot is intended to have specific limited real world application, such as finding leaks in oil pipelines. Bye, Alex

See more popular forum discussions.

Share This Page