Slashdot Mirror


Microsoft Will 'Solve' Cancer Within The Next 10 Years By Treating It Like A Computer Virus, Says Company (independent.co.uk)

Microsoft is serious about finding a cure for cancer. In June, Microsoft researchers published a paper that shows how analyzing online activities can provide clues as to a person's chances of having cancer. They were able to identify internet users who had pancreatic cancer even before they'd been diagnosed, all from analyzing web query logs. Several months later, researchers on behalf of the company now say they will "solve" cancer within the next 10 years by treating it like a computer virus that invades and corrupts the body's cells. The goal is to monitor the bad cells and potentially reprogram them to be healthy again. The Independent reports: The company has built a "biological computation" unit that says its ultimate aim is to make cells into living computers. As such, they could be programmed and reprogrammed to treat any diseases, such as cancer. In the nearer term, the unit is using advanced computing research to try and set computers to work learning about drugs and diseases and suggesting new treatments to help cancer patients. The team hopes to be able to use machine learning technologies -- computers that can think and learn like humans -- to read through the huge amounts of cancer research and come to understand the disease and the drugs that treat it. At the moment, so much cancer research is published that it is impossible for any doctor to read it all. But since computers can read and understand so much more quickly, the systems will be able to read through all of the research and then put that to work on specific people's situations. It does that by bringing together biology, math and computing. Microsoft says the solution could be with us within the next five or ten years.

11 of 259 comments (clear)

  1. BWAH-HA-HA-HA-HA!!! by ewhac · · Score: 5, Insightful

    If there's one thing Microsoft has comprehensively and irrefutably established over the last 35 years of their existence, it's that they haven't the faintest clue how to identify or eradicate viruses.

    1. Re:BWAH-HA-HA-HA-HA!!! by dargaud · · Score: 2, Insightful

      Yeah, I read that title first thing in the morning and I very nearly spit orange juice all over the laptop.

      Seriously though, cancer is a very ancient disease. It comes from when the very first lonely cells decided to band together to increase survival, over 3 billion years ago. They figured out pretty quickly that group rules were necessary, like some cells go some places, others need to suicide at the right time, etc... And when a cell doesn't obey those ground rules and starts reproducing on its own, there you go: cancer.

      So something so grounded in our origins is going to be VERY hard to eradicate. Improve it, yes, we already have. Improve it a lot, probably. But get rid of it I somehow doubt it because any kind of transcription error in a single cell can possibly lead to it.

      --
      Non-Linux Penguins ?
    2. Re:BWAH-HA-HA-HA-HA!!! by silentcoder · · Score: 3, Insightful

      > What we have is an epidemic of cancer

      Nope. We really don't. The increase in cancer cases over time tracks *exactly* the increases in human lifespan. We've always been equally prone to it (barring isolated and regionally limited edge-cases) - but until quite recently almost everybody got killed by something else first.
      Now that we survive most virusses, bacteria and parasites and have basically eradicated just about all our natural predators (with the exception of the mosquito) - we actually live long enough for cancer to happen, and the more people live long - the more get it.

      That said - cancer is not a disease and does not have a cause. Cancer is a collective noun for a whole host of diseases all with different causes, which just happen to have one, single tiny thing in common. The reason we haven't cured cancer is because nothing could possibly do that - no single treatment can deal with so many different diseases, all with different causes (many of which are unknown). Even the shotgun treatments of radiation and chemo are not useful on all of them.
      On the other hand we are making massive progress in curing and preventing specific cancers. In the last few years, for the first time in history, we actually developed a vaccine that can completely prevent several cancers (HPV vaccines grant effective immunity against cervical cancer and several types of throat and lip cancers). The reason is that we discovered that a specific virus causes these cancers - and could create a vaccine against that virus.

      Gene-targetted treatments are already greatly increasing life expectancy, survival rates and quality of life of many cancer patients - with far less negative side effects than the shotgun treatments. More experimental treatments using things like magnetofluids are being investigated which may offer new and uniquely safe types of surgical treatments which are viable on a much larger set of cancers.
      We are making progress - but this is a war against a massive army with a huge variety of different batallayons and there is no one attack to defeat it, not single battle will win this war. Lots of small victories that add up - that's the way to do it, and it won't happen quickly, but it is already happening much quicker than we could have hoped even a decade ago.

      --
      Unicode killed the ASCII-art *
  2. Think and learn like humans by ptaff · · Score: 2, Insightful

    The team hopes to be able to use machine learning technologies -- computers that can think and learn like humans

    If your definition of a human is a retarded 4-year-old that can be trained to name colors with 75% accuracy, yes.

    We're not there, we're not even close; "machine learning" is just the new buzzword in town, rising from the ashes of "big data".

  3. No matter how clueless we are ... by pesho · · Score: 4, Insightful

    ... machine learning is the solution. And cancer is not "like a computer virus that invades and corrupts the body's cells". That is how an actual virus works, hence the analogy by which the "computer virus" term came to be. Cancer is more like when a bit randomly flips in RAM and then by pure coincidence this causes a memory leak within an infinite loop that spreads shit all over the place until everything comes crashing down.

    1. Re:No matter how clueless we are ... by wierd_w · · Score: 5, Insightful

      personally, i would say cancer is much more like wallstreet fatcats.

      a mutation in the rules governing proper behavior causes them to consume all available resources, send ssignals to the regulatory system that they are essential and need protection, while earnestly believing they are the most important part of the system while destroying it from the inside, due to the removal of a system to terminate that behavior early.

    2. Re:No matter how clueless we are ... by ColdWetDog · · Score: 4, Insightful

      NoNoNoNo. The underlying assumption that computers and humans are fundamentally similar is completely incorrect. The term 'computer virus' is a reasonable analogy but you can't push it so far that you impute that the mechanisms are the same. Cancer is way more complex that 'reprogramming a cell'. It involves cell homeostasis mechanisms that have no analogous function in hardware or software.

      "It’s not just an analogy, it’s a deep mathematical insight. Biology and computing are disciplines which seem like chalk and cheese but which have very deep connections on the most fundamental level.”

      (FTFA) Oh yeah. Prove it. Or even give us something other than executive level bullshit.

      Perhaps when you have computers that can handle errors more gracefully than "PC LOAD LETTER" I might think about taking him seriously. But we've barely moved past that level at present.

      --
      Faster! Faster! Faster would be better!
  4. Why not work on gravity and faster-than-light? by Bruce+Perens · · Score: 4, Insightful

    If we can solve the problem of cancer within 10 years by treating it as a computer virus, why not treat gravity as a computer virus and come up with practical, cheap antigravity? Or that pesky light-speed limit, we need to beat that, and 10 years sounds about right

    1. Re:Why not work on gravity and faster-than-light? by Burz · · Score: 3, Insightful

      This is a pitch to bolster the image of invasive data-mining, promising a blue-sky reward in return for watching peoples' browser activity.

      That is the only viral thing about this story--their greed.

  5. Re:Bill Gates and SPAM by Anonymous Coward · · Score: 2, Insightful

    He was right. It was effectively eradicated, at least as far as users were concerned, by 2006. Though the credit for that goes to Google, not Microsoft.

  6. Re:We're doomed by tehcyder · · Score: 3, Insightful

    Am I the only one who is scared about the fact that these clueless fuckwits have enough data on us to diagnose which of us has prostate cancer?

    Diagnosing that someone has prostate cancer because they've googled "symptoms of prostate cancer" isn't rocket surgery.

    --
    To have a right to do a thing is not at all the same as to be right in doing it