Slash Boxes

SoylentNews is people

posted by Fnord666 on Saturday May 16 2020, @11:45PM   Printer-friendly
from the coagulation-is-the-name-of-the-game dept.

New tool helps distinguish the cause of blood clots:

A new tool using cutting-edge technology is able to distinguish different types of blood clots based on what caused them, according to a study published in eLife.

The tool could help physicians diagnose what caused a blood clot and help them select a treatment that targets cause to break it up. For example, it could help them determine if aspirin or another kind of anti-clotting drug would be the best choice for a person who has just had a heart attack or stroke.

[...] To develop a more effective approach to identifying different types of blood clots, Zhou and her colleagues took blood samples from a healthy individual and then exposed them to different clotting agents. The team captured thousands of images of the different types of clots using a technique called high-throughput imaging flow cytometry.

They next used a type of machine-learning technology called a convolutional neural network to train a computer to identify subtle differences in the shape of different types of clots caused by different molecules. They tested this tool on 25,000 clot images that the computer had never seen before and found it was also able to distinguish most of the clot types in the images.

Finally, they tested whether this new tool, which they named the intelligent platelet aggregate classifier (iPAC), can diagnose different clot types in human blood samples. They took blood samples from four healthy people, exposed them to different clotting agents, and showed that iPAC could tell the different types of clots apart.

"We showed that iPAC is a powerful tool for studying the underlying mechanism of clot formation," Zhou says. She adds that, given recent reports that COVID-19 causes blood clots, the technology could one day be used to better understand the mechanism behind these clots too, although much about the virus currently remains unknown.

Journal Reference
Yuqi Zhou, Atsushi Yasumoto, Cheng Lei, et al. Intelligent classification of platelet aggregates by agonist type, (DOI: 10.7554/eLife.52938)

Original Submission

This discussion has been archived. No new comments can be posted.
Display Options Threshold/Breakthrough Mark All as Read Mark All as Unread
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
  • (Score: -1, Flamebait) by Ethanol-fueled on Sunday May 17 2020, @12:30AM (4 children)

    by Ethanol-fueled (2792) on Sunday May 17 2020, @12:30AM (#995189) Homepage

    Masks bad! Masks Good!
    Incubation period 3 days! Incubation period 3 months!
    Clothing stores closed! Pot shops open!
    Antibodies good! Disease never goes away!
    Herd immunity good! Antibodies don't work!
    Disease attacks lungs! No wait, disease attacks nuts! No wait, disease clots blood! No wait, disease turns you Black!
    Medicine helps! No, medicine kills you!
    A Vaccine will end this! No, vaccines won't!
    Buy our vaccine, Goys!

    So while that all slowly sets in to the puny brains of you losers, it is at least accomplishing a few good things: Proving Chinks are bad, Proving Mexicans are filthy, and exposing Jew pharmaceutical companies for selling snake oil to profit as usual from fear.

    • (Score: -1, Offtopic) by Anonymous Coward on Sunday May 17 2020, @01:07AM

      by Anonymous Coward on Sunday May 17 2020, @01:07AM (#995196)

      It only proves that you're an idiot.

    • (Score: 0, Offtopic) by Anonymous Coward on Sunday May 17 2020, @01:14AM

      by Anonymous Coward on Sunday May 17 2020, @01:14AM (#995198)

      What about the Muslims and Japanese? Afraid to tell us their joint role because you're worried about al-jihadi ninja assassins?

    • (Score: 2, Funny) by Azuma Hazuki on Sunday May 17 2020, @04:48AM (1 child)

      by Azuma Hazuki (5086) Subscriber Badge on Sunday May 17 2020, @04:48AM (#995250) Journal

      What are you worried about? Given you've only got one helix yourself, our old friend Covid won't infect you, out of either professional courtesy or because even a virus has standards.

      I am "that girl" your mother warned you about...
      • (Score: 1, Interesting) by Ethanol-fueled on Sunday May 17 2020, @05:12AM

        by Ethanol-fueled (2792) on Sunday May 17 2020, @05:12AM (#995259) Homepage

        I already got it and beat it. The day after my last day working for Boston Dynamics, I went to a concert in an enclosed venue in which a lot of people were coughing. This was in early November, which supports the more conspiratorial theories about the disease taking off in September of last year.

        A day later, I got sick. I didn't realize it at the time because I thought I was hung over but I got sick in the weirdest way: I had coughed up a bit of grey and yellow custard but nothing larger than a marble in the span of a month. I still ran 6 mile runs, puffed a bit of tobacco vape and pot, maybe drank. During the daytime, it was nothing. But during the nighttime, holy shit. My bed was drenched in sweat, feverish as fuck, one time taking literally 8 hot showers in a night to do something about those chills. And I stayed sick for over a month.

        Can't say whether or not the vaping, weed smoking, and drinking helped slow the illness or helped prolong it. But I'm damn convinced I had it. And it was shitty, but just as shitty as the flus I had caught as a kid causing me to spend every Christmas eve for a few years spending my midnights puking my guts out. Except unlike the flus, this disease seemed to disappear completely during the daytime only to rear its ugly head during the nighttime.

  • (Score: -1, Troll) by Anonymous Coward on Sunday May 17 2020, @02:56AM

    by Anonymous Coward on Sunday May 17 2020, @02:56AM (#995229)

    asking for a friend

  • (Score: 2, Insightful) by Frosty Piss on Sunday May 17 2020, @05:03AM (1 child)

    by Frosty Piss (4971) on Sunday May 17 2020, @05:03AM (#995256)

    Because if it doesn’t use AI and have SYNERGY, it’s useless.

    • (Score: 0) by Anonymous Coward on Sunday May 17 2020, @01:19PM

      by Anonymous Coward on Sunday May 17 2020, @01:19PM (#995344)

      AI: check
      Chinese authors: check
      UCLA: check

      If you step into any random dept you will see the same thing. Shitty AI done by weak foreign grad students who hack downloaded code with hard-coded parameters until they get 1 example that works. The professor - obviously doesn't know his shit - presses them for "better quality" results. Nobody learns anything - physics, mathematics, nada. It's the same corrupt shit they're used to in China. The students just need to get a degree and get out. The prof needs MOAR Chinese laborers to demonstrate to the promotion committee he's attracting clients by his voluminous work and reputation in the field. The committe doesn't know shit either, not surprisingly. Up the ladder everyone goes weeeeh!

  • (Score: 0) by Anonymous Coward on Sunday May 17 2020, @01:30PM

    by Anonymous Coward on Sunday May 17 2020, @01:30PM (#995350)

    "New Tool Helps Distinguish the Cause of Blood Clots"

    Not even slightly. More like: Not very new CNN trained to recognize features just like it was designed to do 20 years ago. Kind of works in one small dataset. More data needed - or better yet, license my patent and take this creaky junk that doesn't really work off my hands.