Skip to content

Computers are better at diagnosing?

A new study analyzing the role of IBM’s supercomputer named Watson in medical decision making was just published in Artificial Intelligence in Medicine. While the most acclaimed medical professionals might keep some studies in mind, Watson can check millions of them quickly. Instead of fighting them, doctors should realize we need to include such solutions in the everyday medical decision-making processes.

Using 500 randomly selected patients from that group for simulations, the two compared actual doctor performance and patient outcomes against sequential decision-making models, all using real patient data. They found great disparity in the cost per unit of outcome change when the artificial intelligence model’s cost of $189 was compared to the treatment-as-usual cost of $497.

“This was at the same time that the AI approach obtained a 30 to 35 percent increase in patient outcomes,” Bennett said. “And we determined that tweaking certain model parameters could enhance the outcome advantage to about 50 percent more improvement at about half the cost.”

I’m very glad they added this message at the end:

“Let humans do what they do well, and let machines do what they do well. In the end, we may maximize the potential of both.”


6 Comments Post a comment
  1. BenK #

    This is an interesting paper by the now deceased Bill Costerton and his associates, including Ehrlich, Stoodly, and others, that discusses the way diagnostics lag and impact treatment.

    Towards a New Paradigm in the Diagnosis and Treatment of Orthopedic Infections

    “Imagine an office visit of an older, but still poised, oncologist with her orthopedic surgeon. The surgeon must somehow break the news that a prosthetic knee, installed some 10 months previously, must be removed in a two-stage procedure that will leave the lady immobile while an infection is brought under control, and a new knee can be implanted in 3 months. If this frightening specter was an ovarian cancer, the oncologist would ask if the genome of her cancer had been sequenced and whether estrogen receptors were present on her malignant cells, so that a rational treatment strategy could be devised. The surgeon would say that the aspirate he took on the previous visit had grown “Staphylococcus epidermidis,” in small numbers that might be contaminants from the technician’s hands, and that most strains of Staphylococcus epidermidis these days were resistant to methicillin so she had better go on an aminoglycoside until her surgery early next week. And the twenty-first century would have ground to a sickening halt, on the residue of nineteenth century techniques, right there in the surgeons’ office.”

    March 3, 2014
  2. urigoren #

    WOW! Very interesting. I think it brings new life to the contreversy associated with the statments of Vinod Kushla Back in 2012 where he said:
    “Health care is like witchcraft and just based on tradition.”
    Entrepreneurs need to develop technology that would stop doctors from practicing like “voodoo doctors” and be more like scientists.
    Health care must be more data driven and about wellness, not sick care.
    Eighty percent of doctors could be replaced by machines.

    I dont know if you can replace doctors, but it is sure going to change their work!

    March 3, 2014
  3. I was just looking at Pew Research data regarding who people (in the US) turn to for an accurate medical diagnosis: 91% say a clinician is the most helpful source, rather than a fellow patient, family member, or friend. Not surprising, of course, given where we are, but good to remind ourselves of the current landscape:

    As for Watson, remember that it is only as good as its inputs:

    Thank you for the thought-provoking post!

    March 5, 2014
  4. nice ;)

    November 9, 2014

Trackbacks & Pingbacks

  1. 2014’s Most Popular Medical Stories About The Future of Medicine | ScienceRoll

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 61,000 other followers

%d bloggers like this: