Technology evangelists love the word disruption.
It signals the end of the comfortable, cozy world order where entrenched interests of a select few are toppled by something new-fangled that has a democratizing effect. Select becomes widespread, mainstream.
When Investment Rhymes with Canada
Canada has a proud history of achievement in the areas of science and technology, and the field of biomanufacturing and life sciences is no exception.
Yet this is a somewhat narrow view of the term “disrupt.” A secondary meaning, — one that is often ignored in all the breathless talk of innovation transforming healthcare — according to Merriam-Webster is: to throw into disorder.
And such is the case with artificial intelligence. IBM CEO Ginni Rometty gave an eloquent keynote address at the annual HIMSS meeting in Orlando on Monday painting a picture of a golden era where cognitive technologies and AI allow unprecedented insight to help doctors do what they do best: take care of people.
In her world view, AI enhances what doctors are capable of but will not replace them.
“This is not man versus machine,” Rometty declared. “This is man plus machine. It’s to augment human intelligence, the doctor’s intelligence”
On tweeting this quote, another fellow journalist who manages the Physician’s Practice publication almost immediately responded:
There may be some skepticism w/ that statement. #himss17 https://t.co/Q73qs6W0w0
— Gabriel Perna (@GabrielSPerna) February 20, 2017
He later added, “there’s no way that this does not replace jobs.”
A look at some of the examples of IBM Watson Health’s AI in action may help underscore why Perna is likely correct.
Exhibit A: Watson is amazing at looking through radiology images.
“If radiologists spent their time looking at all images in a CT image stack, they are wasting a lot of time,” explained Deborah DiSanzo, general manager of IBM Watson Health in January at the JP Morgan Healthcare Conference in San Francisco. “Watson can prioritize images in the stack so rather than reading the first image in the stack or some other workflow method, Watson could read the images and say,’Simply look at these images first.'”
This is AI at its best. Improving workflow. But let’s also remember that just as your humble iPhone learns more about you as you use it, AI is also learning, constantly. The more data it consumes the smarter it gets.
Vinod Khosla, the Silicon Valley venture capitalist who is less worried about upsetting the apple cart than IBM executives, believes radiology is ripe for technology-led disruption.
“Nobody argues [about the fact that] radiology will be replaced 80 percent by machine algorithm ….” he told an audience in January at a conference also in San Francisco.
A leading physician-turned health IT expert privately also voiced concerns about radiologists at HIMSS.
Exhibit B: Mayo Clinic’s use of IBM Watson’s Clinical Trial Matching has reduced the time that research coordinators take to screen and match cancer clinical trials and their eligibility criteria to patient profiles to 8 minutes, down from 30. Mayo Clinic oncologist Tufia Haddad believes this will free up the research coordinator’s time for more patient education and interaction.
That’s the beauty of it.
But the flip side needs to be acknowledged too.
We want technology to get smarter so that patient outcomes are improved. But just as automation has replaced humans in a variety of industries, so will AI in healthcare and beyond. A massive displacement across industry sectors will occur, if not today, then in the near future.
And yet, the acknowledgment of this reality is sorely lacking. If IBM doesn’t want to rock the boat, neither does HIMSS leaders because they see fear driving the question and not pragmatism.
“There is a fair amount of fear that underlies that concern,” responded Dr. Michael Zaroukian, chair of the board of directors of HIMSS North America to this question at a media luncheon on Monday. “Physicians have a different set of knowledge than the lay public and high principles that make them trustworthy to help achieve, restore the health of those there entrusted to care for.”
His answer suggested that people are very worried of being treated by machines and not doctors, which was not the purpose of the question of course. The point is not to have a situation where a generation of people replaced by technology feel left behind and unable to participate in the new economy. Something that should resonate after the recent election results.
However, in the end, Zaroukian did touch upon a something that can help to alleviate the disruption: education and retraining.
“All of us need to continue learning, growing, being curious,” he said. “I am not actually worried about massive layoffs, but I am curious about the issue of which job will go away, how people who are currently in those jobs need to be curiosity learning, and how organizations like HIMSS and all the other partners that we have [can] help them anticipate and get ready for the next step.”
Zaroukian is suggesting that people need to be retrained to use these newer technologies — possibly for the “new collar” jobs that Rometty said will exist in the future that are neither blue-collar not white-collar jobs.
I would only offer this: Today’s cars in cruise control mode still need a lot of input from the driver; the first generation of driverless cars being developed right now will still need drivers to be in the front seat. But it’s quite conceivable that future iterations will get smarter and we can simply drink coffee in the back seat. KITT will drive effortlessly while Michael Knight relaxes.
If your job is to drive around, this future may be frightening.
Healthcare will see this too. And we who designed these machines to create a glorious future must prepare for the impending disruption by first acknowledging it and then by putting smart heads together for a solution.
It’s a different moonshot but perhaps equally moral as all our healthcare moonshots.
Photo: DrAfter123, Getty Images
CORRECTION: An earlier version of the story misspelled Zaroukian’s name.
