Search the KHIT Blog

Monday, September 11, 2017

Watson and cancer

"...there’s a rather basic but fundamental problem with Watson, and that’s getting patient data entered into it. Hospitals wishing to use Watson must find a way either to interface their electronic health records with Watson or hire people to manually enter patient data into the system. Indeed, IBM representatives admitted that teaching a machine to read medical records is “a lot harder than anyone thought.” (Actually, this rather reminds me of Donald Trump saying, “Who knew health care could be so complicated?” in response to the difficulty Republicans had coming up with a replacement for the Affordable Care Act.) The answer: Basically anyone who knows anything about it. Anyone who’s ever tried to wrestle health care information out of a medical record, electronic or paper, into a form in a database that can be used to do retrospective or prospective studies knows how hard it is..."
From Science Based Medicine. They've picked up on and run with the reporting first published by STATnews.
"Hospitals wishing to use Watson must find a way either to interface their electronic health records with Watson..."
Ahhh.. that pesky chronic 'interoperababble" data exchange problem.

SBM continues:
What can Watson actually do?
IBM represents Watson as being able to look for patterns and derive treatment recommendations that human doctors might otherwise not be able to come up with because of our human shortcomings in reading and assessing the voluminous medical literature, but what Watson can actually do is really rather modest. That’s not to say it’s not valuable and won’t get better with time, but the problem is that it doesn’t come anywhere near the hype...
Necessarily, Watson has to employ the more difficult "Natural Language Understanding" (NLU) component of Natural Language Processing (NLP). I have previously posted on my NLP/NLU concerns here.

Search Google news for "Watson oncology" or "Watson cancer."

I'm sure you've all seen the numerous Watson TV commercials by now.

Are we now skiing just past the "Peak of Inflated Expectations?"

Everything "OncoTech" is of acute interest to me these days amid my daughter's cancer illness. apropos, see my prior post "Siddhartha Mukherjee's latest on cancer."


THCB has a nice post on the topic.
7 Ways We’re Screwing Up AI in Healthcare

The healthcare AI space is frothy. Billions in venture capital are flowing, nearly every writer on the healthcare beat has at least an article or two on the topic, and there isn’t a medical conference that doesn’t at least have a panel if not a dedicated day to discuss. The promise and potential is very real.

And yet, we seem to be blowing it.

The latest example is an investigation in STAT News pointing out the stumbles of IBM Watson followed inevitably by the ‘is AI ready for prime time’ debate. If course, IBM isn’t the only one making things hard on itself. Their marketing budget and approach makes them a convenient target. Many of us – from vendors to journalists to consumers – are unintentionally adding degrees to an already uphill climb.

If our mistakes led to only to financial loss, no big deal. But the stakes are higher. Medical error is blamed for killing between 210,000 and 400,000 annually. These technologies are important because they help us learn from our data – something healthcare is notoriously bad at. Finally using our data to improve really is a matter of life and death…
Indeed. Good post. Read all of it.

Also of recent relevant note:
Athelas releases automated blood testing kit for home use
Silicon Valley-based startup Athelas today introduced a smartphone app that it says can do simple blood diagnosis at home and return results in just 60 seconds.

The kit itself looks a bit like an Amazon Echo device and is coupled with a smartphone app to reveal the results of the test. In a demonstration, co-founder Deepika Bodapati showed TechCrunch that from taking a sample of blood and sliding it into the device, within seconds users can see their white blood count, neutrophils, lymphocytes and platelets.

Bodapati and co-founder Tanay Tandon are well aware of the fate of a similar device that promised to deliver results but wasn’t exactly what it said it was. That was the blood testing startup, Theranos, that soared to a valuation of $9 billion and then crashed and burned after its effectiveness was called into question.

“Theranos proved there was clear interest in the space, it would have been a great company if it worked,” Tandon said in an interview with Bloomberg. “Now, investors say they need proof before we can raise money.”

Athelas has published papers on the accuracy of its data and has also been FDA-approved as a device to image diagnostics. Before it can be sold over the counter, it will have to receive further approval stating that it’s as accurate as a standard test in lab conditions…
"Theranos?" Remember them? I've had my considerable irascible sport with them here.

Athelas is specifically pitching the utility of their product for oncology blood assay monitoring.

Interesting. My daughter has to run over to Kaiser today for her routine blood draw in advance of her upcoming every-other-week chemo infusion. I'm not sure her oncologist (who is also a hematologist) would be comfortable leaning on DTC single-drop-of-blood assay alternatives.

I think the Athelas people will be at the upcoming Health 2.0 Conference, and we will be hooking up for discussion. I'll have to look back through the Conference agenda to see whether any Watson peeps will be there.

Also, in the wake of my recent cardiology workup, I have to wonder about apps like that now marketed DTC by AliveCor:
Meet Kardia Mobile.
Your personal EKG.

Take a medical-grade EKG in just 30 seconds. Results are delivered right to your smartphone. Now you can know anytime, anywhere if your heart rhythm is normal, or if atrial fibrillation is detected.

Is this widely useful or just another 'Worried Well" toy? I showed this pitch to my cardiologist. He was dubious -- with respect to my case, that is.

On "big data" and "Big Silicon Valley firms." New book release on Sept 12th. Saw a number of articles with and by the author.
"…More than any previous coterie of corporations, the tech monopolies aspire to mold humanity into their desired image of it. They think they have the opportunity to complete the long merger between man and machine - to redirect the trajectory of human evolution. How do I know this? In annual addresses and town hall meetings, the Founding Fathers of these companies often make big, bold pronouncements about human nature - a view that they intend for the rest of us to adhere to. Page thinks the human body amounts to a basic piece of code: "Your program algorithms aren't that complicated," he says. And if humans function like computers, why not hasten the day we become fully cyborg? To take another grand theory, Facebook chief Mark Zuckerberg has exclaimed his desire to liberate humanity from phoniness, to end the dishonesty of secrets.

"The days of you having a different image for your work friends or co-workers and for the other people you know are probably coming to an end pretty quickly," he has said. "Having two identities for yourself is an example of a lack of integrity." Of course, that's both an expression of idealism and an elaborate justification for Facebook's business model…"

Looks interesting. I will be reading and reviewing it. I had a run at some of his issues in 2015. See "The old internet of data, the new internet of things and "Big Data," and the evolving internet of YOU."


Finished the Franklin Foer book. Riveting read. Read it "cover to cover" pretty much straight through in one day. Contextual review coming, stay tuned.



More to come...

No comments:

Post a Comment