@Kim-Locke

Yup, registered. Thanks for sharing!

@Chris-Johnston

So, many decades ago (I'm not saying how many), I did a summer stint with a company that produced script pads. These were those paper prescription pads that doctors would write instructions for the pharmacy to fill your prescription: drug name, dose, how to take, etc. They would write by hand, tear off the top page and hand you the paper. You left the appointment, went to your pharmacy, and handed over the piece of paper. Then the pharmacist would fill your prescription; based on what was written on that piece of paper.

These paper pads were created where every 6 pages were scripts (Patient_Name at the top; doctor's signature line at the bottom) and the 7th page would be a glossy advertisement for a drug. The pharmaceutical companies would fund the production of the pads and provide the ads to be used. The company I worked for would design the script and send the designs to the printers.

To this day, I still remember watching my doctor fill in a script, make a mistake, scratch out, tear off the page and throw it in the garbage... hit an ad, tear-off and throw-in-garbage... and carry on to the 3rd page to properly write out my script. At the time, I found that process somewhat interesting.

I see parallels between these 2 workflows... other than nowadays, the CDS software would probably delay the ads before a manual override-to-clear would activate. (It's like those irritating ads that show up when you're surfing the net. You can never get rid of them fast enough.)

Unfortunately, tear-off and throw-in-garbage will not work here.

@Jenna-Kedy-0

That's great Jenna - maybe I'll bump into you in some sessions 🙂

Wondering if anyone else is attending the symposium today?

@Alies-Maybee

Thanks Alies, I’ve signed up 🙂

@Kim-Locke @Debra-Turnbull

Yet another reason why allowing human-in-the-loop oversight to be tapered down, before the technology has matured sufficiently to provide a level of consistently and quantifiably reliable output is unacceptable.

Humans may be flawed as oversight, but better a flawed human than no oversight at all.

@Chris-Johnston
I loved the simple, down-to-earth explanation. It got techno heavy towards the end, but the approach was basic enough to be easily understood.

I'm going to watch this one again. Well worth the time!

@Chris-Johnston 🤣

Don't you just hate it when that happens...!
Thanks Chris!

The Big Picture: State of AI in Medicine :
Session 1

@Alies-Maybee said in Video on the impact of AI on society:

What can we do?

We are the ones left behind in the villages:

pushback! literacy and learnings get involved

Global warming is happening, and Bezzos, Musk, and Zuckerberg get richer.

Today's Brookings presentation brought up an interesting idea - make product Safety a requirement; build governance around safety. Once regulatory requirement specifies safety, the market shifts from profits; towards safety... and maybe social justice?

@Jenna-Kedy-0 I'm hopeful that we will work out the bad things and be able to work more on the good things. When I think about AI, I think about privacy, consent, and trust. ...and secondly, digital literacy. These things need to be ironed out as laws or regulations before we should be letting children or even young adults use it.

Press the red 'play' button on the video. It will take you live.

@Jenna-Kedy-0

Very well articlulated points.

I have yet to see what transparency will look like for these systems. They are being blindly deployed in order to alleviate clinician pressure. The error-detection piece is still missing.

I honestly did not see the lack-of-informed-consent as being a problem, seeing as I had not experienced that. The more that I learn about these things, the more convinced I become about saying "no" the next time that I am asked to consent.

@Kim-Locke Thank you for sharing! As someone who basically grew up in healthcare spaces because of chronic illness, I think the language we use around AI really matters. I spend a lot of time thinking about trust and what actually helps patients like me feel safe so when I read articles about AI and “emotions,” my brain instantly gets itchy because words matter SO much. I still think we use super anthropomorphic language when talking about AI. Words like “thinking” make people picture something deeply human, when really these systems are processing patterns. It is fascinating! Personally I don’t think it is the same thing as human emotion.
At the same time, I am hopeful about AI in healthcare when it’s used responsibly. I’d honestly much rather see conversations focused on reducing provider burnout and creating more human-centered healthcare experiences because after spending so much of my life as “the patient,” I know firsthand how exhausting healthcare can feel. If AI can help lighten some of that load while still keeping humanity at the center; that’s the kind of AI conversation I want more of!

@Sandra-Holdsworth

I'm not sure what is wrong with your link, but for anyone trying to access, try this link:
Chatham-Kent OHT

@Chris-Johnston
Okay, the mechanism you're describing is a little disturbing.

There is already a parallel system in play in radiology (I believe). The detection skills of the radiologists showed significant deterioration over time - so much so that one of their oversight bodies made a recommendation to turn OFF the AI system for some length of time; in order to force the radiologist to use their brain.

Skills deterioration has been observed and measured. The term used to refer to this deterioration is now called "de-skilling".
(Now, if I can only remember where the heck I read that...)

@Kim-Locke

Yeah... this is in the U.S. Not sure what their privacy & consent practices are. But - something to watch out for if it comes up here.

@Kim-Locke

Very true - what still amazes me is that they've managed to infiltrate the very conservative infrastructure of medicine so rapidly.

We're well acquainted with how reluctant the medical profession are (both individually and as a body) to move forward in any respect. Faxes are one example, the delay of 10-15 years to get research evidence into practice is another.

Yet, entire health systems are jumping on board AI technologies that are neither mature nor sufficiently tested at scale. In just a handful of years, AI projects are consuming large chunks of healthcare budgets despite the alarming lack of regulation or evidence to support efficacy or safety. While other services are squeezed, reduced or defunded, with little or no public awareness or discussion. It's an incredible and concerning feat of manipulation.

@Jenna-Kedy-0 It makes me happy when there's pushback from local communities about having data centers built in their respective areas, and they manage to not have them built at all!

Data centers being built in areas that are experiencing drought just doesn't feel ethical or even right to me, and then all the problems with using a lot of water, the effects on the environment, effects on humans, and wildlife.

"BUT JOBS!" They only really hire very few people to work in these data centers, and they're only there to troubleshoot problems. Most of daily admin things can be done remotely or are automated already.

@Sandra-Holdsworth @Alies-Maybee

Thanks for checking Sandra! It’s pretty steep even for members, and membership ain’t cheap. I’m wondering if there’s a way to escalate the Patients Included initiative, particularly around affordable virtual options. It seems bizarre to me that many conferences are now offering virtual platforms as a bonus to in-person attendees and still won’t offer it to patients who can’t travel for health or financial reasons or both. 😞

@MichelleWithers

True.

Diagnostic AI will not fix the healthcare system. It is, after all, only a tool. A tool to assist clinicians get to an answer... quicker. This translates to less time the patient has to wait as a consequence of mis-diagnosis.

It will not solve long wait times. It will not solve clinician burnout (many are heading for the exits labelled "retirement"). However, if it a tool can alleviate some of a clinician's cognitive load, then I'm all for it. We - as patients / caregivers - have an active role in protecting out clinicians... else, we will lose them altogether.

I recently read an article arguing that Canada does not have a healthcare system. It has a jumbled mess of care delivery services; connected by Band-Aids... hmmm...