The Algorithm Will See You Now

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
An IR fellow offered his (her?) insights on this topic

). Pretty good read 🙂
 
An IR fellow offered his (her?) insights on this topic

). Pretty good read 🙂



Fascinating study. Obviously there is a ton of research on building the better radiologist or dermatologist or diagnostician. I just read this on radiology. Silicon Valley types see this as a software problem, but what they don't get is exactly what that Reddit poster does: computers will just change the problems of false negatives and false positives, will introduce legal challenges and will require physician oversight.

If I were graduating from med school 2017, I'd be scared to go into rads. But no more so than anything else.
 
The problem with Silicon Valley is they are disconnected from medicine. They don't understand the nuances and the full scope. To them, it's just another problem to tackle. Sorry, but this isn't building a Facebook or Snapchat. Medicine is a highly regulated field, and for good reason. If you mess up a consumer tech startup, you lose some money. You mess up a healthcare startup, people can die.

The entire culture right now in Silicon Valley is annoyingly ignorant and arrogant, a very dangerous combination. The people up there feel like they can do no wrong.
 
The problem with Silicon Valley is they are disconnected from medicine. They don't understand the nuances and the full scope. To them, it's just another problem to tackle. Sorry, but this isn't building a Facebook or Snapchat. Medicine is a highly regulated field, and for good reason. If you mess up a consumer tech startup, you lose some money. You mess up a healthcare startup, people can die.

The entire culture right now in Silicon Valley is annoyingly ignorant and arrogant, a very dangerous combination. The people up there feel like they can do no wrong.

I wouldn't make blanket statements like this; there are quite a few MDs who also run startups. Here are just a few I've met/heard about:

Dan Riskin MD, trauma surgeon, founder of Health Fidelity
Darren Schulte MD, CEO of Apixio
Rusty Hoffman, MD, interventional radiologist, Co-founder of Grand Rounds
Greg Albers, MD, co-founder of iSchemaView RAPID

What do all of the above have in common? They all use machine learning and were founded by MDs.
 
I wouldn't make blanket statements like this; there are quite a few MDs who also run startups. Here are just a few I've met/heard about:

Dan Riskin MD, trauma surgeon, founder of Health Fidelity
Darren Schulte MD, CEO of Apixio
Rusty Hoffman, MD, interventional radiologist, Co-founder of Grand Rounds
Greg Albers, MD, co-founder of iSchemaView RAPID

What do all of the above have in common? They all use machine learning and were founded by MDs.

And I bet they are modest. It is only the idiot know nothings who say that computers will replace ____ field of medicine/doctors in 5 years.
 
As a student considering radiology... phew! A lotta the AI hype got me scared for a sec there.
 
As a student considering radiology... phew! A lotta the AI hype got me scared for a sec there.
Same here, but then I thought, it took hospitals years to switch to EMR, pharmacists are still around, EKGs and EEGs are still being read by specialists. If anything of this magnitude is going to happen, it probably won't be in our career lifetime.
 
Until an AI can be sued and have to carry malpractice insurance, none of us have anything to worry about.

The US healthcare system loves to place liability somewhere
 
Last edited:
bump bump so fellow students don't get dissuaded by the false notion of AI replacing radiologists! :laugh:

edit: specifically talking about this: Rads to Anesthesia?

edit2: Lelz. I'mma just post it for y'all. Keep in mind the guy whose reddit comments i'm posting is both an interventional radiologist in MGH and an engineer who has consulted for IBM Watson (as in he actually knows what's what):

 
Last edited:
Top