Most of what your doctor does, a robot can do better

MichaelB

Give me a museum and I'll fill it. (Picasso) Give me a forum ...
Site Team
Joined
Jan 31, 2008
Messages
40,738
Location
Chicagoland
Here's an interesting article, written by a medical student at University of Pennsylvania, about the potential for robots in medicine. Most of what your doctor does, a robot can do better – Quartz One snippet
My medical education focuses on recognizing pathologies under tight deadlines: efficiency and specialized pattern recognition. I'm hard pressed to make a case for my edge over a robot in these domains. And if I have no edge, I would be doing my patients a disservice to guard my job.
This is not a case of technology replacing physicians (IMHO). More like using technology to make highly trained individuals more productive and help reach more users in a cost effective manner. Much of this will never reach me, but my children and their families should benefit enormously, as might countless others living in under-served areas.
 
Interesting. I'm halfway through Daniel Kaneman's Thinking Fast and Slow. He touches on the subject that formulas can do better in many cases than humans can.
 
Not a new story. AI has advanced to a point that SW (I.e, computer, robot, etc) can diagnose better than average doctor can.

... and that's Dr. Robot to (for) you ;).
 
Maybe we will have our own emergency medical holograms some day. It is only a matter of time before symptoms are typed into a database and a list of possible diagnoses, probabilities and potential treatments are displayed back. It has been long overdue.
 
This is not a case of technology replacing physicians (IMHO). More like using technology to make highly trained individuals more productive and help reach more users in a cost effective manner. Much of this will never reach me, but my children and their families should benefit enormously, as might countless others living in under-served areas.

Just like everywhere else in the economy where automation and mechanization were supposed to simple make jobs easier for workers or free them up for other more complex tasks. Instead it's produced millions upon millions of pink slips as jobs are slashed and replaced *entirely* by robots and computers and the like.

It's just another example of how our society has to come to terms with economic and technological progress that makes labor more and more irrelevant with the passing of each year.
 
Robotic surgery has been around for a while and computerized decision support systems for a long while. Uptake has been spotty. A lot of humans are required to service the robots and DSS. There remain many tasks and skills that physicians do that are unlikely to be replaced by technology, though some of those can be done by other humans, e.g. Nurse Practitioners (who are not necessarily cheaper). As always, implementation lags behind the science, and given that it is complex and political, I wouldn't discount the medical profession just yet.
 
Interesting. I'm halfway through Daniel Kaneman's Thinking Fast and Slow. He touches on the subject that formulas can do better in many cases than humans can.

Great book, one of the researchers that the 'Freakonomic' team often references. I happened to be reading it at the time that I had a Doctors appointment, and I recc it to my Doc. Some very relevant sections for the medical field.

I did end up skimming some of it, IIRC, there were a lot of multiple examples of the same points, and after a while I felt like "OK, I got it, don't need another example". But some real thought provoking stuff. It's interesting to be a bit more aware of the times where you 'think fast' and maybe it really needs a 'think slow' approach.

I'll need to follow the original link in the thread, but I imagine that we don't need to wait another generation for this. I would think that much of what can be automated can be done today. I also expect that there will be some push-back from MDs who don't want to off-load their decision making to an algorithm or computer. I saw it among some engineers, and I would imagine some MDs would generally be even more resistant. That may be what slows down progress in this area.

IIRC, the APGAR score is an example of this, and was mentioned in some other thread, by Meadbh I think.

-ERD50
 
Just like everywhere else in the economy where automation and mechanization were supposed to simple make jobs easier for workers or free them up for other more complex tasks. Instead it's produced millions upon millions of pink slips as jobs are slashed and replaced *entirely* by robots and computers and the like.

It's just another example of how our society has to come to terms with economic and technological progress that makes labor more and more irrelevant with the passing of each year.

?

In 1900, ~ 38% of the population were farm workers, that is now ~ 4%. Do we have +34% unemployment due to that? Should we go back to manual farming? Who would be served by higher food prices?

The times they are a-changin'.

-ERD50
 
Sorry, I just don't believe this. Could a robot send me a $500 invoice for a thirty minute visit? I don't think so!
 
In 1900, ~ 38% of the population were farm workers, that is now ~ 4%. Do we have +34% unemployment due to that? Should we go back to manual farming? Who would be served by higher food prices?

Creative destruction no longer works the way it used to. It used to mean that jobs that were eliminated in some industries (by being obsolete) were replaced by even BETTER jobs in a new industry.

That's not been working well for the last decade or two. These days the jobs that are displaced are either eliminated entirely or offshored to some cheap labor haven overseas.
 
Sorry, I just don't believe this. Could a robot send me a $500 invoice for a thirty minute visit? I don't think so!


It may send you a $1000 invoice. Robots can be costly (better service - quality/no waiting, material, maintenance, insurance against law suits, union dues, .... ) :D.
 
Who programs the robot? I can see it now: all males over the age of 63 will have to take a statin. Oh, wait... We already have that with the new guidelines.

I see robot doctors the way I see robotics in cars and aircraft - an important safety feature to help overcome human error, but no substitute for human reasoning.
 
Last edited:
Creative destruction no longer works the way it used to. It used to mean that jobs that were eliminated in some industries (by being obsolete) were replaced by even BETTER jobs in a new industry.

That's not been working well for the last decade or two. These days the jobs that are displaced are either eliminated entirely or offshored to some cheap labor haven overseas.

I'd say that is largely a function of our 'flat earth', not a reflection on technology replacing jobs.

And for the person in a 3rd world country, that new job is replacing their old job with an 'even BETTER jobs in a new industry'. Bottom line, it's tougher for low-end workers in the US to compete.

...

I see robot doctors the way I see robotics in cars and aircraft - an important safety feature to help overcome human error, but no substitute for human reasoning.

In many cases, replacing human reasoning is the thing to do. I'd certainly trust a well programmed algorithm to detect potential drug interactions over counting on my Dr to remember that info, and hope he/she's on their game that day.

The key is - use computers where they add value, use humans where they add value.

-ERD50
 
Bottom line, it's tougher for low-end workers in the US to compete.

If by "lower end" you mean "the lower 80%", I would agree. And if 80% of the people have standards of living that keep creeping closer and closer to the Third World over time, well, I hate to think that's our only option.
 
Last edited:
If by "lower end" you mean "the lower 80%", I would agree. And if 80% of the people have standards of living that keep creeping closer and closer to the Third World over time, well, I hate to think that's our only option.

I guess you might better term in convergence over time. as the standards of living in the 3rd world are rising, for example famine is far less common. Also having skipped wireline phones wireless cell phones and smart phones are common. The question is is the third world rising faster than the 1st world is sinking? (In particular measure against the middle class in the 3rd world not the bottom strata.)
 
Here is one, I bought a sound enhancer, for bird watching/listening. It has Bluetooth tech, an android app that I can run the whole sound spectrum and adjust it basicly like being at the Audiologist on my tablet. Anyway flat world coming, the folks made it primarily for 3rd world countries where everyone has a smartphone, but not an Audiologist. Don't tell but I use it also as a hearing aid.
 
If by "lower end" you mean "the lower 80%", I would agree. And if 80% of the people have standards of living that keep creeping closer and closer to the Third World over time, well, I hate to think that's our only option.

I guess you might better term in convergence over time. as the standards of living in the 3rd world are rising, for example famine is far less common. Also having skipped wireline phones wireless cell phones and smart phones are common. The question is is the third world rising faster than the 1st world is sinking? (In particular measure against the middle class in the 3rd world not the bottom strata.)

This is getting pretty far off-topic from the thread, but I'll say that I agree with meierlde. The third world is creeping closer and closer to us. They don't have clean drinking water or access or basic health care, while most of our our 'poor' have that plus color TV, cable, a game console, AC and heat as needed.

And it isn't our only option because it is not a zero sum game, we just need to come up with technology solutions. I think there are many promising ones, but predicting the future is hard. Maybe for another thread?

-ERD50
 
Maybe we will have our own emergency medical holograms some day. It is only a matter of time before symptoms are typed into a database and a list of possible diagnoses, probabilities and potential treatments are displayed back. It has been long overdue.

It's already here, albeit in crude form.

Symptom Checker from WebMD. Check Your Medical Symptoms.

I used this to diagnose DW's gall stone eruption and took her to the emergency room. The Web MD diagnosis was correct and her gall bladder was removed the next day.
 
It's already here, albeit in crude form.

Symptom Checker from WebMD. Check Your Medical Symptoms.

I used this to diagnose DW's gall stone eruption and took her to the emergency room. The Web MD diagnosis was correct and her gall bladder was removed the next day.

My MD uses something similar, don't think he likes to. The fine conglomerate that owns the practice does it for risk management. He's not afraid to go old school, when 30 years of practice says different.
MRG
 
I want my very own Tricorder.
 
It sounds too much like putting a Merck Manual and PDR on computer with a program to cross reference. It either runs the risk of making the doctor lazy (doesn't look for less obvious symptoms) or the patient using it to diagnose themselves (and not aware of less obvious symptoms). If used as a supplement for a conscientious and knowledgeable doctor it might be useful. However, that could be a big "IF".

Cheers!
 
I've been retired for a year and a half, but waaay back when I was working, the technology wasn't there yet. The system bombarded you with so much bad and useless information, that you got numb and ignored it. After the 500th warning about prescribing Imitrex for someone with a history of migraine (actually, there's no other reason to prescribe it) or metformin for someone with a history of diabetes, it just got tuned out.

A computer is never going to notice the handprint on the little old lady's upper arm or the averted gaze when asking about Tylenol for knee arthritis when the patient really wants to discuss the lump he just found on his privates.
 
the averted gaze when asking about Tylenol for knee arthritis when the patient really wants to discuss the lump he just found on his privates.
Mmm, except that there seems to be evidence to show that he's happier to discuss that lump (and a whole lot of other stuff) with a computer than with a human.
 
Back
Top Bottom