5 ms·

This is a great example of where we need to get humans out of the equation when (if) a machine is conclusively proven to perform consistently better. It was ju

This is a great example of where we need to get humans out of the equation when (if) a machine is conclusively proven to perform consistently better. It was justified (cost wise) to replace many human labourers on auto assembly since machines don't get tired, need breaks, have off days. It could certainly be argued it is even more important in the field of health care (reduce costs and improve outcomes) for all forms of image scanning.

assblaster 8y ago
Machines will never replace dermatologists, machines will only make them more efficient.
- djg123 8y ago
  Well... Never say never. If you said machines would not replace dermatologists in the next 1 or 2 decades, I would agree with you. After that, all bets are off.
- hhs 8y ago
  Interesting point. The authors also seem to note this in their conclusion: “Our findings suggest that artificial intelligence algorithms may successfully assist dermatologists with melanoma detection in clinical practice which needs to be carefully evaluated in prospective trials.” They use the verb assist. I think this will be the case, tools like this will be made available to board certified dermatologists to enhance their work. It won’t replace them anytime soon. This field continues to be among the most competitive for physicians in the US. The barriers to entry are not only high, the number of residency spots get smaller each year when compared to total applicants. If interested, here’s some stats on med school rankings and derm match results: https://escholarship.org/uc/item/59p3z80r https://escholarship.org/uc/item/59p3z80r. But over many years I wonder what will happen. Will there be paradigm shifts that make us rethink about all these specialities and subspecialties? Should we combine them or do something else?
- feral 8y ago
  If you make a dermatologist 5x more efficient, don't you replace 80% of them? Or even better, allow them to spend more time on the hardest cases. And allow people with no access to a dermatologist now, access to a machine almost as good?
  - sgt101 8y ago
    Actually when you make a knowledge worker / service worker in a business processes 5* more efficient the experience is that they spend 400% more time on the cases that they have left. These are the cases that you can't automate and that before automation you couldn't service properly/economically. Now you can, so the workers do.
- rafiki6 8y ago
  What is your efficiency metric here? Number of patients serviced? If so, it's not that simple. Imaging specialties might end up becoming more "efficient" (such as radiology), but medical patient data which is used to diagnose and triage is some of the worst there is (source: I worked with it). The only positive outcome I see from ML for the medical field for the foreseeable future might be to reduce the number of misdiagnoses if all things go very well.
srfilipek 8y ago
> This is a great example of where we need to get humans out of the equation now that a machine performs consistently better. No, it's not. The system is an image classifier with a HUGE false positive rate (and false negative rate). When false positive rates exceed actual incident rates in the population (or far exceed by orders of magnitude, in this case), then it's practically worthless. This is something that the medical community just does not get about statistics (edit: I'm not making a wild generalization here, there are articles out there about this issue). What this study actually shows is that an image-only diagnosis of melanoma sucks and should never be used. It doesn't matter if it out-performs doctors, because in either case the diagnosis is garbage.
- eljimmy 8y ago
  I think it could be useful as a tool for MDs. When booking an appointment with your doctor, you could take a photo and submit it. If the NN detects something abnormal, it could be useful in prioritizing your appointment or at least flagging it for reviewal by the MD.

assblaster 8y ago

Machines will never replace dermatologists, machines will only make them more efficient.

djg123 8y ago
Well... Never say never. If you said machines would not replace dermatologists in the next 1 or 2 decades, I would agree with you. After that, all bets are off.
hhs 8y ago
Interesting point. The authors also seem to note this in their conclusion: “Our findings suggest that artificial intelligence algorithms may successfully assist dermatologists with melanoma detection in clinical practice which needs to be carefully evaluated in prospective trials.” They use the verb assist. I think this will be the case, tools like this will be made available to board certified dermatologists to enhance their work. It won’t replace them anytime soon. This field continues to be among the most competitive for physicians in the US. The barriers to entry are not only high, the number of residency spots get smaller each year when compared to total applicants. If interested, here’s some stats on med school rankings and derm match results: https://escholarship.org/uc/item/59p3z80r https://escholarship.org/uc/item/59p3z80r. But over many years I wonder what will happen. Will there be paradigm shifts that make us rethink about all these specialities and subspecialties? Should we combine them or do something else?
feral 8y ago
If you make a dermatologist 5x more efficient, don't you replace 80% of them? Or even better, allow them to spend more time on the hardest cases. And allow people with no access to a dermatologist now, access to a machine almost as good?
- sgt101 8y ago
  Actually when you make a knowledge worker / service worker in a business processes 5* more efficient the experience is that they spend 400% more time on the cases that they have left. These are the cases that you can't automate and that before automation you couldn't service properly/economically. Now you can, so the workers do.
rafiki6 8y ago
What is your efficiency metric here? Number of patients serviced? If so, it's not that simple. Imaging specialties might end up becoming more "efficient" (such as radiology), but medical patient data which is used to diagnose and triage is some of the worst there is (source: I worked with it). The only positive outcome I see from ML for the medical field for the foreseeable future might be to reduce the number of misdiagnoses if all things go very well.

srfilipek 8y ago

> This is a great example of where we need to get humans out of the equation now that a machine performs consistently better. No, it's not. The system is an image classifier with a HUGE false positive rate (and false negative rate). When false positive rates exceed actual incident rates in the population (or far exceed by orders of magnitude, in this case), then it's practically worthless. This is something that the medical community just does not get about statistics (edit: I'm not making a wild generalization here, there are articles out there about this issue). What this study actually shows is that an image-only diagnosis of melanoma sucks and should never be used. It doesn't matter if it out-performs doctors, because in either case the diagnosis is garbage.

eljimmy 8y ago
I think it could be useful as a tool for MDs. When booking an appointment with your doctor, you could take a photo and submit it. If the NN detects something abnormal, it could be useful in prioritizing your appointment or at least flagging it for reviewal by the MD.