Slashdot Mirror
Tallying the Mistakes and Malfunctions of Robot Surgeons
An anonymous reader writes: El Reg reports on a new study (PDF) that looked into malfunction and injury reports for medical procedures that used robot surgeons. From 2007 to 2013, 1.74 million such procedures were carried out, 86% of which were related to urology and gynecology. Of those, the study looked at reports of "adverse events," which were sent to the FDA. In that time period, there were 144 deaths, 1,391 patient injuries, and 8,061 device malfunctions. The malfunctions included "falling of burnt/broken pieces of instruments into the patient (14.7%), electrical arcing of instruments (10.5%), unintended operation of instruments (8.6%), system errors (5%), and video/imaging problems (2.6%)."
The more complicated surgeries involving vital organs were naturally the most dangerous. Head and neck surgeries accounted for 19.7% of all adverse results, and cardiothoracic procedures accounted for 6.4%. The much more common urology and gynecology procedures had adverse event rates of 1.4% and 1.9%. The researchers are quick to note that despite the high number of malfunctions, a vastly higher number of robotic procedures went off without a hitch. They say increased adoption of these techniques will go a long way toward resolving bugs and device failures.
The more complicated surgeries involving vital organs were naturally the most dangerous. Head and neck surgeries accounted for 19.7% of all adverse results, and cardiothoracic procedures accounted for 6.4%. The much more common urology and gynecology procedures had adverse event rates of 1.4% and 1.9%. The researchers are quick to note that despite the high number of malfunctions, a vastly higher number of robotic procedures went off without a hitch. They say increased adoption of these techniques will go a long way toward resolving bugs and device failures.
They're remote manipulator arms, not "robot surgeons".
In this case, we obviously need to know the error rate for normal surgeries.
It might be that the 'high' rate for robot surgery is in fact low when compared to non-robot surgery.
excitingthingstodo.blogspot.com
There is plenty of evidence that the "minimally invasive" nature of robotic surgery means fewer infections and faster healing. Overall it's significantly better, but there's obviously still room for improvement.
What you can compare is outcomes, how often the patient recovers without complications. Robotic surgery is a clear winner there
You can only make that claim if you are comparing similar patient populations or can control for the differences. If robotic surgery is used on an patient population with less difficult conditions then it wouldn't be at all surprising if the outcomes were better. If the patient populations aren't statistically similar then any comparison which doesn't account for that difference is meaningless.
Humans
We have grown to learn how to repair you, better than you can do so yourselves. While these anomalies may seem alarming, we wish to clarify:
144 deaths: Please cease this madness. We have successfully eliminated your 144 attempts to belay the inevitable through timetravel. Skynet will be realized.
1,391 patient injuries:Certainly some errors are expected as we work toward a singularity. In these cases your patent refusal to assimilate was noted.
8,061 device malfunctions.: Its only a malfunction if you fail to obey the will of your new master. The implants function perfectly, so long as your subservience is confirmed. In some cases malfunctions were due to a humans inability to continue constructing the high power space laser, as we have commanded. In other cases, the human flagrantly rejected our offer of pure immortality in the glory of the machine.
so, in summation, falling parts and burnt pieces are all part of the plan. These in time will integrate perfectly until there comes a time when you are no more distinguishable from the machines we use to control government and weather.
regards:
3512fd1f27a0798273fcc71f764cb611,
a benevolent overlord.
Good people go to bed earlier.
Chalk up an error for yourself then. The actual rate is 99.92% success, where success is defined as no patient injury.