Mind you, human doctors kill and injure as well. But it was fascinating to note the recent story in The Register about robotic surgeons.
I don't normally use the word "cockups" so let me quote directly from the U.K. article: "A team of university eggheads have counted up the number of medical cockups in America reported to the US Food and Drug Administration (FDA) from 2000 to 2013, and found there were 144 deaths during robot-assisted surgery, 1,391 injuries, and 8,061 counts of device malfunctions."
Though that might sound ghastly, consider that 1.7 million robo-operations were carried out between 2007 and 2013. Still, experts want more safeguards.
Two deaths and 52 injuries were caused when the mechanical surgeon spontaneously powered down mid-operation (not good) or made an incorrect movement (also not good). In another 10.5 per cent of recorded malfunctions, electrical sparks burned patients, resulting in 193 injuries.
A major problem, surprisingly, was that one death and 119 injuries were caused by pieces of the robot falling off into the patient, requiring a human surgical team to intervene and retrieve the broken hardware. I guess it's nice to know that there are human backups standing by.
The most dangerous kind of robot surgery is cardiothoracic and head and neck surgeries (6.4 per cent and 19.7 per cent of adverse results respectively), compared to 1.4 per cent and 1.9 per cent for gynecology and urology operations.
Unfortunately, the reports on precise causes are incomplete and the vast majority of deaths and injuries are simply listed as "malfunction," which could mean either the mechanical surgery unit failed or the operator screwed up.
I sent the article to my friend Ed Walters, the CEO of Fastcase, who also teaches robotic law at Georgetown University Law Center. He wrote back (I especially liked #4):
"The key questions are:
- Is the error rate higher or lower than a comparable human surgeon?
- If the error rate is lower than a human surgeon, should there be liability for machine surgical errors?
- Where there is liability, who is liable? Hardware manufacturer, software manufacturer, owner of the machine (a hospital), or operator of the machine?
- Is this a suit for negligence, or products liability? For negligence, can we really even ask what a reasonably prudent robot would do in the same situation?
In any event, I think the bots will be buggy for many years — but in time, they will be demonstrably better than human surgeons for many kinds of operations. Not perfect, and there will be errors, but fewer than by humans."
A man of faith, my friend Ed. And I suppose, from an e-discovery perspective, robots have already taught us that they outperform humans. I just want to be under anesthesia before a robotic doctor begins cutting. Call me old-fashioned, but I like shaking my (human) doctor's hand before and after surgery.
E-mail: email@example.com Phone: 703-359-0700
Digital Forensics/Information Security/Information Technology