Slashdot Mirror
CT Scan "Reset Error" Gives 206 Patients Radiation Overdose
jeffb (2.718) writes "As the LA Times reports, 206 patients receiving CT scans at Cedar Sinai hospital received up to eight times the X-ray exposure doctors intended. (The FDA alert gives details about the doses involved.) A misunderstanding over an 'embedded default setting' appears to have led to the error, which occurred when the hospital 'began using a new protocol for a specialized type of scan used to diagnose strokes. Doctors believed it would provide them more useful data to analyze disruptions in the flow of blood to brain tissue.' Human-computer interaction classes from the late 1980s onward have pounded home the lesson of the Therac-25, the usability issues of which led to multiple deaths. Will we ever learn enough to make these errors truly uncommittable?"
Nonsense, you're still asking users to trust the default, only now the default useless because you lowered it to something 'safe'.
It sounds like the error was simply that the doctors didn't reset the machine after cranking up the dosage for a very particular situation. To expect that the machine would forget this setting, and go back to what it had been doing for years previously, is ridiculous, and goes to show how dangerous what you're suggesting is. They have obviously been trained to trust the machine to fix their mistakes.
This is a lesson in user interface design, though it goes against the dominant ideology, which wants to wrap everything in kid gloves and training wheels, hide complexity and make as many decisions for the user as possible. The user should have a total understanding of the thing they are manipulating. That doesn't mean they need to know physics or software architecture, it only means they share the same domain model in their mind as the author of the software.
These doctors treated this machine like a magic button that takes pictures. This is what happens when a user interface lies to its users, fooling them into thinking something is simple when it is in fact complex and dangerous.
There is a lot of regulatory oversight of these machines. Even relatively benign devices like ultrasound have a lot of process to ensure safety, and the extremely deadly stuff like radiation therapy seems like 99% process to 1% engineering (and you thought those TPS cover sheets were annoying). Many bugs, even simple GUI mistakes, can escalate throughout the company. Which is sort of what you expect with potentially dangerous equipment.
But at the end of the day, you have to beware of the user. You even have to design around the user, to ensure the user does not make a mistake. Allowing the user to make a mistake can often be considered a bug. Using radiation therapy machines is a constant stream of "are you sure?" messages. I heard one story where a machine had two widely separated buttons, both of which had to be held down for safety reasons to perform a certain operation, and then someone discovered a customer that kept a weight near the machine for use on the second button rather than having a colleague help.
This is absolutely one area where you don't want the end user taking control and changing the software or rewiring the machine. You'll never see a GPL license come with these machines.
Your problem appears to be that you think the handbrake is suitable for use while moving. It was designed to prevent the car from starting to move in the first place.
Citation?
I know you Americans don't use the damn thing anyway
I know you whatever-you-are obviously don't read sigs.