Yes risk control measures could impose further risks too. But something like a proven watchdog was missing. AD, I was mentioning about your earlier comment on 8051s.
Well, gentlemen, I am beginning to feel a bit tired, so probably should end this. (Besides, my wife/nurse/watchdog has limited me to an hour at a time on the computer)
sometimes simpler is more reliable. I think Murphy's law (one of them) says that the safety feature is the one that will fail first, creating a hazardous condition.
I was meaning, they have used very simple 8K memory 8051s for such a critical product. No offence, AD. I was worried because it did not have even watchdogs.
oh, dear. Rough rule of thumb is that few people can say "yes" with final authority, but lots of people can say "no" with final authority, and so they will do so just to exercise what little power they have.
They don;t know their physics. My feel that Bernrd Shaw's dictum applies in engieering as follows. Those who can, do. Those who cannot, work for regulatory bodies.
A concern I have with all specifications is the quality of the personnel that do the qualification Some of the guys I have seen in the regulatory bodies have actually challenged Ohm's Law.
I don;t know anyone, although I must admit that it would be an interesting read (as far as specifications can be intersting) just to see what concerns there are and how to approach them and apply them to other fields.
IEC60601-1-4 is the standard for programmable electrical medical systems - means the standard which oversees things particular to programmable ics such as mcus and fpgas.
My overdeveloped sense of guilt prevents me from doing any designs where I fear the consequnces. I don;t that there can be overdesign when it comes to these sort of things. I am worried I will miss something and someone will suffer. As it is my stuff is probably overdesigned if you see the disagreements in approach between Didier and myself on watchdogs.
After designing and test trials, when the first clinical evaluation happens, we feel like a husband of a woman in labour. So much filled with anxiety... After even so many years in the field, I am very nervous every time.
I had blogged on Microcontrollers for medicine at -----http://www.designspark.com/blog/microcontrollers-for-medicine---part-i ------ couple of years back.
This risk management is done both bottom - up and top - down. From user work flow,we work on Fault Tree Analysis. Then work on something related to each component called single fault condition, which is otherwise called Failure Modes and Effects Analysis (FMEA)
Risk management starts with risk analysis. We identify potential hazards at the time of designing. Also the resultant harmful effects. Then a mitigation plan worked out.
From last year a governmental regulatory body has been formed which does things similar to FDA. However, most manufacturers work on getting CE marking and comply to standard IEC standards and ISO13485
BB, trouble with SCA is its hard to diagnose. If your computer has a blue screen then you turn it off, it's hard to figure out why there was a blue screen. SCA is an electrical fault, and it could have been very temporary, even a timing race condition. Once the heart is in fib, no telling how it got there. Once it stops, there's no clue why
I didn't know SCA was such a danger. I have not seen much about it in the press. Getting AEDs more available in schools seems a no brainer- but I guess even no brainers take lots of work to move thru the school system...
Other devices I did app notes on are Blood Oximeters (using LEDs and photo sensors instead of drawing blood) and regular blood drawing devices (like blood glucose meters). The chemistry of measuring blood characteristics is facinating...
BB turns out SCA is a #1 killer of student athletes. We've seen cases as young as 4 at the Parent Heart Watch, a collection of parents who've lost children due to SCA and are advocating screening tests and AEDs in schools.
I worked with the MCU vendor to white the app note. They had some code fragments but needed an app note to explain the functionality, the peripheral devices, do some estimates for code size and performance- that kind of stuff. Learned alot doing it!
I don't have much to say on the topic other than I was speaking to my cardiologist and he didn't like engineers as patients since they demanded way too much precision than medicine could measure. For instance I am now monitoring my blood pressure twice a day. The machine takes 3 readings over a ~2 minute period and provides an average. I have looked at the 3 readings that make up the average and they can vary by about 20mmHg- all the way from good to unacceptable- how do you really know which reading to take when you go to the doctor and he only takes one.
Join me here on Friday at 11am Eastern US Time, 1600 GMT, for a live chat about medical electronics based on MCUs. What are the requirements? What are the opportunities?
To save this item to your list of favorite Microcontroller Central content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Tweet This
