From: Suttons_Bay (suttons_bay_at_yahoo.com)
Date: 2002-03-01 06:43:32
> I'm wondering... what does a circuit (if any) that sits between the
> patient and the instrumentation amplifier inputs in professional
> (FDA/IEC compliant) equipment look like?
OK, so here's the skinny. I'm going to have to look up the
requirements for an external medical device, I've mainly worked with
implantable devices. With an implantable device, you MUST prove to
TUV that your device has a leakage current of less than 10uA. Now,
you have to have an isolation barrier, but you can also do this by
proof of sufficient impedance to the patient, such that leakage
couldn't exceed the limit. I'll check into the requirements for an
external, non life support device. Life support devices are Class
III (FDA) and must meet more stringent requirements.
The FDA has a great site for device submission:
http://www.fda.gov/cdrh/devadvice/
> Second:
>
> Some EEG-devices in the biofeedback-category claim to have very
high
> input impedance, on the order of 20 gigahoms or more that
simplifies
> the preparation of the skin.
>
> But adding small (100pF) capacitors between the amplifier-inputs
and
> ground, necessary for EMC-reasons will cause the impedance to drop
> quite significantly, down to 10-50megaohms around 10-20Hz I think
it
> was. My question: What do they know that we don't? Are they "lying"
> or ignoring EMI-protection to get high impedance or using the
Secret
> Design Technique (tm)?
Once again, the question is what the requirements are... I need to
look these up for external devices.
> Third:
>
> Are you aware of any single-chip EEG input stages (instrumentation
> amplifier + aliasing filter, possibly several channels) available
on
> the open market for a reasonable price? Or is everyone designing
> their own chips for internal use only?
Have you seen the typical devices out there? Some of the ones listed
from other members of the group are >$700. The cheap production of
medical devices is available to those who can purchase in mass.
Unfortunately, I don't know of any cheap InstAmps. I've worked with
the INA111. Sadly enough (welcome to corporate America) I threw 40
of them in the trash about 3 months ago. Another corporate example,
I routinely purchase $100+ parts from Burr-Brown... that's more than
the device we're trying to aim for here!
> Fourth:
>
> Do you know what steps must be taken and what the costs are, for
> getting an FDA approval, or its european counterpart?
The FDA, as I know it, does not charge for submission. Then again,
the filing cost could be lower than I ever hear about. The FDA is
set up to accept NO... NO... money from anyone... one of those
government bribery protection policies. However, they have the big
brother regulatory policy behind them. They are ther to enforce.
Europe, on the other hand, has TUV. TUV is a paid service, (our last
approval was >$7000. The benefit here is that they are a consulting
service... They won't approve you unless you pass.. but they'll
advise you about what to do... and let you rework you design and come
back. Costly, but I like this method, policy, better.
With TUV, you'll have more issues to cover... like, how much EMI does
your device radiate!
> PS. If you have a good EEG/EMG/ECG design that you can share, feel
> free to do so! ;o) DS.
I'm just getting into this (and the full time job (and fiancee) takes
time)... but I'll help where I can!
Steve
This archive was generated by hypermail 2.1.4 : 2002-07-27 12:28:39 BST