PulseOx (SpO2) measurement and display with QMP

Rayman · 2013-03-24 16:58

Just getting started on our MicroMedic contest stuff...

Thought I'd share some progress with a PulseOx (SpO2) sensor...

My doctor friend sent me Nellcor SpO2 sensor, which I think is the same one that
comes in the Micromedic kits.

The resistive touchscreen controller of the QMP (Quick Media Player) happens to have a few spare
analog inputs. I'm using one of them to read it 12-bit analog data from the SpO2 sensor.

Started with the Parallax code and added some extra stuff and now it seems to work pretty well:

Here's the circuit:

Here's the code:
QMP_SpO2_DEMO2 - Archive [Date 2013.03.24 Time 19.50].zip

Ken Gracey · 2013-03-24 20:00

Ray, are you able to turn the SPO2 data into a calibrated saturation output? I think you might be the first to have achieved this step.

Hopefully you're headed into a DVI display project. This could do well in the contest!

Rayman · 2013-03-25 02:55

Wasn't thinking about displaying a calibrated reading, but that's a good idea.
I think I have the info, just need a calibration factor...

Was reading some things that Google turned up, but not sure how to pick the calibration factor yet...

David Betz · 2013-03-25 06:56

Rayman wrote: »

Just getting started on our MicroMedic contest stuff...

Thought I'd share some progress with a PulseOx (SpO2) sensor...

My doctor friend sent me Nellcor SpO2 sensor, which I think is the same one that
comes in the Micromedic kits.

The resistive touchscreen controller of the QMP (Quick Media Player) happens to have a few spare
analog inputs. I'm using one of them to read it 12-bit analog data from the SpO2 sensor.

Started with the Parallax code and added some extra stuff and now it seems to work pretty well:

Here's the circuit:

Here's the code:
QMP_SpO2_DEMO2 - Archive [Date 2013.03.24 Time 19.50].zip

Hi Ray,

Where did you get your SpO2 sensor? Also, the one pictured as part of the MicroMedic contest kit looks like it's intended for only a single use. Do you know where you can get the clip on ones that can be used multiple times?

Thanks,
David

Edit: Duh. I just re-read your message and you say you got yours from a friend. Anyone know a place to buy these individually?

Rayman · 2013-03-25 08:55

AFAIK, the one I'm using is identical to the one in the MicroMedic kit...

I have no idea where to get different ones.

But, I'm testing without taking it out of it's clear plastic wrapper.
It seems to work this way, so I can test many times with the same unit.

But, you do need to keep it still when taking a measurement...

David Betz · 2013-03-25 10:07

Rayman wrote: »

AFAIK, the one I'm using is identical to the one in the MicroMedic kit...

I have no idea where to get different ones.

But, I'm testing without taking it out of it's clear plastic wrapper.
It seems to work this way, so I can test many times with the same unit.

But, you do need to keep it still when taking a measurement...

Thanks. I'll look on the web for someone who sells these sensors. I think Parallax only makes them available as part of the Micromedic kits but not separately.

David Betz · 2013-03-25 10:57

David Betz wrote: »

Thanks. I'll look on the web for someone who sells these sensors. I think Parallax only makes them available as part of the Micromedic kits but not separately.

I found a place selling them through amazon.com for a little over $2 each and was about to order 5 when I noticed that the shipping for 5 sensors was going to be over $42. Ugh...

Invent-O-Doc · 2013-03-25 18:55

There isn't a difference in the LEDs between the clip on and the portable. One just is less durable because of the band aid material they are constructed with. You can try rough association with a cheap wal-mart pulse ox on the same person, but getting lower oxygen level readings will be a problem.

Invent-O-Doc · 2013-03-25 18:56

Oh yes, Cool demo!

dgately · 2013-03-25 20:58

Rayman wrote: »

I have no idea where to get different ones.

There are several Nellcor non-disposable units on eBay... Not sure if they have the same pinouts as the disposable type.

http://www.ebay.com/sch/i.html?_trksid=p2050601.m570.l1313&_nkw=UPS+Nellcor+DS-100A+Adult+Finger+Clip+Spo2+Sensor+&_sacat=0&_from=R40

dgately

David Betz · 2013-03-25 21:20

dgately wrote: »

There are several Nellcor non-disposable units on eBay... Not sure if they have the same pinouts as the disposable type.

http://www.ebay.com/sch/i.html?_trksid=p2050601.m570.l1313&_nkw=UPS+Nellcor+DS-100A+Adult+Finger+Clip+Spo2+Sensor+&_sacat=0&_from=R40

dgately

Thanks! I just ordered one. I'll let you know how it works out once it arrives from China.

Rayman · 2013-03-26 05:42

I just ordered one too... It is kindof a pain to hold the thing steady while playing with this... This looks like a better way... For $14, it's seems like a good deal.

Rayman · 2013-03-26 17:58

Just tried reducing the value of my filter capacitor a few orders of magnitude, from 100 nF to 100 pF...

The signal got bigger and noisier looking, maybe as you'd expect. Had to add a 5X division factor to get it from clipping on the screen...

This looks a little noisy to me, so I might try a bit bigger value next:

kbash · 2013-03-26 22:55

Rayman,

I just did the design of a "Slightly different" type of sensor for pulse-ox systems. ( I'm not allowed to say much about it ) . In the process I took apart several of the different sensors. By far, the best LED/sensor pairs were the ones in the permanent finger clip-on units. One thing to keep in mind... the data that you are working with from these things is right on the edge of "Doable" There's a lot of noise of many different types to contend with. If you take one of the "disposable" sensors apart, even there, you will see that they have a fine copper mesh noise (ground) shield around everything. Be careful to keep all the grounds, shields, etc, intact as the noise can easily overshadow the data.

http://cinc.mit.edu/current/preprints/340.pdf

That's a link to an article that I found helpful in some of my research.

I've also attached a write-up I did early in the project. I had to delete the job specific information and photos, but some of it might be helpful. You're probably beyond the point where you'll learn anything from it, but you might enjoy seeing some of my "Odd" data.

Good luck with this... I believe the Prop has the horsepower to do this and do it well, however, even with a reasonably high end (Masimo) pulse-oximeter I used for my baseline studies, ... I still got some really weird data from time to time if you get any valid data at all... you're doing good!

Ken

Rayman · 2013-03-27 02:50

Thanks kbash, this gives me a lot of ideas... Plus, I didn't know there was a reflective version, that's interesting. Funny how the table gives a reading though...

One thing I see is the first pdf is that they dynamically control the LED output. That's something I think I can easily add here....

skylight · 2013-03-27 05:40

Good luck with the project Rayman, it looks good so far

PulseOx (SpO2) measurement and display with QMP

Comments