pulse oximeter

[pbajaj]

Hi,

I am a Biomedical Engineering student and I am working on my senior design project that deals with a CPR monitoring system. My part is to design a pulse oximeter for the team. I am trying to find a circuit diagram and a LabView code for the same. I found some designs but since I do not have much experience with circuits it was really difficult to follow for me. Kindly help me with it.

Thanks,

Piyush

[bburress]

Hey Piyush,
    Could you possibly tell us more about your application.  What type of circuit are you looking for?  How are you going to interface with other components in the system?  Where are the designs that you have found already?

[unclebump]

See this thread.

http://forums.ni.com/ni/board/message?board.id=170&message.id=141418&requireLogin=False

[pbajaj]

Hello Brian,

I am trying to find the pulse of the victim when the user is giving the CPR to the victim. So we have decided to use the pulse oximeter. I am trying to find the circuit diagram for a pulse oximeter. I have looked at the other posting here on the forum that is for the pulse ox. I have 2 LEDs and 2 photodiodes that I can use to make the pulse ox. We will have that connected to the SC 2345 from National Instruments which is connected to a computer. I have to write a LabView code too so that I can display the pulse and the oxygen saturation on the computer.

thanks,

Piyush

[bburress]

Hey Piyush,
    That other post seems to contain alot of details that should get you going.  From the LabVIEW side, all we need to do is read voltages from the DAQ card.  This should be simple, and is actually already done for you in the LabVIEW examples, as described in the other post.  Once you have the voltages, then you can scale them as you see fit and graph them accordingly.

[johnsold]

Piyush,

As a former paramedic as well as electrical engineer I suspect that you will have some difficulty getting good results. I recall a case where a young Emergency Medical Technician got very excited because the commercial pulse oximeter was indicating a pulse all the way to the hospital, where the patient was pronounced dead. The instrument had been responding to various stray light and motion artifacts. Further, the typical pulse oximeter measures variations in peripheral capillary blood flow which is likely almost non-existent under CPR conditions.

Lynn

[pbajaj]

Hi Lynn,

Could you suggest me to use some other device that can be built in the lab and which can be used to measure the pulse of the victime during CPR?

Thanks,

Piyush

[johnsold]

Piyush,

I had not thought about that until you asked.

I suspect that doppler ultrasound might be a possibility, perhaps monitoring the carotid or brachial arteries. The carotid would give better results but is likely to be inaccessible due to airway support and medication injection activities. Also, wrapping something around the neck to keep the transducer in place has its own problems. (Tape on one side only?) General motion artifact will be a major problem regardless of your approach.

Have you searched the literature to see what is being done? I have been away from emergency medicine for several years and am not up to date on the latest advances. Does your faculty advisor have any practical suggestions?

Lynn

[Japex]

Hi everyone,

I'm a new member of this forum, I have been trying to figure it out how to build a pulse oximeter using the correct theory. I've got a emitting drive circuit to switch both leds at two different times but I'm not obtaining the right signal from the output of a transimpedance circuit. Not sure if I'm using the right swtching times (60us ON in a period of 1.2ms) or if I'm not measuring the right output of the circuit.

I'm a bit confused about how those times work,  the whole circuit has several phases it starts with a emitting drive circuit which turns on and off each led, after that there is transimpedance circuit which captures the recieved light and transforms it into a voltage value, immediately after there is aBPF(Band Pass Filter) that has cut off frecuencies of (0.5-20)Hz which eliminates the DC level filters the noise signals from ambient light. Am I not supposed to get the PPG signal at this point?

Can somebody help me please??

Thanks

[johnsold]

Japex,

It is generally better to start a new thread rather than responding to an old one.  Especially since your question is about a similar topic, but considerably different in detail.

If you are switching the LED 0.06 ms ON and 1.14 ms OFF, then filtering the detector output with a 20 Hz band pass filter, I doubt that anything gets through the filter.  You are looking for a 5% duty cycle signal pulsed at 833 Hz after it passes through a 20 Hz filter.  The filter completely eliminates your signal.

What you need to do is to use something comparable to a sample and hold circuit or synchronous detection circuit between the photodetector/amplifier and the BPF.  This will reconstruct the pulse signal which can then be filtered.

Lynn