Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Optical heartbeat monitor

Status
Not open for further replies.

fabiorov

Newbie level 1
Joined
Mar 13, 2005
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,298
heartbeat optical

Hi everybody, I'm new in this comunity.

I'm trying to make an optical heartbeat monitor, but I don't have much information of this.

Please If someone have documentation of this please help me
 

bobcat1

Advanced Member level 4
Joined
Jul 10, 2002
Messages
1,284
Helped
99
Reputation
198
Reaction score
34
Trophy points
1,328
Activity points
8,545
hemon pulse monitor

Hi

Here is some file and schematic hope you will find them usefull

All the best


Bobi
 

ezpcb

Banned
Joined
Mar 2, 2005
Messages
210
Helped
8
Reputation
16
Reaction score
3
Trophy points
1,298
Activity points
0
finger heartbeat monitor

Hi fabiorov
I'vd done such a project during my master's project. It's quite easy for just counting the heartbeat, a infrared optical coupler will do that. Just put one's finger between the emitter and receiver, then amplify the received signal, you'll see the heartbeating on the oscillscopes.

mike

---------------------------------
www.ezpcb.com
 

RegUser_2

Full Member level 2
Joined
Dec 24, 2001
Messages
125
Helped
11
Reputation
22
Reaction score
9
Trophy points
1,298
Activity points
1,507
heart beat monitor schematic

ezpcb said:
Hi fabiorov
I'vd done such a project during my master's project. It's quite easy for just counting the heartbeat, a infrared optical coupler will do that. Just put one's finger between the emitter and receiver, then amplify the received signal, you'll see the heartbeating on the oscillscopes.

mike

---------------------------------
www.ezpcb.com

There's more than that because the measured AC signal due to the change of optical thickness due to pulsatile arterial blood flow is a tiny fraction (couple of procents) of the total transmitance of the finger and have to be detected in presence of several fundamental noise sources:

1. There is a stray light interfering (if the photodiode don't have proper optical filter then for instance luminiscent source produces large AC signals which need to be filtered)

2. There are the so called "motion artefacts" due to the fact the change of the optical path due to mechanical interferences may and do garble severely the pulsatile fraction of the signal.

3. The frequency range of typical human pulse is in the range of say 50 to 150 bpm (say 0.5 to 2 Hz) what lays well within the 1/f noise range of a standard readout electronics and hence the 1/f noise (thermal drifts, popcorn noise, contact noise, ...) may and will severely decrease the dynamic range of the measurement.

In general some have to modulate the LED and use synchronous demodulation or double correlated sampling to get rid of the stray light and 1/f noise of the signal amplifiers, then use some 0.1 to 5 Hz bandpass filter to get rid of out of signal band interferences. The resulting pulstile AC signal should resemble a sow-tooth waveform.

The period of this signal equals to the pulse rate you want to measure. The cleanest way to measure this period is to digitize this signal and perform FFT, then pickup the first local maxima of the resulting spectra.

The process is very slow so some could easily do the task using low-power microcontrollers with built-in hardware multiplier, ADC and LCD driver like for instance the TI MSP430F449.
 

ezpcb

Banned
Joined
Mar 2, 2005
Messages
210
Helped
8
Reputation
16
Reaction score
3
Trophy points
1,298
Activity points
0
finger heartbeat monitors

RegUser_2 said:
ezpcb said:
Hi fabiorov
I'vd done such a project during my master's project. It's quite easy for just counting the heartbeat, a infrared optical coupler will do that. Just put one's finger between the emitter and receiver, then amplify the received signal, you'll see the heartbeating on the oscillscopes.

mike

---------------------------------
www.ezpcb.com

There's more than that because the measured AC signal due to the change of optical thickness due to pulsatile arterial blood flow is a tiny fraction (couple of procents) of the total transmitance of the finger and have to be detected in presence of several fundamental noise sources:

1. There is a stray light interfering (if the photodiode don't have proper optical filter then for instance luminiscent source produces large AC signals which need to be filtered)

2. There are the so called "motion artefacts" due to the fact the change of the optical path due to mechanical interferences may and do garble severely the pulsatile fraction of the signal.

3. The frequency range of typical human pulse is in the range of say 50 to 150 bpm (say 0.5 to 2 Hz) what lays well within the 1/f noise range of a standard readout electronics and hence the 1/f noise (thermal drifts, popcorn noise, contact noise, ...) may and will severely decrease the dynamic range of the measurement.

In general some have to modulate the LED and use synchronous demodulation or double correlated sampling to get rid of the stray light and 1/f noise of the signal amplifiers, then use some 0.1 to 5 Hz bandpass filter to get rid of out of signal band interferences. The resulting pulstile AC signal should resemble a sow-tooth waveform.

The period of this signal equals to the pulse rate you want to measure. The cleanest way to measure this period is to digitize this signal and perform FFT, then pickup the first local maxima of the resulting spectra.

The process is very slow so some could easily do the task using low-power microcontrollers with built-in hardware multiplier, ADC and LCD driver like for instance the TI MSP430F449.

Yes of course. I described the situation in a few simple words, but all the conditions you mentioned must be concerned. We use a special finger probe to eliminate the light interfering. And the patient is asked to be quiet when they are in the measurement process. The signals was filtered by a analog notch and then with a digital filter (our kits are based on dsPIC)

mike


------------------------------------------
www.ezpcb.com
 

RegUser_2

Full Member level 2
Joined
Dec 24, 2001
Messages
125
Helped
11
Reputation
22
Reaction score
9
Trophy points
1,298
Activity points
1,507
heart beat schematic

The best place for some to look for info on optical pulse oximetry ('cos optical pulsometer is a simplified and crippled pulse oximeter) is in the US patent database. It have abundance of info including schematics for some industrial instruments and also code and interesting ideas about the construction of the optical probes and modes of operation.

One of the basic cluster of inventions is related to reducing the influence of the motion artefacts. There is a big patent war going between Massimo and Nelcor regarding implementation of adaptive filtering algorithms.
 

jaltiti

Newbie level 1
Joined
Mar 26, 2007
Messages
1
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,286
heartbeat monitor (hemon) explanation

Hi,

I built Kieth Wilson Hearbeat mointor, but faced issues with amplifying the AC
signal. I could not get it to be 3V. Also I faced issues with noise. I going to use
an active bandpass filter and see if this will solve the issue. looking at his doc and
the osc. snap shoot he does not seem to have any issues with noise.
 

leoneqba

Newbie level 3
Joined
May 20, 2009
Messages
3
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,299
I made a pulse oximeter recently. Simply add a higher capacitance than normal over the feedback resistor to limit the bandwidth also you can also find increased improvement by adding a low pass filter in the feedback just like if you were amplifying a capacitive load. Another important consideration is that signal strength in the transimpedance amplifier is S = -ipRf; where as the noise due to the resisitor is N = [ 4*Kb*T*Rf]^(1/2); therefore the SNR is -ipRF/ [ 4*Kb*T*Rf]^(1/2) this means that as you increase the feedback resistance your SNR increases by (RF)^(1/2);
 

pranam77

Advanced Member level 5
Joined
Apr 22, 2008
Messages
2,777
Helped
553
Reputation
1,104
Reaction score
369
Trophy points
1,363
Location
INDIA
Activity points
16,406
leoneqba said:
I made a pulse oximeter recently. Simply add a higher capacitance than normal over the feedback resistor to limit the bandwidth also you can also find increased improvement by adding a low pass filter in the feedback just like if you were amplifying a capacitive load. Another important consideration is that signal strength in the transimpedance amplifier is S = -ipRf; where as the noise due to the resisitor is N = [ 4*Kb*T*Rf]^(1/2); therefore the SNR is -ipRF/ [ 4*Kb*T*Rf]^(1/2) this means that as you increase the feedback resistance your SNR increases by (RF)^(1/2);
If you are suggesting a new dimention to the previous posts in the thread, note that the last reply is more than 2 years old. If you are telling about your achievements, you may start a new thread for better views and replies. Cheers
 

sakurarandom

Junior Member level 3
Joined
Mar 26, 2011
Messages
30
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,286
Activity points
1,500
May I know anyone have the PIC coding in C language? I am using PIC18F4550. Now I am facing problem on how to calculate the heart beat from sensor in PIC. Can anyone teach me? Thank you so miuch
 

nikhilele

Advanced Member level 3
Joined
Jan 11, 2006
Messages
805
Helped
101
Reputation
202
Reaction score
76
Trophy points
1,308
Location
Bangalore India
Activity points
6,375
here you will find demo code and heart beat sensor both
**broken link removed**
you can buy sensor online from
Heart Beat Sensor [1157] : Sunrom Technologies, Your Source for Development

---------- Post added at 09:28 ---------- Previous post was at 09:27 ----------

I made a pulse oximeter recently. Simply add a higher capacitance than normal over the feedback resistor to limit the bandwidth also you can also find increased improvement by adding a low pass filter in the feedback just like if you were amplifying a capacitive load. Another important consideration is that signal strength in the transimpedance amplifier is S = -ipRf; where as the noise due to the resisitor is N = [ 4*Kb*T*Rf]^(1/2); therefore the SNR is -ipRF/ [ 4*Kb*T*Rf]^(1/2) this means that as you increase the feedback resistance your SNR increases by (RF)^(1/2);

Can you post your ckt diagram
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Top