Group Actions
Engagement
  • Author Author: sunnyiut
  • Date Created: Date Created
  • Views 4018 views
  • Likes 4 likes
  • Comments 11 comments
Related
Recommended

Part1.3 ECG output of AD8232

sunnyiut

Posted by Md. Kamrul Hussain March 17, 2015

Title: Total Player Monitoring

By: Md. Kamrul Hussain

Project Category: Design Challenge

Project Name: Sudden Impact Wearables Design Challenge

Blog post: 04

 

 

Intro:

It’s my fourth blog post in the Sudden Impact Wearables Design Challenge. Last couple of months was a bit hard for me. Got stuck with other stuffs and could not manage enough time to carry on this deign challenge. However, I am back again.

In this blog post I’ll be presenting the analysis of the ECG signals I have got from the Analog Front End, AD8232.

 

Setup:

Image of the full setup -                                                    Analog Front End AD8232 and Electrodes -

imageimage


Conventional Ag-AgCl electrodes were used for this experiment.

 

Supply and Isolation -


image

The Breadboard circuit is to provide the required 3.3v supply from a 6v battery and the opto isolation circuitry to observe the signal onto a scope without connecting it’s GND with the AFE. This keeps the signal clean from noise and also provides the minimum medical safety.

 

ECG output:

 

image

 

The captured ECG signal from AD8232 displayed on TBS1202B scope with 250msec time scale. The signal was inverted due to the use of opto isolator. By inverting the scope input I got the conventional output.

image

 

Output analysis:

The output of the AD8232 was taken as the Fast Response switch was disabled.

image

 

With enabling the Fast response switch I got an output almost similar to the previous one.

image

 

If I compare the output of AD8232, with the primary AFE designed by me then I can observe some differences.

image

 

This was the output of the primary AFE. It shows that the AD8232 is missing detail information in the signal. Like the P-wave is flatten and the whole T-wave is superimposed with the 50Hz noise.

 

May be that’s because of the filter bandwidth. In my design I had captured in between a range of 0.3Hz to 159Hz where AD8232 limits the low pass far below 50Hz to get rid of common noise. However, this won’t be necessary as it’s not for diagnostic purpose. I just need to get the heart rate with single LEAD ECG, so it’s pretty good enough. Moreover, AD8232 produces a better, sharp QRS complex which will be more helpful.


However, AD8232 responded effectively for removing motion artifacts. For general movement, it maintains the output signal pretty good. But for fast strong movements or contraction of the muscle the signal gets distorted. I have to check it's performance over motion artifacts once again after mounting it on human body.


image

All the outputs of the AD8232 is presented inversely because of the opto-isolator. The original output without isolator looks like below, where the 50Hz noise is added from oscilloscope.

 

image

 

Spectrum view:

image

 

Fast response:

With fast response enabled, the output was restored almost immediately after reconnecting the LEAD. With fast response disabled, the output becomes steady after about 0.5 to 1.5 seconds.

image

 

Shutdown:

The shutdown feature works perfectly.

imageimage

     image: Shutdown is turned on.                                        image: Signal restored after shutdown is turned off

 

LEAD off detection:

  Unfortunately, the LOD+/- didn’t perform as I thought. I placed the switch into DC position and removed one LEAD. I was expecting something like a digital high/low output swing. But when the LEAD was taken off the LOD output started oscillating.


image

 

In my next blog I’ll try to post the full circuit for detecting the heart rate from the ECG signal.

Parents

  • Hii ... can you please put some light on exact interfacing of ECG probe and configuration of AD8232 board.... Also I am very much interested to know on which place of body you have connected ECG probe...

  • in reply to ravi_butani

    Hello ,

    I don't know how much I can explain in detail, but I am trying my best -

    about ECG probe:

    I have used conventional ECG electrodes with Ag-AgCl coating generally used in clinical diagnosis. U can use the disposable electrodes too, I had used it in my primary design consists of general purpose instrumentation amplifier and the output was posted in my previous blog [part 1.2].

    imageimage

    The interfacing is very straight forward. The AD8232 evaluation board has 3 LEADs combination. U can use it for 2 LEADs combination too, but I have used 3 LEADs in my experiment.

    LEAD RL : the evaluation board has electrode terminal named 'RL' which stands for RIGHT LEG and they have used a bootstrapped circuit to drive the Right Leg electrode to reduce the common noise. So, our common ['RL'] Electrode should be placed in Right LEG [in between the ankle and the calf muscle].

    LEAD LA : there is another terminal on the board named LA, which goes to LEFT ARM. I have placed the electrode right above the Left Wrist.

    LEAD RA : the RA terminal of the board goes to Right Arm. I have placed the electrode right above the Right Wrist.

     

    ** now this combination actually represents one of the 3 LIMB LEADs called LEAD 01. I have checked the output of the AD8232 for LEAD 01 combination in this experiment. But I have planned to use it in LEAD 02 combination in my final design, because it gives a high QRS complex in most cases [varies man to man] which helps to detect the heart rate [I hope]. if u want, u can go through my previous blogs where I have explained the LEAD combinations graphically.

     

    about AD8232:

    there are three switches to vary the performance of the Analog Front End. in my experiment I tried these following combinations -

         1) [S1] FR - fast response disabled [right side], [S2] LOD - lead off detection switched to DC [right side], [S3] Shutdown - disabled

              it gave me the output without fast response. so when I removed one of the three electrodes, and then replaced it back - the output took about 0.5 - 1.5 seconds to get restored back.

         2) FR - enabled, LOD - DC, Shutdown - disabled

              it gave me a better and quick response to immediately restore the ECG signal after removing and reconnecting an electrode.

         3) finally I enabled the shutdown switch to see whether the output goes low or not.

    ** the LOD - lead off detection switch was kept in DC position all the time, because I used 3 LEADs combination. If u use 2 LEADs combination, u have to switch it into AC position [left side].

     

    ** u have to select the "INPUT BIAS SELECTION" - in my experiment, I have biased the LA and RA electrodes with Vs [supply voltage]. This is required for 3 LEADs combination and when the S2 [Lead off detection] is switched into DC mode. Therefore, i have selected the last terminal [biased by Vs] by putting on a jumper at P3.

     

    ** the "REFOUT" TP19 can be used if u want to process the output further more [like - driving the opto isolator using an opamp with dual supply to check the output in oscilloscope]

     

    Placement of electrodes:

    I hope u have got the placement of my electrodes for this experiment already. But in my final design, i'll change the position of the electrodes in such a way that it will be suitable for the players [ i hope ]. if u r interested to know in detail, u can check out my second blog for the modified electrode placements.

     

    -thank you for your interest, plz let me know if I made any mistake and plz feel free to ask me if u have any query

  • in reply to sunnyiut

    Thanks a lot ...

    Actually I have tried a lot with AD8232 for heart rate monitor in dynamic condition but I failed every time ... When there is no body movement it gives me stable reading but with body movement its heard to detect pulse... So finally I have gave up and moved to optical pulse sensor...

    With amazing results achieved by you with AD8232 its inspired me to once again do some tests on AD8232..

     

    Thanks Again..

    Ravi

Comment
  • in reply to sunnyiut

    Thanks a lot ...

    Actually I have tried a lot with AD8232 for heart rate monitor in dynamic condition but I failed every time ... When there is no body movement it gives me stable reading but with body movement its heard to detect pulse... So finally I have gave up and moved to optical pulse sensor...

    With amazing results achieved by you with AD8232 its inspired me to once again do some tests on AD8232..

     

    Thanks Again..

    Ravi

Children
  • in reply to ravi_butani

    for large and fast movements the signal also gets distorted for me. but I would like to work on it. so, I'll try to find the optimum placements of electrodes and beat detection algorithm.

    let's hope for the best.

    on the other hand, optical pulse sensor [plethysmograph] has the same limitation. It also gets distorted due to motion artifact. However, i'll try to check the performance of plethysmograph too, if i get enough time. I have wasted so much time... I am afraid that I may not be able to implement all the features I had planned.

    however, best of luck with your second try. hopefully, this time you will find it better. please let us know about your findings.

    -thank you.