Measuring Auditory Brainstem response

Afsheen JInan

Other Parts Discussed in Thread: ADS1299, INA128

Auditory Brainstem response "ABR" (0.1 - 0.5 uV) signal lies inside the EEG recordings (5-10uV).

In order to extract the ABR signal from noise, it should be amplified (Using Differential Amplifier) and filtered (Low pass and high pass).

In this case, EEG is considered as noise.

To avoid latencies in recorded ABR signal, digital signal processing is preferred.

However, to have a high SNR, it is better to have an amplifier with the gain of 100,000 (As the signal amplitude is between 100 - 500 nV) and an analog bandpass filter (30Hz to 3kHz), before converting the signal from analog to digital.

I have come across the IC ADS1299, which best suits the need, contains a differential programmable gain amplifier (PGA) along with Right leg driven circuit (RLD) - [To reduce the Common mode noise] and lead off detection circuits.

However the gain of the PGA is limited to 24, which is unlike the Instrumentation amplifiers (Gain is 1000) and it doesn't contain any analog filters embedded.

Can you please suggest any IC, in which an analog filter is Embedded

Also an amplifier which generates less noise (As the signal amplitude is between 100 - 500 nV), and has got a high gain.

Thanks and Regards

Afsheen

over 8 years ago

0 Pete%20Semig over 8 years ago

TI__Mastermind 25805 points

Hello Afsheen,

Given a gain of 100,000V/V, you will have to cascade gain stages. After looking through our instrumentation amplifier portfolio, the only device that I think may interest you is the INA128/9. The maximum gain is listed as 10,000V/V, but please note that the gain error increases with gain. Nonetheless, the 0.1Hz to 10Hz noise is listed as 0.2uVpp (typical).

I hope this helps!

0 Afsheen JInan over 8 years ago in reply to Pete%20Semig

Prodigy 60 points

Dear Pete,

Thank you for the reply.

Can you please suggest any ADC IC, in which an analog filter is Embedded with amplifier

Regards,
Afsheen

0 Pete%20Semig over 8 years ago in reply to Afsheen JInan

TI__Mastermind 25805 points

Hello Afsheen,

I have moved this post to the Precision Data Converters forum...perhaps they can help you select an ADC.

0 Brian Pisani over 8 years ago in reply to Pete%20Semig

TI__Mastermind 38745 points

Hey Afsheen,

We do not have any products that integrate an active filter with an ADC. The only thing that is similar is the PGA on the ADS1299 that has low-pass properties (refer to Table 5 in the ADS1299 datasheet). Being a delta-sigma, the device also digitally low-pass filters the data, but to get the exact response you are looking for, you will likely have to digitally filter the output data in the processor.

Regards,
Brian Pisani

0 Zhang jian over 8 years ago in reply to Brian Pisani

Prodigy 90 points

Hi Engineer,

I also use ads1299 to acquire the signal of Auditory Brainstem response "ABR" (0.1 - 0.5 uV) .As shown below,It is th standard

ABR signal.

Here is my circuit,just like your ads1299 EVM without any preamplifier before ads1299.

I use this board to get background noise and the spectrogram of the background noise .They are below.

However,when I use this board to get ABR signal by repeating 1500 times(for decreasing random noise) ,and them averaging with Matlab.I just get a strange result. The signal by averaging 1500 and the spectrogram of the signal are follow. I have not get the ABR signal,but a signal seems not random noise.Where the strange signal comes from,How could I do to get ABR?

Sincerely

0 Brian Pisani over 8 years ago in reply to Zhang jian

TI__Mastermind 38745 points

Hello Zhang,

It seems like you are getting some interference from signals in the kHz range. However, the signal you wish to measure is only 100 Hz (probably +/- 100 Hz as well). I would recommend you use a lower data rate. This will low pass filter out some of that interference. Perhaps try setting the data rate to 1 kSPS instead.

Brian

0 Zhang jian over 8 years ago in reply to Brian Pisani

Prodigy 90 points

HI，Engineer

However,The ABR signal just lsats about 10ms，so if I use a lower data rate,I will get little point.

0 Brian Pisani over 8 years ago in reply to Zhang jian

TI__Mastermind 38745 points

Hello Zhang,

I was referring to the figure that you posted above in which it seems the signal lasts about 1s. Is this not the case?

Brian

0 Zhang jian over 8 years ago in reply to Brian Pisani

Prodigy 90 points

Hi，Engineer
I just record 15ms signal， the ABR signal is just about 10ms，after sound stimulation.

0 Brian Pisani over 8 years ago in reply to Zhang jian

TI__Mastermind 38745 points

Hello Zhang,

It will then be necessary to identify and eliminate the sources of interference. I would also recommend you look at the input referred noise tables in the ADS1299 datasheet (Tables 1-4) to ensure that your signal is detectable among the noise inherent in the AD1299 for given settings.

Brian

0 Zhang jian over 8 years ago in reply to Brian Pisani

Prodigy 90 points

Thank you,Engineers

What’s more ,I want to know the AC input impedance of ads1299,not the DC one in datasheet.

Sincerely

0 Zhang jian over 8 years ago in reply to Brian Pisani

Prodigy 90 points

Thank,you,Engineers

what's more,I want to know the AC input impedance of ads1299,not the DC one in datasheet.

Sincerely

0 Brian Pisani over 8 years ago in reply to Zhang jian

TI__Mastermind 38745 points

Hey Zhang,

This value is not specified. However we do specify input capacitance as 20 pF typically. This will help you to predict the impedance you can predict for an ac input.

Regards,
Brian Pisani

Data converters

Data converters forum

Measuring Auditory Brainstem response