3. Signal processing of biomedical signals a) Describe the most common artifacts observed in the ECG...

 

Transcribed Image Text:

3. Signal processing of biomedical signals a) Describe the most common artifacts observed in the ECG signal and explain if and how those artifacts can be removed. Given a Matlab variable ecg containing the data of a 3 seconds recording of the ecg of a person running. sampled at 1kHz, write the code for removing the expected artifacts from the recorded signal. Make sensible assumptions on the characteristics of the noise for this particular recording. (20%) b) How does a median filter work? When is it used (e.g. for which kind of noise)? How can you set the cut-off frequency of a median filter? (20%) c) The process of averaging across trials aims to provide a measure of the true underlying physiological signal of interest, by reducing the standard deviation of the noise by a factor of sqrt(N), where N is the number of trials. What are the assumptions underlying this assertion? (20%) d) The variable EMG contains the data of 10s of emg recording of 3 isometric muscular contractions from the biceps brachii muscle, sampled at 1kHz. Write the Matlab code for: 1) pre-processing the EMG, removing the most common artifacts present in the EMG; 2) extract the envelope of the EMG, using a cutoff frequency of 10 Hz; 3) Normalize the amplitude of the whole envelope to its maximum value; 4) define a variable activation that is 0 when the amplitude of the normalized envelope is smaller than 0.1 and 1 when the amplitude of the normalized envelope is greater than 0.1. (40%) 3. Signal processing of biomedical signals a) Describe the most common artifacts observed in the ECG signal and explain if and how those artifacts can be removed. Given a Matlab variable ecg containing the data of a 3 seconds recording of the ecg of a person running. sampled at 1kHz, write the code for removing the expected artifacts from the recorded signal. Make sensible assumptions on the characteristics of the noise for this particular recording. (20%) b) How does a median filter work? When is it used (e.g. for which kind of noise)? How can you set the cut-off frequency of a median filter? (20%) c) The process of averaging across trials aims to provide a measure of the true underlying physiological signal of interest, by reducing the standard deviation of the noise by a factor of sqrt(N), where N is the number of trials. What are the assumptions underlying this assertion? (20%) d) The variable EMG contains the data of 10s of emg recording of 3 isometric muscular contractions from the biceps brachii muscle, sampled at 1kHz. Write the Matlab code for: 1) pre-processing the EMG, removing the most common artifacts present in the EMG; 2) extract the envelope of the EMG, using a cutoff frequency of 10 Hz; 3) Normalize the amplitude of the whole envelope to its maximum value; 4) define a variable activation that is 0 when the amplitude of the normalized envelope is smaller than 0.1 and 1 when the amplitude of the normalized envelope is greater than 0.1. (40%)

Answer rating: 100% (QA)

a The most common artifacts observed in the ECG signal are noise power line interference and muscle ... View the full answer