Acquisition and Semi-Automated Analysis of Respiratory Muscle Surface Electromyography

I'm interested in the motor control of the respiratory muscles, and this methodology will help us understand their coordination by examining both the timing and the magnitude of their activation. The most significant recent findings are how the cognitive demands of even a low level of loaded breathing with associated dyspnea can result in more errors of a cognitive task and decrease the speed of the physical task. In addition, the pattern of breathing also becomes more variable.

Our protocol addresses the need for a method that removes ECG artifacts from EMG signals without compromising the amplitude or the timing, thereby preserving muscle activation details that are essential for accurately analyzing respiratory muscle coordination during breathing. Our protocol basically removes only the ECG frequency content from the combined ECG, EMG signal. That's only the ECG artifacts.

It retains all other essential details of the EMG signals and the temporal accuracy while accommodating variable heart widths with precision. Consider the cognitive demands of a patient with COPD crossing a busy street. This requires thinking about traffic, coordination to walk, processing breathlessness, and processing increased recruitment of the ventilatory muscles.

My research focuses on how the cognitive load and motor control of ventilatory and limb muscles can interfere with physical activities. To begin, position the participant in a seated position. Rub the skin with an alcohol wipe and let it dry completely to reduce the skin impedance.

Identify the muscles of interest through landmarking and palpation and place the surface electrode on the right side of the thorax. For the costal diaphragm and intercostal muscles, identify the anterior axillary line and midclavicular line. Place the paired electromyography electrodes vertically between these two lines at the level of the seventh or eighth intercostal space.

Then identify the posterior triangle of the neck for locating the scalene muscle. This muscle can be visualized more clearly by having the participant take a big breath or by right-sided isometric resisted side flexion. Now place the paired electrodes along the longitudinal axis of the muscle at the level of the cricoid process for the parasternal intercostal muscles, locate the second intercostal space one to two centimeters lateral to the right side of the sternum.

The location of this rib is facilitated by palpating the clavicle, the manubrium, and the sternal angle of Louis at the junction between the manubrium and the sternum. The second parasternal intercostal muscle is just to the right and inferior to the sternal angle of Louis. Position the paired electrodes along the longitudinal axis of the muscle.

Next landmark the sternocleidomastoid muscle after locating the suprasternal notch and mastoid process. Place the hand on the left side of the participant's chin asking the participant to gently perform isometric left rotation against the hand to accentuate the right sternocleidomastoid muscle belly. Then position the paired electrodes at the midpoint of the muscle belly along its longitudinal axis.

If required, attach the ground sensor to C7 or T1 spinous process and place the negative ECG electrode over the manubrium. Then place the positive ECG electrode over the fifth intercostal interspace at the left anterior axillary line. Now attach electromyography sensor clips to the electromyography electrodes.

Apply double-sided tape under each electromyography sensor to fix it to the skin securely. Ensure that wires from different electromyography sensors do not overlap or create crosstalk between muscles. Place medical grade hypoallergenic tape over the electrodes and sensor to secure them further to the skin without applying excessive pressure.

For signal acquisition, select the preset template on the data acquisition software and press Open. The preset parameters will have a 0.5-20 hertz high pass filter in the EMG signal to reduce low frequency artifacts and facilitate realtime visualization. Ensure the sampling rate is set to 1 kilohertz and the gain of the EMG signal is 1, 000.

After selecting the template, acquire a synchronized recording of ECG and respiratory flow. Acquire surface EMG and ECG data according to protocol, such as during an inspiratory threshold loading in a healthy volunteer participant. For pre-processing, open the software and confirm the parameters of a bidirectional high pass filter of 5 hertz, the least mean square adaptive filter to remove ECG contamination, and a root means square transform with a moving window of 0.02 seconds, then press Continue.

Next, select the file to be analyzed and press OK.If analyzing the entire duration, set the range from 0 seconds to the maximum time. Press Select the range, Continue and then Conditioning. Press the analyse button to apply preset parameters and visualize the analyzed EMG signal.

Press the rescaled on 1 button to normalize the EMG signal by its maximum value during the recording. Press continue to calculate on off. The software will detect EMG activity onset timing based on the derivative function of the EMG signal.

Now press the on and off button. Select the EMG signal from the desired muscle for visualization. Alternate between muscles to inspect all recorded signals.

Press stop looking and go to saving and click Saving. Select data to save with an option to reduce signal frequency. Press Save processed data.

Select the folder for saving. Name the file and press Save. For post-processing, open the saved file using calculation software for post-processing.

Identify each breath by the on and off times of the flow signal. Calculate the EMG peak RMS and mean RMS for each breath. For EMG onset, calculate the absolute difference between EMG onset and inspiratory flow onset in milliseconds.

And then for EMG offset, calculate the absolute difference between EMG offset and the end of inspiratory flow. For EMG onset relative to the duration of inspiratory time, calculate the relative difference between EMG onset and inspiratory flow onset. Finally, for EMG offset relative to the duration of inspiratory time, calculate the relative difference between the EMG offset and the inspiratory flow offset.

ECG artifacts were reduced in the diaphragms filtered electromyography compared to the raw electromyography. At low inspiratory threshold loading, onset activity of the scalene and parasternal intercostal muscles occurred before the onset of inspiratory flow, whereas the diaphragm and sternocleidomastoid muscles activated after inspiratory flow onset. At high inspiratory threshold loading, earlier activation of all four muscles relative to inspiratory flow was observed.

The duration of electromyography activity for the diaphragm, parasternal intercostal, and scalene muscles remained similar between low and high loads. The duration of sternocleidomastoid muscle activity was longer at high load compared to low load. The root mean square of electromyography for all muscles was higher at high load than at low load, indicating increased muscle activity.