EMGProcessing

The video introduces the difficulty of interpreting raw EMG data, which appears as voltage versus time graphs. It highlights the need for processing techniques to make the data understandable and comparable across different trials, especially given the inherent challenges of EMG collection.

One primary processing technique is rectification, which involves converting all negative voltage values to positive values. This makes the entire signal positive, simplifying interpretation. In MATLAB, this can be achieved using the 'absolute value' function.

Another method is to identify the envelope of the EMG data, which represents the general shape or outline of the signal. MATLAB has a dedicated 'envelope' command for this purpose, and users are encouraged to consult MATLAB's help documentation for its usage.

The root mean square (RMS) value is a common technique to analyze EMG data. Squaring the signal effectively rectifies it. MATLAB provides an 'rms' command for this calculation, offering an alternative to direct rectification.

Normalizing EMG outputs to a maximum voluntary contraction (MVC) is crucial for comparing data across trials and days. This involves performing an MVC at the start of data collection and then dividing subsequent EMG activity by this maximum value. While generally leading to values below 100%, it's possible for activity during an actual task to exceed the recorded MVC. In MATLAB, this is done by dividing all data points by the average MVC value.

Analyzing the frequency components of an EMG signal provides insights into muscle activity. The Fast Fourier Transform (FFT) converts a time-domain signal into the frequency domain, revealing the 'power spectrum'. The power spectral density (PSD) indicates the strength of different frequency components. MATLAB's FFT function can be used to generate this power spectrum.

To access the EMG data in MATLAB, use 'filename.analog.data'. Specific channels for EMG (columns 17 and 18) and goniometer data (column 23) are provided, with a recommendation to check 'analog.labels' for verification. The video concludes by instructing viewers to plot EMG activity versus time and to prepare to convert goniometer voltage data into angle measurements for future analysis of EMG as a function of joint angle.