A mathematical model of the electromyogram has been developed and implemented in a program which runs in real-time. I have tried to keep the model as realistic as possible. The model simulation includes a waveform display of the EMG accompanied by an audio output.
The interpretation of the EMG depends considerably on the recording method and the architecture of the muscle as well as the physiological activity during the recording. Here the main factors affecting the recorded EMG are outlined and from which its analysis can be explained. The validity of the concepts underlying this explanation is best tested by constructing a formal model and simulating it.
The following figure shows a block schematic of the components contributing to the shape of a single motor unit action potential. Each action potential generated by the motor neuron excites the group of muscle cells comprising the motor unit. The shape of the MUAP depends on the muscle architecture (end-plate distribution), the fibre propagation properties (action potential shape), and the electrode positioning, dimensions, etc. The MUAP is the composite of all the fibre potentials from the motor unit. For a simple model of how these factors affect the MUAP shape see this program:MUAP simulation
Voluntary EMG involves the firing of several motor neurons and the activity of the corresponding motor unit fibres. The number of motor neurons that are active (recruitment) and the rate at which they are activated by the motor neuron (firing rate) depends on the force required from the muscle.
In this simulation the user first specifies the muscle parameters, namely, the general range of sizes of the motor units, and the average firing rate of the fibres. The exact motor unit features are generated by the program using these inputs. Therefore, everytime the muscle is respecified a new "muscle" is synthesised even if the user specifications are the same. This ensures that a number of different muscles, which are still in the category of normal or abnormal can be generated. This reflects the natural variation in the real world. The user can also control the muscle force which then determines the number of active units and their firing rate.
Clinical diagnostic EMG is usually performed with needle electrodes. The needle electrode is moved inside the muscle so that it is close to a particular motor unit of interest. In this simulation the user can move the needle diagonally into the muscle and thereby move near or away from motor units that are present in the simulated muscle.
One of the most important features of clinical EMG is the audio output. This simulation includes such an audio output. Therefore, it requires a properly configured sound card.
Some feature found in EMG machines like triggered raster display, Turns-Amplitude plots, are also available in the simulation.
This simulation will run in your browser provided it: (a) has a recent version of the Java Environment, and (b) has a properly configured sound card. Click here to begin: [EMG model simulation]
References
Suresh R. Devasahayam, Signals and Systems in Biomedical Engineering: Signal Processing and Physiological Systems Modeling. Kluwer Academic/ Plenum Publishers, New York, June 2000.
Suresh Devasahayam