Can you help please with suggestions or references which are show the state of the art for mental fatigue classifier in real environments ? such as experiences in driver fatigue based on EEG features.
In the EEG signal, components with very low and very high frequencies are not clinically meaningful (unless under certain conditions). For this reason, normal tapping is considered to be a lower frequency range, which is most important for neuropsychological physiology.
This range ranges from 0.1 to 100 Hz and in a finer state between 0.3 and 70 Hz. In a normal adult, the wake-up frequency range is low (0.3 to 7) Hz and high frequencies (above 30 Hz) are rarely seen. The dominant waves of these people are the average frequencies between 8 and 13 Hz and frequencies between 14 and 30 Hz. These rhythms are grouped into frequency bands:
Delta waves: Under 3.5 Hz,
Theta waves: Between 4 and 8 Hz,
Alpha waves: between 8 and 13 Hz
Beta waves: above 13 Hz.
The recorded EEG signal amplitude from the skin is typically between 10 and 100 microvolts and is rarely greater than 50 microvolts. Signals that are recorded from inside the kraft range from 500 to 1500 micro volts. This indicates that the main activity of the cortex from the fatty layers, skin, bone, spinal fluid, and ... filtered the main signal and severely weakens it.
Two methods for recording EEG signals are:
1- Registration as reference
2- Bipolar registration
In the first method, the potential of all electrodes is measured relative to one or two reference electrodes. The choice of appropriate reference in this method is very effective and important because in this method it is assumed that the reference does not have any electrical activity. Because of this, routers usually connect the reference electrodes to the left or right eardrum, or both.
In the second method, the potential difference between the two desired electrodes is measured and selected according to a specific standard, and therefore there is no specific reference here.
Each of the two methods has some advantages and disadvantages, depending on the type of application and the conditions to be chosen. In the reference electrode method, because of the large distances between the active electrodes with the reference electrode, the amplitude of the received signals is higher and, of course, the received noise is stronger, but the signal-to-noise ratio in this case is more than bipolar. Interpreting the strips obtained from the reference method is very difficult due to the different distances of the electrodes to the reference, and in some cases definitive results can not be obtained. In contrast to this method, in the bipolar method, due to the equal distances of electrodes, the signals obtained between two electrodes can be compared in terms of amplitude. However, in this method, the detection of polarities is a bit confusing, and this problem is solved by computer processing methods.
With a channel recording ocipital EEG and another eye movements, patterns can be found Ex: the akpha band diminishes or disappears substitutted by theta or delta and the eys show pendular movements, carcteristic of falling sleep. some people used this for controlling fatigue in drivers