- Slides: 18
Neuro. Phone : Brain-Mobile Phone Interface using a wireless EEG Headset 11 -03 -2014 Ilho nam
BCI (Brain–computer interface) Science and technology of devices and systems responding to neural and/or cognitive processes, which enable their user - who may have neural pathways that have been damaged by amputation, trauma or disease – to : - Control their environment through reanimation of paralyzed limbs - Control robotic devices - Control computer devices
Emotiv Headset • 14 Channel EEG(Electroencephalograph) • Designed for gaming, • consumer biofeedback • Affordable –$399 basic headset –$799 research edition • Wireless, highly portable
Some Important Things they Tell you about the Hardware • Includes gyro output (captures head movements) • Samples at 128 Hz from a 2048 Hz internal rate • Records voltage from easy to use saline sponge sensors • SDKs available in C++ & Python
Some Important Things they Don’t Tell you about the Hardware • Hair is a huge issue for getting a good connection with the saline sensors. • When batteries are low the sensors will continue operating but start returning increasingly random data (always keep charged) • For some reason you can’t record while the device is plugged in • Saline + exposed metal parts = corrosion
The Neuro. Phone A dial system for smartphones based on the P 300 BCI, and using the EMOTIV headset For EEG acquisition
The Neuro. Phone : System architecture
EEG(Electroencephalograph) • An EEG records electrical signals from the brain – Measures postsynaptic potentials of neurons, via electrodes on the scalp • An EEG detects the summed ionic currents of thousands of pyramidal neurons beneath each electrode. • The signals relayed to the EEG are typically amplified 10, 000 times and filtered.
Brain Wave Types • EEGs record brain waves which are oscillating electrical voltages in the brain measuring a few volts. • There are six widely recognized brain waves: – Delta: 1 -4 Hz. – Theta: 4 -7 Hz. – Alpha: 8 -12 Hz. – Mu rhythm is alpha-range activity that is seen over the sensorimotor cortex. – Beta: 12 -30 Hz. – Gamma: 30– 100 Hz.
The P 300 -based BCI Imagine you are waiting in your car… The traffic light is red…
The P 300 -based BCI
The P 300 -based BCI About 300 ms after it turns green, a specific phenomenon appears in your brain : the << P 300 Potential >>
What Does the Signal Look Like? Three Example Electrodes
The simplest BCI EYES OPEN EYES CLOSED 1 sec
Emotiv Epoc Test. Bench - 3 D Brain Activity Map Windows 8 EEG http: //youtu. be/Erb-N 3 OD 6 nc
EVALUATION This preprocessing separately for all 6 stimuli corresponding to the six images of the Dial Tim application. The different time durations of data accumulation for averaging. Contact pictures are flashed once every half a second in random order; each of the 6 pictures has a 1/6 chance for each flash.
Conclusion • The EMOTIV headset, although cheap, is able to detect most important EEG features exploited in medical Brain-Computer Interfaces • As all EGG system, EMOTIV is very sensitive to all kinds of artifacts. Some solutions are emerging to address this issue but anyway, extra care should be taken to avoid them as much as possible Future? • Whole brain coherence measures, recurrence analysis, synchrony, entrainment etc…. • Sniffing packets could take on a very new meaning if brain-mobile phone interfaces become widely used.