Our team is working on the development of the brain-computer interface technology that will allow effective mental communication between the user and the computer with the help of a mobile EEG device and a computer screen. At the heart of the project is a system of a fast and unambiguous identification of the user’s attentional locus on the screen, which allows us to tell where the users are concentrating their attention in the real time. Our current project is designed to address existing drawbacks of the current BCI solutions, namely poor precision and slow identification rate . To this end, we employ two technical solutions: cutting edge machine learning algorithms and an innovative spatial frequency techniques,.
The machine learning techniques implemented in our project were designed and tested for the EEG-based classification of the visually presented stimuli. These techniques consist of the pretrained fine-tuned convolutional neural network algorithm supported by the advanced ...
Our team is working on the development of the brain-computer interface technology that will allow effective mental communication between the user and the computer with the help of a mobile EEG device and a computer screen. At the heart of the project is a system of a fast and unambiguous identification of the user’s attentional locus on the screen, which allows us to tell where the users are concentrating their attention in the real time. Our current project is designed to address existing drawbacks of the current BCI solutions, namely poor precision and slow identification rate . To this end, we employ two technical solutions: cutting edge machine learning algorithms and an innovative spatial frequency techniques,.
The machine learning techniques implemented in our project were designed and tested for the EEG-based classification of the visually presented stimuli. These techniques consist of the pretrained fine-tuned convolutional neural network algorithm supported by the advanced EEG data-analysis procedure.
The spatial frequency tagging technique relies on the Fast Periodic Visual Stimulation paradigm, a research technique that records periodic neural response to a periodically modulated stimulation. Recent research identified that this periodic response can be reliably captured by the EEG; and that the power and the localization of this response on the scalp are dependent on the attentional locus of the viewer. This phenomenon represents a convenient probe into the attentional locus of the potential user. We intend to devise screens with weak subliminal frequency stimulation that will allow us identifying the locus using a simplified dry EEG system.
In comparison to the current neural based techniques, our technology will provide a higher level of accuracy and superior speed due to the combination of the machine learning algorithms with the advanced visual presentation paradigm.
The potential applications of this technology vary from assistive implementations in cases of patients with severe motor problems, including advanced cases of neural degenerations like ALS or paralysis to the implementation of this technology in consumer products, like TV sets, phones, and computers to allow mental control over these appliances. Finally, this technology can be useful for industrial and security purposes, when it’s imperative to know users locus of attention, for control and visual information projection purposes, like in case of pilots.
More information

Employees

Maxim Kazimirov
Admin
Maxim Kazimirov CEO Master of Science, Biology. Rich business development background.