Replacing EEG Sensors by AI Based Emulation
| Autor: | Daniel Kolb, Dieter Kranzlmüller, Fabio Genz, Johannes Starck, Clemens Hufeld, Simone Müller |
|---|---|
| Rok vydání: | 2021 |
| Předmět: | |
| Zdroj: | Lecture Notes in Computer Science ISBN: 9783030875947 AVR |
| DOI: | 10.1007/978-3-030-87595-4_6 |
| Popis: | Electroencephalography (EEG) has become a widely used non-invasive measurement method for brain-computer interfaces (BCI). Hybrid BCI (hBCI) additionally incorporate other physiological indicators, also called bio-signals, in order to improve the decryption of brain signals evaluating a variety of different sensor data. Although significant progress has been made in the field of BCI, the correlation of data from different sensors as well as the possible redundancy of certain sensors have been less frequently studied. Based on deep learning our concept presents a theoretical approach to potentially replace one sensor with the measurements of others. Hence, a costly or difficult to sensor measurement could be left out of a setup completely without losing its functionality. In this context, we additionally propose a conceptual framework which facilitates and improves the generation of scientifically significant data through their collection within a corresponding VR application and set-up. The evaluation of these collected sensor data, which is described in five consecutive steps, is to cluster the data of one sensor and to classify the data from other sensors into these clusters. Afterwards, the sensor data in each cluster are analysed for patterns. Through the predictive data analysis of existing sensors, the required number of sensors can be reduced. This allows valid statements about the output of the original sensor with no need to use it effectively. An artificial intelligence (AI) based EEG emulation, derived from other directly related bio-signals, could therefore potentially replace EEG measurements which indirectly enables the use of BCI in situations where it was previously not possible. Future work might clarify relevant questions concerning the realisation of the concept and how it could be further developed. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|