BESP: A novel BCI experiment simulation platform to assist BCI-related algorithm validation

In recent decades, many brain-computer interface (BCI) software platforms have emerged. However, there are still some limitations. First, integrating an algorithm on online BCI software platform is difficult and time-consuming. Second, there is no guarantee that an unproven algorithm will work prope...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) s. 1 - 6
Hlavní autoři: Chen, Qingzhi, Ke, Jingshuang, Huang, Zhihua
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 28.10.2023
Témata:
Brain-computer interfaces
Measurement
Signal processing algorithms
Software
Software algorithms
Statistical analysis
Visualization
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:In recent decades, many brain-computer interface (BCI) software platforms have emerged. However, there are still some limitations. First, integrating an algorithm on online BCI software platform is difficult and time-consuming. Second, there is no guarantee that an unproven algorithm will work properly. Last, existing platforms do not support automatic recording of result metrics during experiments, and frequent validation of multiple algorithms is inefficient. In this paper, we designed a novel BCI experiment simulation platform (BESP) based on offline analysis to satisfy the needs of algorithm validation. BESP reorganizes the experimental steps and provides some interfaces, which makes algorithm validation simple and efficient. In addition, BESP is able to record the results produced during the experimentation. Then the results of multiple algorithms can be easily analyzed, compared and visualized by the statistical analysis system provided by BESP. With these characteristics, BESP is able to verify the correctness of the algorithm. Finally, we designed a simulated P300 spelling experiment based on BESP, and completed the validation and analysis of Algorithms SPLUCB and TSPLUCB (based on PLUCB and TPLUCB). Furthermore, we used BESP to analyze the differences between the stimulus signals of the 20 subjects. Through the analysis, we found some areas worthy of further study. In summary, BESP is well suited for the process of analysis, optimization and validation among multiple algorithms.
DOI:10.1109/CISP-BMEI60920.2023.10373290