LAUSR

Objective. Asynchronous brain-computer interfaces (BCIs) show significant advantages in many practical application scenarios. Compared with the rapid development of synchronous BCIs technology, the progress of asynchronous BCI research, in terms of containing multiple targets and training-free detection, is still relatively slow. In order to improve the practicability of the BCI, a spatio-temporal equalization multi-window algorithm (STE-MW) was proposed for asynchronous detection of steady-state visual evoked potential (SSVEP) without the need for acquiring calibration data. Approach. The algorithm used SIE strategy to intercept EEG signals of different lengths through multiple stacked time windows and statistical decisions-making based on Bayesian risk decision-making. Different from the traditional asynchronous algorithms based on the ‘non-control state detection’ methods, this algorithm was based on the ‘statistical inspection-rejection decision’ mode and did not require a separate classification of non-control states, so it can be effectively applied to detections for large-scale candidates. Main results. Online experimental results involving 14 healthy subjects showed that, in the continuously input experiments of 40 targets, the algorithm achieved the average recognition accuracy of 97.2±2.6% and the average information transfer rate (ITR) of 106.3±32.0 bitsmin−1 . At the same time, the average false alarm rate in the 240 s resting state test was 0.607±0.602 min−1 . In the free spelling experiments involving patients with severe amyotrophic lateral sclerosis, the subjects achieved an accuracy of 92.7% and an average ITR of 43.65 bits min−1 in two free spelling experiments. Significance. This algorithm can achieve high-performance, high-precision, and asynchronous detection of SSVEP signals with low algorithm complexity and false alarm rate under multi-targets and training-free conditions, which is helpful for the development of asynchronous BCI systems.