4 results about "Event-related potential" patented technology

The present invention relates to systems and methods for facilitating product preferences and / or product recommendations. These systems and methods take into account biometric data of a subject in order to determine product preferences and / or product recommendations. Although other factors of the subject are optional, these factors can also be taken into account when determining product preferences and / or product recommendations for the subject. These product preferences or recommendations can be presented to the subject automatically via a display device or with the assistance of a product consultant. In order to obtain the biometric data, the subject will be exposed to a variety of stimuli, such as a fragrance / aroma stimulus. Biometric data will then be collected from the subject based on the subject's response to this fragrance / aroma stimulus. In some instances, the biometric data collected is related to the subject's event-related potentials (ERPs), i.e., the measured brain responses that are a direct result of a specific sensory, cognitive, or motor event.

The application discloses a self-supervised auditory electroencephalogram decoding method under a low signal-to-noise ratio condition and application thereof, and comprises the following steps: S1) collecting multi-channel original electroencephalogram signals generated by an auditory perception paradigm of a subject under a low signal-to-noise ratio condition of -9 dB, wherein the auditory perception paradigm simultaneously induces an auditory steady-state response (ASSR) and an event-related potential (P300) through a complex sound; and S2) sequentially performing a 0.3-75 Hz band-pass filtering and a 50 Hz notch filtering preprocessing operation on the original electroencephalogram signals to obtain denoised electroencephalogram signals. The self-supervised learning framework is constructed for the auditory electroencephalogram decoding task under the low signal-to-noise ratio condition, the ASSR and P300 features are synchronously induced by using the -9 dB low signal-to-noise ratio complex sound paradigm, compared with the traditional high signal-to-noise ratio experimental paradigm, the self-supervised learning framework is closer to the real complex acoustic application scene, and the applicability and practicability of the method under strong noise interference are significantly improved; the self-supervised learning framework can continuously provide the model with a reconstruction target with appropriate difficulty, and the discriminability and learning efficiency of the self-supervised representation are significantly improved.

This invention belongs to the field of biomedical engineering technology. It discloses a stem cell therapy efficacy evaluation system based on brain-computer interfaces and genetic information, comprising: acquiring multi-channel EEG signals and genomic polymorphic site data from patients; constructing a priority mapping matrix for repair needs in target brain regions; extracting micro-functional domain gene neural function coupling strength distribution data; identifying neural plasticity-sensitive regions based on changes in coupling strength gradients; calculating the regulatory enhancement coefficient and migration retardation coefficient of stem cell differentiation signals; constructing a multi-scale pathway synergy map; generating a composite scheme of treatment efficacy characterization parameters based on the map and mapping matrix; monitoring event-related potential waveform features to identify characteristic plasticity response peaks; constructing stem cell therapy efficacy evaluation results; and significantly improving the accuracy of stem cell therapy efficacy evaluation.