Examples of analytics that can be grouped into the first layer of signal and data processing:
Signal decomposition and source separation (filtering by frequency bands and separating overlapping biological sources)
Signal referencing and normalisation (adjusting how raw signals are represented to improve consistency, comparability, and signal quality.
Signal quality and latency analytics (assessing whether the data are reliable, timely and usable)
Filtering an artifact removal (cleaning the raw signals by removing noise and unwanted frequency components)
Data acquisition, buffering (handling data)
Motion correction (compensating for movement-induced distortions in the recorded signals)
Examples of analytics that can be grouped into the second layer of interpretation and decoding:
Feature extraction and inference (converting signals into informative representations)
Classification of certain mental states
Image processing such as AI-based hippocampal segmentation and volume calculation
Pattern recognition
Threshold detection (identifying events of interest in filtered signals)
Connectivity mapping (analysing interactions between different signals or brain regions)
Frequency analysis (extracting spectral characteristics from signals)
Real-time BCI decoding to extract motor intention (interpreting neural signals in real time to infer user intent)
Event detection and annotation (identifying meaningful occurrences in the signal)
Quantitative analytics and calculation metrics such as spectral power and connectivity (going beyond data quality by comparing and interpreting brain activity)
Examples of analytics that can be grouped into the third or second layer of application and interaction:
Manual or AI post-processing of images
Stimulation-response profiling (examining how signals change in response to applied stimulation)
Aggregation for longitudinal insights (combining and analysing data collected over multiple time points)
Statistical modelling (identifying patterns, relationships, or predictions from processed signals)
Machine learning-based processing (using trained models to infer meaning)
Predictive AI (inferring or predicts states, intents, outcomes)