Brain-Computer Interface Clustering for Signal Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Brain-computer interface technologies using EEG signals face challenges with low signal-to-noise ratios, inter-subject variability, and the need for time-consuming customized models for each user, making it difficult to provide reliable and efficient user intention recognition.
Innovation Solution
A brain-computer interface apparatus utilizing clustering technology to minimize signal correction processes by extracting clustering features from frequency powers of brain signals across multiple subjects, generating clustering models, and constructing intention determination models for reduced machine learning and faster user intention classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual customized models are created for each user to account for inter-subject variability, then recognition performance is improved, but the time required for signal correction and model creation increases significantly
Solution Approach 1:
The patent segments users into distinct clusters based on their brain signal characteristics (e.g., spectral power distribution patterns). Instead of treating each user individually, the system divides the user population into groups with similar neural patterns, allowing shared model usage within clusters. This segmentation reduces the need for extensive individual customization while maintaining recognition accuracy.
Solution Approach 2:
The patent creates representative clustering models that serve as templates for groups of users with similar brain activity patterns. Once a clustering model is established for a particular user group, it can be copied and applied to new users belonging to the same cluster, significantly reducing the time required for signal correction and model creation while maintaining recognition performance.
2Measurement precision
If deep learning technology is used to improve brain wave analysis accuracy, then recognition performance is improved, but the complexity of the system increases
Solution Approach 1:
The patent applies deep learning technology selectively to specific aspects of brain wave analysis rather than implementing comprehensive deep learning across the entire system. By focusing computational complexity only on critical analysis tasks and using simpler methods for other processing stages, the system achieves high accuracy while controlling overall complexity.
Solution Approach 2:
The patent performs preliminary processing and feature extraction on brain wave signals before applying deep learning algorithms. By pre-processing the data to extract relevant features and reduce dimensionality, the system reduces the computational burden on deep learning models, thereby improving efficiency and reducing system complexity while maintaining analysis accuracy.
3Ease of manufacture
If EEG signals are measured non-invasively using electrodes on the scalp, then cost and ease of application are improved, but signal-to-noise ratio deteriorates
Solution Approach 1:
The patent introduces clustering models as an intermediary layer between the noisy EEG signals and the intention recognition system. These models learn to filter and interpret the signals by identifying patterns specific to each user cluster, effectively separating the useful brain activity information from the noise introduced by non-invasive measurement, thereby improving signal-to-noise ratio while maintaining ease of application.
Data Source
AI summary
The present disclosure relates to a technical idea for minimizing a signal correction process between users using brain activity-based clustering technology. More specifically, the present disclosure relates to technology for minimizing a signal correction process between users by clustering a brain signal of a measurement subject into a specific clustering model and determining an intention of the measurement subject using an intention determination model learned on the specific clustering model. The brain-computer interface apparatus according to one embodiment of the present disclosure may include a feature extractor for extracting a plurality of clustering features using frequency powers for each band of brain signals measured from a plurality of learning subjects; a clustering model generator for generating a plurality of clustering models based on the extracted clustering features; and a brain wave processor for constructing an intention determination model by performing machine learning of brain signals for each of the generated clustering models, determining a newly measured brain signal of a measurement subject as any one of the clustering models, and determining an intention of the measurement subject using the constructed intention determination model.