Invasive BCI Chinese Character Decoding with Stroke-State Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing brain-computer interface (BCI) technologies struggle to accurately decode Chinese character writing due to the instability of neural signals caused by the plasticity of neurons during different writing states, such as writing strokes and breaks, leading to inaccurate decoding results.
Innovation Solution
A method for decoding Chinese character writing using an invasive BCI that involves filtering and preprocessing neural signals, dividing them into writing stroke and break states, and employing a state discriminator and predictors trained on Kalman filters and Hidden Markov Models (HMM) to distinguish and decode these states separately.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single decoder is used to decode both writing stroke and writing stroke break states, then the device complexity is reduced, but the decoding accuracy deteriorates due to neural signal instability across different writing states
Solution Approach 1:
The patent divides the decoding system into separate decoders for different writing states. Specifically, it segments the neural signal decoding process into at least two distinct decoders: one for writing stroke states and another for writing stroke break states. This segmentation allows each decoder to be specialized for its specific state, improving decoding accuracy despite the increased system complexity.
2Reliability
If neural signals are processed without state discrimination, then the processing speed is maintained, but the reliability of decoding results deteriorates due to signal instability during different writing behaviors
Solution Approach 1:
The patent applies preliminary action by first discriminating the writing state (stroke or break) before performing the actual decoding. The system pre-processes the neural signal to identify which state is currently active, then routes the signal to the appropriate specialized decoder. This preliminary state discrimination ensures reliable decoding results while maintaining processing efficiency through automated state-based routing.
3Manufacturing precision
If traditional filtering methods are used without state-specific processing, then the ease of operation is maintained, but the manufacturing precision of decoding results deteriorates due to inability to handle different writing states
Solution Approach 1:
The patent implements dynamics by making the decoding system adaptive to different writing states. The system dynamically selects which decoder to use based on the current writing state (stroke or break), allowing the decoding precision to be optimized for each specific state. This dynamic state-dependent processing achieves high decoding precision while the system automatically handles the complexity, maintaining ease of operation for the end user.
Data Source
AI summary
Disclosed in the present invention is a method for decoding Chinese character writing for an invasive brain-computer interface. In a practical application, a corresponding motor neural signal is divided into two states of a writing stroke and a writing stroke break in view of inconsistency of the writing stroke and the writing stroke break during Chinese character writing, and different filters are trained. A hidden markov model (HMM algorithm) and a Viterbi algorithm are used to judge a task state of the motor neural signal, and the corresponding signal is put into a corresponding decoder. The present invention effectively reduces influence of difference of neural data in different states on the decoder, and improves the performance and robustness of the decoder.