Automated EEG Classifier Builder Using Evolutionary Algorithms
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current EEG equipment is inadequate for real-time, objective evaluation of electrical brain signals, making it difficult for non-neurodiagnostic personnel to interpret data accurately in emergency settings, and relies heavily on human selection of features for classification, which is time-consuming and prone to subjectivity and error.
Innovation Solution
The development of fully-automated methods using evolutionary algorithms and genetic algorithms to build classifiers for classifying electrical brain signals, selecting features randomly and encoding them into bit strings to optimize classification performance, and applying Linear Discriminant Functions for accurate brain function assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual feature selection by human designers is used, then classification systems can be built, but the process is time-consuming and prone to subjectivity and human error
Solution Approach 1:
The system performs automatic feature selection and classifier construction without human intervention. The computer automatically evaluates multiple features, selects the optimal subset, and builds the classification model, eliminating the need for manual feature selection by human designers while maintaining or improving classification accuracy
Solution Approach 2:
The patent replaces the mechanical process of manual feature selection with an automated computational system. Instead of human experts manually selecting features based on experience and judgment, the system uses algorithms to automatically evaluate and select features, substituting human cognitive processes with computational mechanisms
2Measurement precision
If exhaustive search of all feature combinations is performed, then the best classifier can be found, but the computational task becomes very time-consuming and intensive
Solution Approach 1:
The system extracts only the most relevant features from the complete set of available features. Instead of evaluating all possible feature combinations, the automated system identifies and extracts the subset of features that contribute most to classification accuracy, reducing computational requirements while maintaining performance
Solution Approach 2:
The system performs partial evaluation of feature combinations rather than exhaustive search. It evaluates a strategically selected subset of feature combinations that are most likely to yield optimal results, avoiding the computational burden of evaluating all possible combinations while still finding effective classifiers
3Measurement precision
If more features are selected for the classifier, then classification accuracy may improve, but the complexity of the classifier increases
Solution Approach 1:
The system dynamically adjusts the number and type of features used in the classifier based on their contribution to classification accuracy. It changes the feature set parameters automatically, selecting only those features that provide meaningful improvement in accuracy while keeping the classifier complexity manageable
Data Source
AI summary
Methods for constructing classifiers for binary classification of quantitative brain electrical activity data is described. The classifier building methods are based on the application of one or more evolutionary algorithms. In one embodiment, the evolutionary algorithm used is a genetic algorithm. In another embodiment, the evolutionary algorithm used is a modified Random Mutation Hill Climbing algorithm. In yet another embodiment, a combination of a genetic algorithm and a modified Random Mutation Hill Climbing algorithm is used for building a classifier. The classifier building methods are fully automated, and are adapted to generate classifiers (for example, Linear Discriminant Functions) with high sensitivity, specificity and classification accuracy.