DEVELOPMENT OF FULLY-AUTOMATED CLASSIFIER BUILDERS FOR NEURODIAGNOSTIC APPLICATIONS
First Claim
1. A fully-automated method of building a binary classifier for classification of brain function based on electrical brain signals, comprising the steps of:
- acquiring quantitative signal features from a reference database of brain electrical activity data;
organizing the quantitative features into hierarchical classes based on one or more quantitative measures indicative of the performance of the features;
selecting at least one set of features at random from the highest class in the hierarchical organization;
encoding the set of features into at least one bit string; and
applying one or more evolutionary algorithms to the at least one bit string to obtain a binary classifier with near-optimal performance.
7 Assignments
0 Petitions
Accused Products
Abstract
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.
53 Citations
31 Claims
-
1. A fully-automated method of building a binary classifier for classification of brain function based on electrical brain signals, comprising the steps of:
-
acquiring quantitative signal features from a reference database of brain electrical activity data; organizing the quantitative features into hierarchical classes based on one or more quantitative measures indicative of the performance of the features; selecting at least one set of features at random from the highest class in the hierarchical organization; encoding the set of features into at least one bit string; and applying one or more evolutionary algorithms to the at least one bit string to obtain a binary classifier with near-optimal performance. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16)
-
-
17. A method of building a Linear Discriminant Function for classification of electrical brain signals, comprising the steps of:
-
selecting multiple sets of quantitative features from a larger available pool of features derived from electrical brain activity data; forming an initial population of chromosomes using the selected sets of features; and applying genetic algorithm operators to the population of chromosomes. - View Dependent Claims (18, 19, 20, 21, 22, 23, 24, 25)
-
-
26. A method of building a Linear Discriminant Function for classification of electrical brain signals, comprising the steps of:
-
selecting a set of quantitative features from a larger available pool of features derived from electrical brain activity data; encoding a chromosome as a binary bit string using the selected set of features; inverting the value of at least one bit at a random location on the bit string to generate a new bit string; and computing an objective function value of the new bit string. - View Dependent Claims (27, 28, 29, 30, 31)
-
Specification