SYSTEMS AND METHODS FOR AUTOMATIC SEGMENT SELECTION FOR MULTI-DIMENSIONAL BIOMEDICAL SIGNALS

• US 20140379620A1
• Filed: 01/25/2013
• Published: 12/25/2014
• Est. Priority Date: 01/25/2012
• Status: Active Grant

• Current Assignee: Regents of the University of California (University of California)
• Original Assignee: Regents of the University of California (University of California)
• Inventors: Sarrafzadeh, Majid, Lan, Mars
• Granted Patent: US 11,170,310 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/372 Analysis of electroencephalograms
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• CPC Class Codes: G06F 18/211 Selection of the most significant subset of features
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/047 Pattern matching networks; Rete networks
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems
• CPC Class Codes: G16Z 99/00 Subject matter not provided for in other main groups of this subclass

Abstract

Systems and methods for automatically analyzing and selecting prominent channels from multi-dimensional biomedical signals in order to detect particular diseases or ailments are provided. Such systems and methods may be applied in different ways to obtain numerous benefits, such as lowering of power and processing requirements, reducing an amount of data acquired, simplifying hardware deployment, detecting non-trivial patterns, obtaining, clinical episode prognosis, improving patient care, and/or the like.

Claims

1. A computer-implemented method of analyzing signal data, the method comprising:

retrieving, by a computing device, a set of sample data, wherein the set of sample data includes sample data from a plurality of channels, and wherein the set of sample data is annotated with one or more occurrences of events of interest;

determining, by the computing device, one or more channels of the plurality of channels that include patterns associated with occurrences of the events of interest; and

labeling, by the computing device, the determined one or more channels as useful.

2. The computer-implemented method of claim 1, wherein determining one or more channels that include patterns associated with occurrences of the events of interest includes performing a multi-scale analysis of the set of sample data.

3. The computer-implemented method of claim 2, wherein performing a multi-scale analysis of the set of sample data comprises applying a wavelet transform to at least a portion of the set of sample data.

4. The computer-implemented method of claim 3, wherein applying a wavelet transform includes using a Haar wavelet.

5. The computer-implemented method of claim 2, wherein performing a multi-scale analysis of the set of sample data comprises generating a group of coefficients and normalizing the coefficients to generate features;

and

wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes filtering the features using a feature selection algorithm.

6. The computer-implemented method of claim 5, wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes verifying the accuracy of the features using a supervised learning algorithm.

7. The computer-implemented method of claim 1, further comprising discarding sample data from one or more channels not labeled as useful.

8. A nontransitory computer-readable medium having computer-readable instructions stored thereon that, if executed by one or more processors of a computing device, cause the computing device to perform actions for analyzing signal data, the actions comprising:

retrieving, by the computing device, a set of sample data, wherein the set of sample data includes sample data from a plurality of channels, and wherein the set of sample data is annotated with one or more occurrences of events of interest;

determining, by the computing device, one or more channels of the plurality of channels that include patterns associated with occurrences of the events of interest; and

labeling, by the computing device, the determined one or more channels as useful.

9. The computer-implemented method of claim 8, wherein determining one or more channels that include patterns associated with occurrences of the events of interest includes performing a multi-scale analysis of the set of sample data.

10. The computer-implemented method of claim 9, wherein performing a multi-scale analysis of the set of sample data comprises applying a wavelet transform to at least a portion of the set of sample data.

11. The computer-implemented method of claim 10, wherein applying a wavelet transform includes using a Haar wavelet.

12. The computer-implemented method of claim 9, wherein performing a multi-scale analysis of the set of sample data comprises generating a group of coefficients and normalizing the coefficients to generate features;

and wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes filtering the features using a feature selection algorithm.

13. The computer-implemented method of claim 12, wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes verifying the accuracy of the features using a supervised learning algorithm.

14. The computer-readable medium of claim 8, wherein the actions further comprise discarding sample data from one or more channels not labeled as useful.

15. A system for collecting and analyzing signal data, comprising:

a plurality of sensors;

a data store configured to store signal data collected from the plurality of sensors; and

a signal processing device communicatively coupled to the data store and configured to;

retrieve a set of sample data from the data store, wherein the set of sample data includes sample data from a plurality of channels, each channel associated with at least one sensor of the plurality of sensors, and wherein the set of sample data is annotated with one or more occurrences of events of interest;

determine one or more channels of the plurality of channels that include patterns associated with occurrences of the events of interest; and

label the determined one or more channels as useful.

16. The system of claim 15, wherein determining one or more channels that include patterns associated with occurrences of the events of interest includes performing a multi-scale analysis of the set of sample data.

17. The system of claim 16, wherein performing a multi-scale analysis of the set of sample data comprises applying a wavelet transform to at least a portion of the set of sample data.

18. The system of claim 17, wherein applying a wavelet transform includes using a Haar wavelet.

19. The system of claim 16, wherein performing a multi-scale analysis of the set of sample data comprises generating a group of coefficients and normalizing the coefficients to generate features;

and wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes filtering the features using a feature selection algorithm.

20. The system of claim 19, wherein determining one or more channels that include patterns associated with occurrences of the events of interest further includes verifying the accuracy of the features using a supervised learning algorithm.

21. (canceled)