PRECISE FERQUENCY-PATTERN ANALYSIS TO DECOMPOSE COMPLEX SYSTEMS INTO FUNCTIONALLY INVARIANT ENTITIES

• US 20140107979A1
• Filed: 10/16/2013
• Published: 04/17/2014
• Est. Priority Date: 10/16/2012
• Status: Abandoned Application

• Current Assignee: New York University
• Original Assignee: New York University
• Inventors: Llinas, Rodolfo, Ustinin, Mikhail N.
• Application Number:
  US14/055,692
• Publication Number:
  US 20140107979A1
• Time in Patent Office: Days
• US Class Current: 702/189
• CPC Class Codes: A61B 5/245 specially adapted for magnetoencephalographic [MEG] signals
• CPC Class Codes: A61B 5/7257 using Fourier transforms
• CPC Class Codes: G01R 15/00 Details of measuring arrangements of the types provided for in groups G01R17/00 - G01R29/00, G01R33/00 - G01R33/26 or G01R35/00
• CPC Class Codes: G06F 17/14 Fourier, Walsh or analogous domain transformations , e.g. Laplace, Hilbert, Karhunen-Loeve, transforms for correlation function computation G06F17/156; spectrum analysers G01R23/16
• CPC Class Codes: G06F 18/00 Pattern recognition
• CPC Class Codes: G06F 2218/14 by matching peak patterns

Abstract

Methods, systems, and apparatuses, including computer programs encoded on computer readable media, for registering a multichannel time series from data collected from a plurality of channels. Each channel corresponds to one of one or more sensors. A precise Fourier transform of the data from all channels is done to transform the data in a frequency domain. Frequencies from the frequency domain are selected. The selected frequencies from all channels are inverse Fourier transformed. A restored time is selected, and activity for the selected frequencies at the restored time is localized. Coherence for the selected frequencies is estimated and coherent components are extracted. Partial spectrum is assembled comprising frequencies with similar patterns of the coherent components. Inverse transform of the partial spectrum is used to reconstruct a time series for functional entities producing the similar patterns of the coherent components.

Claims

1. A method comprising:

registering a multichannel time series from data collected from a plurality of channels, wherein each channel corresponds to one of one or more sensors;

precise Fourier transforming, using a processor, the data from all channels to transform the data in a frequency domain;

selecting frequencies from the frequency domain;

inverse Fourier transforming, using the processor, for the selected frequencies from all channels;

selecting a restored time; and

localizing, using the processor, activity for the selected frequencies at the restored time.

2. The method of claim 1, further comprising estimating a multichannel spectrum based upon the precise Fourier transform.

3. The method of claim 1, further comprising tuning the precise Fourier transform to a selected frequency.

4. The method of claim 1, further comprising determining if activity associated with a selected frequency is external noise.

5. The method of claim 4, wherein the sensors are magnetic field sensors.

6. The method of claim 5, wherein determining if activity associated with a selected frequency is external noise comprises:

determining if localized activity is near borders of one or more of the magnetic field sensors; and

determining the localized activity is noise based upon the localized activity being located near borders of one or more of the magnetic field sensors.

7. The method of claim 1, further comprising estimating coherence for the selected frequencies.

8. The method of claim 7, further comprising extracting coherent components at the selected frequencies.

9. The method of claim 8, further comprising assembling a partial spectrum comprising frequencies with similar patterns of the coherent components.

10. The method of claim 9, further comprising inverse transforming the partial spectrum to reconstruct a time series for functional entities producing the similar patterns of the coherent components.

11. A system comprising:

one or more processors configured to;

register a multichannel time series from data collected from a plurality of channels, wherein each channel corresponds to one of one or more sensors;

precise Fourier transform the data from all channels to transform the data in a frequency domain;

select frequencies from the frequency domain;

inverse Fourier transform for the selected frequencies from all channels;

select a restored time; and

localize activity for the selected frequencies at the restored time.

12. The system of claim 11, wherein the one or more processors are further configured to estimate coherence for the selected frequencies.

13. The system of claim 12, wherein the one or more processors are further configured to extract coherent components at the selected frequencies.

14. The system of claim 13, wherein the one or more processors are further configured to assemble a partial spectrum comprising frequencies with similar patterns of the coherent components.

15. The system of claim 14, wherein the one or more processors are further configured to inverse transform the partial spectrum to reconstruct a time series for functional entities producing the similar patterns of the coherent components.

16. A non-transitory computer readable medium having instructions stored thereon that, when executed by a computing device, cause the computing device to perform operations comprising:

registering a multichannel time series from data collected from a plurality of channels, wherein each channel corresponds to one of one or more magnetic field sensors;

precise Fourier transforming the data from all channels to transform the data in a frequency domain;

selecting frequencies from the frequency domain;

inverse Fourier transforming for the selected frequencies from all channels;

selecting a restored time; and

localizing activity for the selected frequencies at the restored time.

17. The non-transitory computer readable medium of claim 16, wherein the operations further comprise estimating coherence for the selected frequencies.

18. The non-transitory computer readable medium of claim 17, wherein the operations further comprise extracting coherent components at the selected frequencies.

19. The non-transitory computer readable medium of claim 18, wherein the operations further comprise assembling a partial spectrum comprising frequencies with similar patterns of the coherent components.

20. The non-transitory computer readable medium of claim 19, wherein the operations further comprise inverse transforming the partial spectrum to reconstruct a time series for functional entities producing the similar patterns of the coherent components.