Method for processing brainwave signals

• US 10,080,506 B2
• Filed: 07/30/2015
• Issued: 09/25/2018
• Est. Priority Date: 08/02/2010
• Status: Active Grant

• Current Assignee: Chi Yung Fu
• Original Assignee: Chi Yung Fu
• Inventors: Fu, Chi Yung
• Primary Examiner(s): HULBERT, AMANDA K
• Application Number:
  US14/814,364
• Publication Number:
  US 20150327813A1
• Time in Patent Office: 1,153 Days
• Field of Search: 607 45, 600544
• CPC Class Codes: A61B 2560/0214 of power generation or supply
• CPC Class Codes: A61B 2562/046 in a matrix array
• CPC Class Codes: A61B 2562/14 Coupling media or elements to improve sensor contact with skin or tissue
• CPC Class Codes: A61B 2576/026 for the brain
• CPC Class Codes: A61B 5/0036 including treatment, e.g., using an implantable medical device, ablating, ventilating
• CPC Class Codes: A61B 5/0042 for the brain
• CPC Class Codes: A61B 5/0075 by spectroscopy, i.e. measuring spectra, e.g. Raman spectroscopy, infrared absorption spectroscopy A61B5/0071 takes precedence
• CPC Class Codes: A61B 5/165 Evaluating the state of mind, e.g. depression, anxiety
• CPC Class Codes: A61B 5/246 using evoked responses
• CPC Class Codes: A61B 5/291 for electroencephalography [EEG]
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/369 Electroencephalography [EEG] devices for psychotechnics A61B5/16
• CPC Class Codes: A61B 5/377 using evoked responses
• CPC Class Codes: A61B 5/6803 Head-worn items, e.g. helmets, masks, headphones or goggles
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7235 Details of waveform analysis A61B5/349 takes precedence
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• 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: A61N 1/0408 Use-related aspects
• CPC Class Codes: A61N 1/0456 : Specially adapted for transcutaneous electrical nerve stimulation [TENS]
• CPC Class Codes: A61N 1/0476 : Array electrodes (including any electrode arrangement with more than one electrode for at least one of the polarities)
• CPC Class Codes: A61N 1/0526 : Head electrodes A61N1/0551 takes precedence
• CPC Class Codes: A61N 1/36021 : for treatment of pain
• CPC Class Codes: A61N 1/36025 : for treating a mental or cerebral condition
• CPC Class Codes: A61N 1/36031 : using physiological parameters for adjustment
• CPC Class Codes: A61N 1/40 : Applying electric fields by inductive or capacitive coupling microwave apparatus A61N5/00; Applying radio-frequency signals
• CPC Class Codes: A61N 2/006 : for magnetic stimulation of nerve tissue
• CPC Class Codes: A61N 2/008 : for pain treatment or analgesia
• CPC Class Codes: A61N 2/02 : using magnetic fields produced by coils, including single turn loops or electromagnets A61N2/12 takes precedence
• CPC Class Codes: A61N 5/02 : using microwaves
• CPC Class Codes: A61N 5/10 : X-ray therapy; Gamma-ray therapy; Particle-irradiation therapy A61N5/01 takes precedence
• CPC Class Codes: A61N 7/00 : Ultrasound therapy lithotripsy A61B17/22, A61B17/225; massage using supersonic vibration A61H23/00 ; using ultrasound for introducing media into the body A61M37/0092
• CPC Class Codes: G16H 20/40 : relating to mechanical, radiation or invasive therapies, e.g. surgery, laser therapy, dialysis or acupuncture
• CPC Class Codes: G16H 20/70 : relating to mental therapies, e.g. psychological therapy or autogenous training
• 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

A signal processing method and system combines multi-scale decomposition, such as wavelet, pre-processing together with a compression technique, such as an auto-associative artificial neural network, operating in the multi-scale decomposition domain for signal denoising and extraction. All compressions are performed in the decomposed domain. A reverse decomposition such as an inverse discrete wavelet transform is performed on the combined outputs from all the compression modules to recover a clean signal back in the time domain. A low-cost, non-drug, non-invasive, on-demand therapy braincap system and method are pharmaceutically nonintrusive to the body for the purpose of disease diagnosis, treatment therapy, and direct mind control of external devices and systems. It is based on recognizing abnormal brainwave signatures and intervenes at the earliest moment, using magnetic and/or electric stimulations to reset the brainwaves back to normality. The feedback system is self-regulatory and the treatment stops when the brainwaves return to normal. The braincap contains multiple sensing electrodes and microcoils; the microcoils are pairs of crossed microcoils or 3-axis triple crossed microcoils.

Claims

1. A method for processing brainwave signals of a patient having a neurological or mental disorder, the method comprising:

receiving, by one or more processors, a brainwave signal obtained by a device coupled to a patient, the brainwave signal containing noise;

performing, by the one or more processors, an n-level decomposition of the brainwave signal into wavelet components of different scales to separate the brainwave signal from the noise;

for each individual wavelet component among the wavelet components of different scales, inputting, by the one or more processors, the individual wavelet component into a corresponding compression module to squeeze out the noise from the individual wavelet component in a decomposed domain, wherein the compression module is self-supervised and has been trained by creating an output behavior of the compression module that closely matches an input behavior of the compression module;

performing, by the one or more processors, an inverse decomposition on outputs from the compression modules to recover a clean brainwave signal in a time domain; and

causing, by the one or more processors, physiological stimulation of the patient with signals generated based on the neurological or mental disorder of the patient as determined by comparing the clean brainwave signal to at least one of a plurality of normal or abnormal brainwave signals stored in a database, wherein each abnormal brainwave signal is indicative of a neurological or mental disorder.

2. The method of claim 1, wherein the n-level decomposition is a wavelet decomposition into different scales, or a multi-resolution pyramid decomposition into different scales.

3. The method of claim 1, wherein the compression modules are vector quantizers or auto-associative neural networks.

4. The method of claim 1, wherein the method is implemented in a data processing or analyzing system, or as firmware or hardware that includes an application-specific integrated circuit (ASIC), a full-custom chip, a digital signal processing (DSP) chip, or a graphic processor chip (GPC) for real time processing.

5. The method of claim 1, wherein the n-level decomposition is a multi-resolution pyramid scheme performing successive averaging to obtain smoother scales as decomposed signal moves up the pyramid and scales with lower resolutions.

6. The method of claim 5, wherein the successive averaging is performed by multiple neural networks.

7. The method of claim 5, further comprising:

performing a preliminary denoise process by eliminating components from lower parts of the multi-resolution pyramid.

8. The method of claim 5, wherein the n-level decomposition using artificial neural networks in the multi-resolution pyramid scheme and the compression modules provide resilience in the performance of the process of the signal through a flexibility of denoising in the multi-resolution pyramid scheme.

9. The method of claim 2, wherein the decomposed components include a highest-order smooth component and additional rough components.

10. The method of claim 9, further comprising:

performing a preliminary denoising process by retaining only the smoothest component thus allowing quick assessment of signal of interest to determine whether further processing is needed.

11. The method of claim 9, further comprising:

performing a preliminary denoising process by eliminating a plurality of lower-level rough components from the wavelet stage or the multi-resolution pyramid scheme resulting in performing both decomposition for subsequent neural network denoising and early denoising function in one single procedure.

12. The method of claim 9, further comprising:

performing a preliminary denoising process by eliminating a plurality of lower-level components or selective rough components, enabling progressive signal processing which selectively trades off between extracted signal quality and processing complexity and speed;

wherein the preliminary denoising process impacts real-time processing with capability of adding more of eliminated components to improve signal quality.

13. The method of claim 1, wherein the n-level decomposition and the compression modules provide resilience to performance of the overall process of the signal by permitting suboptimal denoising in the decomposition stage that may be compensated in a subsequent compression stage.

14. The method of claim 1, wherein the brainwave signal includes an Electroencephalography (EEG) signal, an infrared (IR) signal, a magnetometer signal from an alkali vapor magnetometer or a superconducting quantum interference device (SQUID), or a signal from other sensor that can sense brain activities.

15. The method of claim 1, wherein the decomposition and compression allow for efficient data storage because relatively intense noise over trace signals occupying more storage space has been removed.

16. The method of claim 1, wherein achieving optimization control is through gradient descent.

17. The method of claim 1, wherein the brainwave signal is one dimension of a multi-dimensional signal that has data of multiple dimensions.

18. A method for processing brainwave signals for treating a patient having a neurological or mental disorder, comprising:

receiving, by one or more processors, a brainwave signal obtained by a cap or pad placed on a head of the patient, the brainwave signal containing noise, and the cap or pad including a plurality of microelectrodes or microsensors for measuring the brainwave signals from the patient and a plurality of coils for applying therapeutic treatment to the patient;

performing, by the one or more processors, an n-level decomposition of the brainwave signal into components of different scales;

for each individual component among the components of different scales, inputting, by the one or more processors, the individual component into a corresponding compression module to squeeze out the noise from the individual component in a decomposed domain, wherein the compression module is self-supervised and has been trained by creating an output behavior of the compression module that closely matches an input behavior of the compression module;

performing, by the one or more processors, an inverse decomposition on outputs from the compression modules to recover a clean brainwave signal in a time domain; and

causing, by the one or more processors, physiological stimulation of the patient via the plurality of coils with signals generated based on the neurological or mental disorder of the patient as determined by comparing the clean brainwave signal to at least one of a plurality of normal or abnormal brainwave signals stored in a database, wherein each abnormal brainwave signal is indicative of a neurological or mental disorder.

19. The method of claim 18, further comprising:

supplying power to the microelectrodes and microsensors of the cap using radiofrequency (RF) coupling to operate the cap without using a battery.

20. The method of claim 18, wherein the coils of the cap are operated in synchrony as a phased array for providing deep brain stimulation.

21. The method of claim 18, wherein the plurality of coils work at lower power in tandem in a phased-array arrangement to minimize chance of heating up electrodes and electronic circuits.

22. The method of claim 18, wherein the microelectrodes and microsensors are in close proximity to treatment devices to allow effective local feedback for therapy.

23. The method of claim 18, wherein the coils of the cap operate at high magnetic fields on the order of one Tesla at a frequency of less than 20 Hz.

24. The method of claim 18, wherein the coils of the cap operate at low magnetic fields on the order of 1000 gauss or less, at a frequency of from 100 Hz to 10 KHz.

25. The method of claim 18, wherein the coils of the cap include large coil bands or small cross coils or a combination thereof that work in tandem and are driven by phased-array driving electronics to improve deep brain stimulation.

26. The method of claim 18, wherein the transcranial direct current stimulation (tDCS) system or magnetic transcranial stimulation (MRS) with coils for the cap include large coil bands or small cross coils of a combination thereof that work in tandem to function as nerve block for pain treatment without surgically severing nerves.

27. The method of claim 26, wherein the nerves are in a primary somatic area (SI), a second somatic area (SII), an insular cortex (IC), a prefrontal cortex (PFC), or a parietal cortices.

28. The method of claim 18, wherein the cap or coating of the coils includes material with low thermal conductivity to reduce heat damages to the skin.

29. The method of claim 18, wherein the cap or pad includes soft body-conforming material to allow precise coil-to-skull or coil-to-body separation and spatial positioning to give better control and adaptation to different head shapes or head sizes or body shapes or body sizes.

30. A method for processing body-generated signals of a patient having a disorder, the method comprising:

receiving, by one or more processors, a body-generated signal obtained by a device coupled to a patient, the body-generated signal containing noise;

performing, by the one or more processors, an n-level decomposition of the body-generated signal into components of different scales to separate the body-generated signal from the noise;

for each individual component among the components of different scales, inputting, by the one or more processors, the individual component into a corresponding compression module to squeeze out the noise from the individual component in a decomposed domain, wherein the compression module is self-supervised and has been trained by creating an output behavior of the compression module that closely matches an input behavior of the compression module;

performing, by the one or more processors, an inverse decomposition on outputs from the compression modules to recover a clean body-generated signal in a time domain; and

causing, by the one or more processors, physiological stimulation of the patient with signals generated based on the disorder of the patient as determined by comparing the clean body-generated signal to at least one of a plurality of normal or abnormal body-generated signals stored in a database, wherein each abnormal body-generated signal is indicative of a disorder.

31. The method of claim 30, wherein the body-generated signal is an Electroencephalography (EEG) signal, an electrocardiogram (EKG) signal, or an internal emission signal from a human body.

32. The method of claim 30, wherein the body-generated signal is induced when the patient'"'"'s body responds to external stimulation, the external stimulation being X-ray, magnetic field, radioactive-nuclide treatment, microwave, radio-frequency wave, ultrasound, or a stimulation that provokes a human body to response.