System and method for the analysis of biopotential signals using motion artifact removal techniques

• US 10,061,891 B2
• Filed: 04/14/2014
• Issued: 08/28/2018
• Est. Priority Date: 04/15/2013
• Status: Active Grant

• Current Assignee: Stichting Imec Nederland
• Original Assignee: Stichting IMEC Nederland
• Inventors: Grundlehner, Bernard, Mihajlovic, Vojkan
• Primary Examiner(s): Barbee, Manuel L
• Application Number:
  US14/252,611
• Publication Number:
  US 20140309943A1
• Time in Patent Office: 1,597 Days
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• 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: G16H 30/20 for handling medical images, e.g. DICOM, HL7 or PACS
• CPC Class Codes: G16H 30/40 for processing medical images, e.g. editing
• CPC Class Codes: G16H 50/20 for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

A system and method for the analysis of biopotential signals using motion artifact removal techniques is disclosed. The system includes a motion classification module configured to receive at least one biopotential signal and at least one reference secondary input signal. The motion classification module performs motion artifact classification for determining motion meta-information, comprising a type and/or severity indication about motion phenomena causing artifacts in the biopotential signal. The motion classification module communicates motion meta-information to a motion artifact reduction module configured to remove motion artifacts from the biopotential signal based on the information received from the motion classification module. The system is further configured to evaluate the effectiveness of at least one algorithm used to remove motion artifacts, and generate feedback information between the motion classification module and the motion artifact reduction module to optimize motion artifact classification.

Claims

1. A system for analysis of biopotential signals, comprising:

a motion classification module; and

a motion artifact reduction module,

wherein the motion classification module is configured to receive and store at least one biopotential signal and at least one reference secondary input signal, the motion classification module further configured to perform motion artifact classification for determining motion meta-information comprising a type and/or severity indication about motion phenomena producing artifacts in the biopotential signal that are capable of contaminating the biopotential signal and communicating said motion meta-information to the motion artifact reduction module;

wherein the motion artifact reduction module is configured to select one of a plurality of filtering techniques based on the information received from the motion classification module and perform motion artifact removal by filtering the received biopotential signal using the selected filtering technique;

wherein one or more filter or other motion artifact reduction settings are adjusted to remove the artifacts that result from motion occurring during measurement of the biopotential signals to generate clean signals, wherein an improvement of analysis of the biopotential signals is achieved as a result of using the clean signals to produce an output signal that compensates for the motion, the output signal thereby providing calculated results of the subject'"'"'s biosigns having improved accuracy;

wherein the motion classification module and the motion artifact reduction module are implemented with software and a programmable processor;

wherein the system is configured to evaluate an effectiveness of at least one algorithm used for motion artifact removal;

wherein the system is further configured to generate feedback information between the motion classification module and the motion artifact reduction module for optimizing performance of the motion artifact classification, the optimizing being automatically selected based on the motion meta-information extracted via the motion artifact classification; and

wherein the system produces an output based on the biopotential signals measured, the output comprising treatment data that is utilized to treat the subject, the treatment data being derived as a function of the calculated results.

2. The system of claim 1, wherein the system is further configured to store and process outputs of the motion classification module and the motion artifact reduction module in different periods of time in order to generate said feedback information for optimizing the performance of the motion artifact classification.

3. The system of claim 1, wherein evaluating the effectiveness of at least one algorithm used for motion artifact removal comprises comparing the biopotential signal after motion artifact removal with a reference baseline signal of such biopotential signal.

4. The system of claim 1, wherein the at least one reference secondary input signal is a signal comprising information about one or a combination of contact impedance, contact force, motion acceleration, temperature or humidity.

5. The system of claim 1, wherein the motion classification module is further configured to determine and communicate setting parameters for artifact removal to the motion artifact reduction module.

6. The system of claim 5, wherein the setting parameters comprise an artifact removal technique selection indication and/or coefficients related to a given artifact removal technique and/or signal selection indication for a given artifact removal technique.

7. The system of claim 1, wherein the motion artifact reduction module is further configured to apply, configure, and/or optimize a certain artifact removal technique based on information received from the motion classification module.

8. The system of claim 1, wherein the motion classification module and the motion artifact reduction module are configured for performing one or a combination of classification, statistical analysis, spectral analysis, cross-signal analysis, principal component analysis, independent component analysis, canonical component analysis, adaptive filtering, Bayesian filtering or empirical mode decomposition techniques.

9. The system of claim 1, further comprising a pre-processing module configured to adapt the received signals in order to be processed by the motion classification module and/or a post-processing module configured to adapt the output signals provided by the motion classification module and/or the motion artifact reduction module.

10. A method for the analysis of biopotential signals, comprising, in a system according to claim 1:

receiving and storing at least one biopotential signal;

receiving and storing at least one reference secondary input signal;

performing motion artifact classification for determining motion meta-information comprising a type and/or severity indication about motion phenomena causing motion artifacts in the biopotential signal;

selecting one of a plurality of filtering techniques based on that motion artifact classification;

performing removal of the motion artifacts from the biopotential signal by filtering the biopotential using the selected filtering technique, wherein the performing motion artifact classification and the performing removal of the motion artifacts are implemented with software and a programmable processor;

evaluating the effectiveness of the removal of the motion artifacts from the biopotential signal;

generating feedback information for optimizing performance of the motion artifact classification; and

treating the subject via the output generated based on the biopotential signals measured and the calculated results.

11. The method of claim 10, further comprising storing and processing outputs of the motion classification module and the motion artifact reduction module in different periods of time in order to generate said feedback information for optimizing the performance of the motion artifact classification.

12. The method of claim 10, wherein evaluating the effectiveness of the removal of the motion artifacts from the biopotential signal comprises comparing the biopotential signal after motion artifact removal with a reference baseline signal of such biopotential signal.

13. The method of claim 10, wherein the at least one reference secondary input signal is a signal comprising information about one or a combination of contact impedance, contact force, motion acceleration, temperature or humidity.

14. The method of claim 10, further comprising determining and communicating setting parameters for artifact removal to the motion artifact reduction module, wherein the setting parameters comprise an artifact removal technique selection indication and/or coefficients related to a given artifact removal technique and/or signal selection indication for a given artifact removal technique.

15. The method of claim 10, further comprising applying, configuring, and/or optimizing a certain artifact removal technique based on information received from the motion classification module.

16. The method of claim 10, further comprising performing one or a combination of classification, statistical analysis, spectral analysis, cross-signal analysis, principal component analysis, independent component analysis, canonical component analysis, adaptive filtering, Bayesian filtering or empirical mode decomposition techniques.

17. The system of claim 1 wherein extraction of the clean signals from the biopotential signals containing the artifacts contaminating the biopotential signals is performed based on the meta-information extracted from one or more reference secondary input signals, including electrode-to-tissue impedance, force applied to an electrode, contact force of the electrode, or acceleration of the electrode, and wherein the meta-information comprises signal properties, including standard deviation, slope, and/or higher-order statistics.

18. A system for the analysis of biopotential signals, comprising:

means for receiving at least one biopotential signal;

means for receiving at least one reference secondary input signal;

one or more storage devices that store the at least one biopotential signal and the at least one reference secondary input signal;

means for performing motion artifact classification for determining motion meta-information comprising a type and/or severity indication about motion phenomena causing motion artifacts in the biopotential signal;

means for selecting one of a plurality of filtering techniques based on that motion artifact classification;

means for performing removal of the motion artifacts from the biopotential signal by filtering the biopotential using the selected filtering technique; and

means for evaluating the effectiveness of the removal of the motion artifacts from the biopotential signal;

wherein feedback information is generated, based on at least one output of a motion artifact reduction module and/or a motion classification module for optimizing the performance of the motion artifact classification;

wherein at least the performing removal of the motion artifacts is implemented with software and a programmable processor;

wherein one or more filter or other motion artifact reduction settings are adjusted to remove the artifacts that result from motion occurring during measurement of the biopotential signals to generate clean signals, wherein an improvement of analysis of the biopotential signals is achieved as a result of using the clean signals to produce an output signal that compensates for the motion, the output signal thereby providing calculated results of the subject'"'"'s biosigns having improved accuracy; and

wherein the subject is treated by utilizing an output generated based on the biopotential signals measured, the output comprising treatment data that is derived as a function of the calculated results.

19. The system of claim 18, wherein the biopotential signal receiving means comprises a motion classification module, wherein the reference signal receiving means comprises the motion classification module, wherein the classification performing means comprises the motion classification module, wherein the removal performing means comprises a motion artifact reduction module, wherein the evaluating means comprises the motion artifact reduction module, and wherein the feedback information is generated via the motion artifact reduction module.

20. A non-transitory computer readable medium, comprising computer executable instructions for causing a processor to perform a method for the analysis of biopotential signals, the method comprising:

receiving at least one biopotential signal;

receiving at least one reference secondary input signal;

storing the at least one biopotential signal and the at least one reference secondary input signal;

performing motion artifact classification for determining motion meta-information comprising a type and/or severity indication about motion phenomena causing motion artifacts in the biopotential signal;

selecting one of a plurality of filtering techniques based on that motion artifact classification;

performing removal of the motion artifacts from the biopotential signal by filtering the biopotential using the selected filtering technique, wherein the performing motion artifact classification and performing removal of the motion artifacts are implemented with software and a programmable processor;

evaluating the effectiveness of the removal of the motion artifacts from the biopotential signal;

generating feedback information for optimizing the performance of the motion artifact classification, wherein one or more filter or other motion artifact reduction settings are adjusted to remove the artifacts that result from motion occurring during measurement of the biopotential signals to generate clean signals, wherein an improvement of analysis of the biopotential signals is achieved as a result of using the clean signals to produce an output signal that compensates for the motion, the output signal thereby providing calculated results of the subject'"'"'s biosigns having improved accuracy; and

treating the subject by utilizing an output generated based on the biopotential signals measured, the output comprising treatment data that is derived as a function of the calculated results.

21. The computer-readable medium of claim 20, the method further comprising:

evaluating outputs of the motion artifact removal algorithms with respect to expected signal properties; and

building a knowledge database regarding estimated effectiveness of the algorithms, the knowledge database storing information regarding how to optimize motion classification as well as motion artifact reduction.