Identification and implementation of locomotion modes using surface electromyography

• US 9,700,439 B1
• Filed: 08/05/2014
• Issued: 07/11/2017
• Est. Priority Date: 04/15/2008
• Status: Active Grant

• Current Assignee: Rehabilitation Institute of Chicago (Tenet Healthcare Corporation)
• Original Assignee: Rehabilitation Institute of Chicago (Tenet Healthcare Corporation)
• Inventors: Huang, He, Kuiken, Todd
• Primary Examiner(s): Woznicki, Jacqueline
• Application Number:
  US14/452,227
• Time in Patent Office: 1,071 Days
• CPC Class Codes: A61F 2/60 Artificial legs or feet or parts thereof
• CPC Class Codes: A61F 2/72 Bioelectric control, e.g. myoelectric
• CPC Class Codes: A61F 2/80 Sockets, e.g. of suction type
• CPC Class Codes: A61F 2002/607 Lower legs
• CPC Class Codes: A61F 2002/701 operated by electrically controlled means, e.g. solenoids or torque motors
• CPC Class Codes: A61F 2002/704 computer-controlled, e.g. robotic control
• CPC Class Codes: A61F 2002/7615 Measuring means
• CPC Class Codes: A61F 2002/762 for measuring dimensions, e.g. a distance
• CPC Class Codes: A61F 2002/7625 for measuring angular position
• CPC Class Codes: A61F 2002/7635 for measuring force, pressure or mechanical tension
• CPC Class Codes: A61F 2002/764 for measuring acceleration
• CPC Class Codes: A61F 2002/7645 for measuring torque, e.g. hinge or turning moment, moment of force

Abstract

Apparatus and methods are provided for determining a locomotion mode that can be provided to a controller of a lower prosthesis limb in order to accurately control the prosthesis. One or more prosthesis sensors are provided that break a gait cycle down into a plurality of gait phases. EMG sensors provide signals to a processor that directs them to a gait phase specific classifier that is used to determine a particular locomotion mode for the wearer. With the locomotion mode accurately known, the prosthetic device can be accurately controlled.

Claims

1. A method comprising:

affixing a plurality of prosthesis sensors to a lower limb prosthesis;

affixing or connecting a plurality of EMG sensors to the prosthesis;

determining, using a gait phase algorithm running on a processor, a gait phase based on an output of the prosthesis sensors;

measuring signals from the plurality of EMG sensors;

inputting the signals from the EMG sensors and information reflecting the gait phase to the processor;

determining, using a locomotion mode algorithm running on the processor, a locomotion mode based on the information reflecting the gait phase and the signals from the EMG sensors; and

providing the locomotion mode at the gait phase to an input of a controller of the lower-limb prosthesis,

wherein the locomotion mode algorithm can determine a level walking mode, an obstacle mode, a stair ascent mode, a stair descent mode, an ipsi-turn mode, a contra-turn mode, and a standing mode.

2. The method according to claim 1, further comprising:

storing in a memory the information reflecting the gait phase.

3. The method according to claim 1, further comprising:

converting the EMG sensor signals from analog to digital before inputting them to the processor.

4. The method according to claim 1, wherein the gait phase algorithm can determine a post-heel-contact phase, a pre-toe-off phase, a post-toe-off phase, and a pre-heel-contact phase.

5. The method according to claim 1, wherein the locomotion mode algorithm is configured as a pattern recognition classifier.

6. The method according to claim 5, wherein the pattern recognition classifier is a linear discriminant analysis-based classifier.

7. The method according to claim 5, wherein a classifier for the pattern recognition classifier is selected from the group consisting of:

a Gaussian mixed model, a multilayer perceptron network, a hidden Markov model, an artificial neural network, and a fuzzy logic classifier.

8. The method according to claim 5, wherein the pattern recognition classifier comprises a plurality of pattern recognition sub-classifiers, wherein each sub-classifier in the plurality of pattern recognition sub-classifiers is associated with a portion of gait.

9. The method according to claim 8, wherein the plurality of pattern recognition sub-classifiers comprises a post-heel-contact classifier, a pre-toe-off classifier, a post-toe-off classifier, and a pre-heel contact classifier.

10. The method according to claim 1, wherein the lower limb prosthesis is an above-knee lower limb prosthesis.

11. The method according to claim 1, wherein the signals from the EMG sensors input to the processor represents at least one EMG feature.

12. The method according to claim 11, wherein the at least one EMG feature is extracted during a phase window between about 120 ms and 200 ms in duration.

13. The method according to claim 11, wherein the at least one EMG feature comprises at least one time domain feature.

14. The method according to claim 13, wherein the at least one time domain feature represents a mean absolute value of the signals from the EMG sensors.

15. The method according to claim 13, wherein the at least one time domain feature represents a number of zero-crossings of the signals from the EMG sensors.

16. The method according to claim 13, wherein the at least one time domain feature represents a waveform length of the signals from the EMG sensors.

17. The method according to claim 13, wherein the at least one time domain feature represents a number of slope sign changes of the signals from the EMG sensors.