Detecting and Using Body Tissue Electrical Signals

Abstract

Systems and methods for gesture control are disclosed. In some embodiments, a system may include a plurality of electrode pairs, a motion sensor, a controller, and a classifier. The system may be configured to: enter a monitoring state in which the system is configured to receive data; receive a first set of data; determine that the first set of data does not satisfy one or more action criteria; return to the monitoring state without transmitting the first set of data to the classifier; receive a second set of data; determine that the second set of data satisfies the one or more action criteria; transmit the second set of data to the classifier; using the classifier, analyze the second set of data to generate an interpreted output indicating a gesture performed by the person; and based on the interpreted output, generate a machine instruction.

1 . A system for biopotential-based gesture control, the system comprising:
 a plurality of electrode pairs, each electrode pair defining a channel configured to receive electrical signals at a respective position on a portion of a body of a person;   a motion sensor, the motion sensor being configured to generate a motion sensor output indicating a motion of the portion of the body of the person;   a controller, the controller being configured to control processing of the electrical signals received by the plurality of electrode pairs;   a classifier, the classifier being configured to analyze data derived from the plurality of electrode pairs and the motion sensor and, based on this analysis, generate interpreted outputs indicating gestures performed by the person;   wherein the system is configured to:   enter a monitoring state in which the system is configured to receive data from the plurality of electrode pairs and the motion sensor;   receive a first set of data, the first set of data being derived from the electrical signals of one or more of the plurality of electrode pairs and/or the motion sensor output;   determine that the first set of data does not satisfy one or more action criteria;   in response to determining that the first set of data does not satisfy the one or more action criteria, return to the monitoring state without transmitting the first set of data to the classifier;   receive a second set of data, the second set of data being derived from the electrical signals of one or more of the plurality of electrode pairs and/or the motion sensor output;   determine that the second set of data satisfies the one or more action criteria;   in response to determining that the second set of data satisfies the one or more action criteria, transmit the second set of data to the classifier;   using the classifier, analyze the second set of data to generate an interpreted output indicating a gesture performed by the person; and   based on the interpreted output, generate a machine instruction.   
     
     
         2 . The system of  claim 1 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that an amplitude of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         3 . The system of  claim 2 , wherein the threshold is determined based on electrical noise in an environment of the plurality of electrode pairs. 
     
     
         4 . The system of  claim 1 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that a duration of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         5 . The system of  claim 1 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that a measured parameter of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         6 . The system of  claim 1 , wherein:
 the system is further configured to compare a first measurement from a first electrode pair of the plurality of electrode pairs to a second measurement from a second pair of the plurality of electrode pairs; and   determining that the second set of data satisfies the one or more action criteria is based, at least in part, on the comparison between the first measurement and the second measurement.   
     
     
         7 . The system of  claim 6 , wherein the system is configured to assess, based on the comparison between the first measurement and the second measurement, that a phenomenon captured in both the first measurement and the second measurement is a neuromuscular activation intended by the person. 
     
     
         8 . The system of  claim 7 , wherein the comparison between the first measurement and the second measurement comprises determining that the phenomenon registered at the first electrode pair earlier than at the second electrode pair. 
     
     
         9 . The system of  claim 7 , wherein the comparison between the first measurement and the second measurement comprises determining that the phenomenon had a larger amplitude as measured at the first electrode pair than at the second electrode pair. 
     
     
         10 . The system of  claim 1 , wherein:
 the plurality of electrode pairs, the motion sensor, and the controller are disposed on a wearable device configured to be worn on a wrist of the person; and   the classifier is disposed in a smartphone, the wearable device being configured to wirelessly transmit the second data to the smartphone.   
     
     
         11 . A method for biopotential-based gesture control, the method being performed using a system comprising a plurality of electrode pairs, a motion sensor, a controller, and a classifier, the method comprising:
 entering a monitoring state in which the system is configured to receive data from the plurality of electrode pairs and the motion sensor, each electrode pair of the plurality of electrode pairs defining a channel configured to receive electrical signals at a respective position on a portion of a body of a person, and the motion sensor being configured to generate a motion sensor output indicating a motion of the portion of the body of the person;   receiving a first set of data, the first set of data being derived from the electrical signals of one or more of the plurality of electrode pairs and/or the motion sensor output;   using the controller, determining that the first set of data does not satisfy one or more action criteria, the controller being configured to control processing of the electrical signals received by the plurality of electrode pairs;   in response to determining that the first set of data does not satisfy the one or more action criteria, returning to the monitoring state without transmitting the first set of data to the classifier;   receiving a second set of data, the second set of data being derived from the electrical signals of one or more of the plurality of electrode pairs and/or the motion sensor output;   using the controller, determining that the second set of data satisfies the one or more action criteria;   in response to determining that the second set of data satisfies the one or more action criteria, transmitting the second set of data to the classifier, the classifier being configured to analyze data derived from the plurality of electrode pairs and the motion sensor and, based on this analysis, generate interpreted outputs indicating gestures performed by the person;   using the classifier, analyze the second set of data to generate an interpreted output indicating a gesture performed by the person; and   based on the interpreted output, generate a machine instruction.   
     
     
         12 . The method of  claim 11 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that an amplitude of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         13 . The method of  claim 12 , wherein the threshold is determined based on electrical noise in an environment of the plurality of electrode pairs. 
     
     
         14 . The method of  claim 11 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that a duration of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         15 . The method of  claim 11 , wherein determining that the second set of data satisfies the one or more action criteria comprises determining that a measured parameter of one or more signals received by the plurality of electrode pairs exceeds a threshold. 
     
     
         16 . The method of  claim 11 , wherein:
 the method further comprises comparing a first measurement from a first electrode pair of the plurality of electrode pairs to a second measurement from a second pair of the plurality of electrode pairs; and   determining that the second set of data satisfies the one or more action criteria is based, at least in part, on the comparison between the first measurement and the second measurement.   
     
     
         17 . The method of  claim 16 , wherein the method further comprises assessing, based on the comparison between the first measurement and the second measurement, that a phenomenon captured in both the first measurement and the second measurement is a neuromuscular activation intended by the person. 
     
     
         18 . The method of  claim 17 , wherein the comparison between the first measurement and the second measurement comprises determining that the phenomenon registered at the first electrode pair earlier than at the second electrode pair. 
     
     
         19 . The method of  claim 17 , wherein the comparison between the first measurement and the second measurement comprises determining that the phenomenon had a larger amplitude as measured at the first electrode pair than at the second electrode pair. 
     
     
         20 . The method of  claim 11 , wherein:
 the plurality of electrode pairs, the motion sensor, and the controller are disposed on a wearable device configured to be worn on a wrist of the person; and   the classifier is disposed in a smartphone, the wearable device being configured to wirelessly transmit the second data to the smartphone.