System and method for motion detection using a PPG sensor
A photoplethysmography (PPG) circuit obtains PPG signals at one or more wavelengths. The PPG signal is processed to identify motion artifacts. The motion artifacts are correlated with predetermined PPG signal patterns associated with movement of a body part or a control command for a user device. The PPG signals may thus be used to detect movement of the body part or determine a control command. A user device may be controlled in response to the determined control command.
1. A biosensor, comprising:
a memory configured to store a database including a plurality of predetermined patterns and associated control commands for a device;
a sensor configured for positioning over an area of tissue of a user and configured to obtain an optical signal, wherein the optical signal includes a spectral response around a first wavelength of light reflected from the area of the tissue of the user;
a processing device configured to:
process the optical signal at the first wavelength to identify a first motion artifact at a first time interval;
correlate the first motion artifact to a first predetermined pattern of the plurality of predetermined patterns in the database and determine a first control command for the device associated with the first predetermined pattern using the database;
process the optical signal at the first wavelength to identify a second motion artifact at a second time interval; and
correlate the second motion artifact to a second predetermined pattern of the plurality of predetermined patterns in the database and determine a second control command for the device associated with the second predetermined pattern using the database.
2. The biosensor of claim 1 , wherein the first and second control commands are input to the device, wherein the device includes a user device or vehicle.
3. The biosensor of claim 2 , wherein the associated control commands include at least one of: selection of a key on a virtual keyboard, selection of an icon on a GUI, control movement of a cursor on a display of the user device, or control movement of the vehicle.
4. The biosensor of claim 1 , wherein the first motion artifact reflects changes in blood flow through vessels in the area of the tissue due to movement of a first body part at the first time interval and wherein the second motion artifact reflects changes in blood flow through vessels in the area of the tissue due to the movement of a second body part at the second time interval.
5. The biosensor of claim 4 , wherein the database further includes motion data of a body part associated with each of the plurality of predetermined patterns.
6. The biosensor of claim 5 , wherein the processing device is further configured to:
obtain first motion data for the first body part associated with the first predetermined pattern; and
obtain second motion data for the second body part associated with the second predetermined pattern.
7. The biosensor of claim 6 , wherein the first motion data includes an identification of the first body part, a direction of movement of the first body part and a speed of the movement of the first body part; and
wherein the second motion data includes an identification of the second body part, a direction of movement of the second body part and a speed of the movement of the second body part.
8. The biosensor of claim 7 , wherein the processing device is configured to:
obtain a direct current (DC) level or a maximum amplitude of the first motion artifact; and
determine a force of movement of the first body part using the DC level or the maximum amplitude of the first motion artifact.
9. The biosensor of claim 1 , wherein the sensor is included in a band configured for positioning over the area of the tissue of a wrist or arm of the user and comprises:
a photodetector configured to detect light from the area of the tissue of the wrist or arm of the user; and
at least one light source configured to emit light at the first wavelength.
10. The biosensor of claim 1 , wherein the processing device is configured to:
obtain a heart rate signal using the optical signal;
determine a quality factor of the optical signal; and
process the optical signal when the quality factor exceeds a predetermined threshold.
11. The biosensor of claim 1 , wherein the processing device is a neural network processing device.
12. A wearable device, comprising:
a sensor configured for positioning over an area of tissue of a user and configured to obtain an optical signal, wherein the optical signal includes a spectral response around a first wavelength of light reflected from the area of the tissue of the user;
a processing circuit in communication with the sensor and configured to:
determine a heart rate signal from the optical signal;
determine a quality factor of the heart rate signal;
when the quality factor exceeds a predetermined threshold, identify a first motion artifact at a first time interval in the optical signal and a second motion artifact at a second time interval in the optical signal;
compare the first and second motion artifacts to a plurality of predetermined patterns in a database, wherein each of the plurality of predetermined patterns is associated with a control command;
correlate the first motion artifact to a first predetermined pattern of the plurality of predetermined patterns in the database and determine a first control command associated with the first predetermined pattern using the database;
correlate the second motion artifact to a second predetermined pattern of the plurality of predetermined patterns in the database and determine a second control command associated with the second predetermined pattern using the database; and
a transmitter configured to transmit the first and second control commands to a user device for controlling the user device.
13. The wearable device of claim 12 , wherein the processing circuit is further configured to:
control the wearable device in response to the first and second control commands.
14. The wearable device of claim 12 , wherein the processing circuit is further configured to:
initiate a training mode for customization of the first predetermined pattern in a database;
request movement of a first body part of the user;
obtain an optical signal of the user during the requested movement of the first body part;
identify a motion artifact in the optical signal due to the requested movement of the body part; and
update the first predetermined pattern in the database using the motion artifact in the optical signal.
15. The wearable device of claim 12 , wherein the first motion artifact reflects changes in blood flow through vessels in the area of the tissue due to movement of a first body part at the first time interval and wherein the second motion artifact reflects changes in blood flow through vessels in the area of the tissue due to the movement of a second body part at the second time interval.
16. The wearable device of claim 12 , wherein the processing circuit is further configured to:
when the quality factor of the heart rate signal is not within the predetermined threshold, generate a request to reposition the wearable device.
17. The wearable device of claim 12 , wherein each of the plurality of predetermined patterns in the database is further associated with motion data of a body part, wherein the motion data includes an identification of the body part, a direction of movement of the body part and a speed of the movement of the body part.
18. A biosensor, comprising:
a sensor configured to obtain an optical signal, wherein the optical signal includes a spectral response around a wavelength of light detected from tissue of a user;
a processing device configured to:
identify a fluctuation in the optical signal, wherein the fluctuation in the optical signal reflects neural activity or vasodilation of vessels in the tissue of the user;
correlate the fluctuation in the optical signal to a predetermined signal pattern, wherein the predetermined signal pattern is associated with an intended movement of a body part due to a mental initiating of movement with little to no actual movement of the body part;
determine a control command associated with the intended movement of the body part; and
control operation of a device in response to the control command.
19. The biosensor of claim 18 , wherein the processing circuit is configured to obtain the predetermined signal pattern by:
initiating a training mode for the biosensor;
generating by the device a request to the user to mentally intend to move the body part without actual movement of the second body part;
obtaining a training optical signal around the wavelength of light in response to the request;
identifying a fluctuation in the training optical signal; and
update the predetermined signal pattern using the fluctuation.