System and method of biomechanical posture detection and feedback

1. A method of providing postural feedback comprising:

receiving by a microprocessor at repeated intervals data from a tri-axial accelerometer, the microprocessor and the tri-axial accelerometer comprising a sensor device attached to a user, the sensor device further comprising memory, an actuator, and a power source;

normalizing by the microprocessor the received accelerometer data;

determining by the microprocessor a postural description of the user based on the normalized received accelerometer data, wherein the postural description comprises a movement state and posture quality, wherein determining the postural description of the user comprises:

computing a sequence of accelerometer data features from the normalized received accelerometer data,

analyzing the sequence of accelerometer data features according to a rule-based model to determine, in part, the movement state and the postural description; and

triggering by the microprocessor the actuator to output sensory feedback based on the postural description of the user.

2. The method of claim 1 further comprising transmitting across a wireless electronic communication connection to a computing device the postural description of the user and the normalized received accelerometer data.

3. The method of claim 1 wherein determining the postural description based on the normalized received accelerometer data further comprises:

computing feature vectors for the normalized received accelerometer data;

quantizing the feature vectors;

assigning the feature vectors to Gaussian distributions;

acquiring distribution histograms for multiple movement states; and

comparing the distribution histograms to determine a current movement state from the multiple movement states.

4. The method of claim 1 wherein determining the postural description based on the normalized received accelerometer data comprises calculating a pelvic tilt angle of the user.

5. The method of claim 1 wherein determining the postural description of the user comprises determining a movement state of the postural description from at least the rule-based model and a statistical model that is based on the normalized received accelerometer data.

6. The method of claim 1 wherein determining the postural description of the user comprises determining a movement state of the postural description from at least the rule-based model and a library comparison model that is based on the normalized received accelerometer data.

7. The method of claim 1 wherein a pelvic tilt angle model is used in part to determine the postural description of the user based on the normalized received accelerometer data.

8. A postural feedback apparatus comprising:

a sensor device configured to be attached on a user, the sensor device comprising:

a tri-axial accelerometer;

an actuator; and

a microprocessor configured to

receive data from the tri-axial accelerometer about movement of the user;

normalize the received accelerometer data;

determine a postural description of the user based on the normalized received accelerometer data, wherein the postural description comprises a movement state and posture quality, wherein the microprocessor is further configured to:

compute a sequence of accelerometer data features from the normalized accelerometer data,

analyze the sequence of accelerometer data features according to a rule-based model to determine in part the movement state and the postural description; and

trigger the actuator to output sensory feedback based on the postural description of the user.

9. The apparatus of claim 8 wherein the sensor device further comprises a wireless communication module.

10. The apparatus of claim 8 wherein the microprocessor is further configured to communicate with the wireless communication module to transmit the postural description of the user and the normalized received accelerometer data to a computing device.

11. The apparatus of claim 8 wherein the microprocessor configured to determine the postural description of the user based on the normalized received accelerometer data comprises the microprocessor configured to

compute feature vectors for the normalized received accelerometer data;

quantize the feature vectors;

assign the feature vectors to Gaussian distributions;

acquire estimated distribution histograms for multiple movement states; and

compare the distribution histograms to determine the current movement state from the multiple movement states.

12. The apparatus of claim 8 wherein the microprocessor configured to determine the postural description of the user based on the normalized received accelerometer data additionally uses a statistical model.

13. The apparatus of claim 8 wherein the microprocessor configured to determine the postural description of the user based on the normalized received accelerometer data additionally uses a library comparison model.

14. The apparatus of claim 8 wherein the microprocessor configured to determine the postural description of the user based on the normalized received accelerometer data additionally uses a pelvic tilt angle model.

15. A non-transitory computer readable medium having stored thereupon computing instructions comprising:

a code segment to receive by a microprocessor at repeated intervals data from a tri-axial accelerometer, the microprocessor and the tri-axial accelerometer comprising a sensor device attached to a user, the sensor device further comprising memory, an actuator, and a power source;

a code segment to normalize by the microprocessor the received accelerometer data;

a code segment to determine by the microprocessor a postural description of the user based on the normalized received accelerometer data wherein the code segment to determine the postural description comprises:

a code compute a sequence of accelerometer data features from the normalized received accelerometer data,

a code segment to analyze the sequence of accelerometer data features with a rule-based model and determine the movement state and the postural quality; and

a code segment to trigger by the microprocessor the actuator to output sensory feedback based on the postural description of the user.

16. The non-transitory computer readable medium of claim 15 having stored thereupon computing instructions further comprising:

a code segment to transmit from the microprocessor across an electronic communication connection to a computing device the postural description of the user to be displayed on the computing device as an avatar with an appearance, the appearance of the avatar representing the postural description of the user.

17. The non-transitory computer readable medium of claim 16 having stored thereupon computing instructions further comprising:

a code segment to transmit from the computing device across a network connection to a cloud server the postural description of the user, the cloud server using the postural description of the user to tailor biofeedback for the user.

18. The method of claim 1 wherein a pelvic tilt angle model is used to detect at least a lying state.

19. The method of claim 1 wherein the rule-based model determines the transition between at least a sitting state and a standing state.

20. The method of claim 1 , wherein computing a sequence of accelerometer data features comprises computing a power value; and wherein the rule-based model to determines the movement state comprises identifying a moving state from a first power value threshold and identifying a stationary state from a second power threshold.

21. The method of claim 1 , wherein the rule-based model is used in part to determine at least a standing state, sitting state, lying state, running state, and walking state.