System and method for automatically time labeling repetitive data

• US 9,654,234 B2
• Filed: 08/26/2016
• Issued: 05/16/2017
• Est. Priority Date: 08/28/2015
• Status: Active Grant

• Current Assignee: Globus Medical, Inc.
• Original Assignee: Focus Ventures, Inc.
• Inventors: Feng, Shuo
• Primary Examiner(s): Huang, David S
• Application Number:
  US15/249,115
• Publication Number:
  US 20170063475A1
• Time in Patent Office: 263 Days
• Field of Search: 375316
• CPC Class Codes: A61B 5/112 Gait analysis
• CPC Class Codes: A61B 5/1123 Discriminating type of movement, e.g. walking or running A61B5/1116, A61B5/112 take precedence
• CPC Class Codes: A61B 5/7225 Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation A61B5/30, A61B5/316 take precedence; transformers for specific applications H01F38/00; amplifiers H03F
• CPC Class Codes: A61B 5/7239 using differentiation including higher order derivatives
• CPC Class Codes: A61B 5/725 using specific filters therefor, e.g. Kalman or adaptive filters A61B5/316 take precedence
• CPC Class Codes: A61B 5/7267 involving training the classification device
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A63B 24/0003 Analysing the course of a movement or motion sequences during an exercise or trainings sequence, e.g. swing for golf or tennis
• CPC Class Codes: A63B 24/0062 Monitoring athletic performances, e.g. for determining the work of a user on an exercise apparatus, the completed jogging or cycling distance
• CPC Class Codes: G16H 50/70 for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: H04L 43/028 by filtering

Abstract

An exercise assistance system labels sensor signals received from a movement measurement device to identify repetitions within the sensor signals. The movement measurement device can transmit motion data describing the user'"'"'s movement to the exercise assistance system, and includes the sensor signals generated by one or more sensors in the movement measurement device. The exercise assistance system generates a combination signal from the received sensor signals and passes the combination signal through a low pass filter to generate a labeling signal. The exercise assistance system can generate three types of labels for the labeling signal identifying different parts of repetitions. The exercise assistance system labels the sensor signals using the generated labels and can identify repetitions in the labeled sensor signals using a classification model.

Claims

1. A method comprising:

receiving motion data from a motion measurement device, the motion data comprising a plurality of sensor signals from a plurality of sensors in the motion measurement device, the plurality of sensor signals describing a plurality of repetitions;

generating a combination signal by computing a combination of the plurality of sensor signals;

generating a labeling signal by passing the combination signal through a low pass filter;

generating a first order derivative signal by computing the first order derivative of the labeling signal;

generating a plurality of labels based on the labeling signal and the first order derivative signal, the plurality of labels comprising;

a first label that identifies a beginning of a repetition of the plurality of repetitions,

a second label that identifies an extremum of a repetition of the plurality of repetitions, and

a third label that identifies an end of a repetition of the plurality of repetitions; and

labeling the plurality of sensor signals based on the plurality of labels.

2. The method of claim 1, wherein the plurality of sensors includes at least one of an accelerometer, a gyroscope, a magnetometer, a GPS module, and an electric compass.

3. The method of claim 1, wherein the second label is generated by identifying a local extremum of the labeling signal.

4. The method of claim 1, wherein the first label and the third label of the plurality of labels are generated by identifying a local extremum of the first order derivative signal.

5. The method of claim 1, wherein labeling the sensor signals with the plurality of labels comprises identifying a plurality of label triplets, each label triplet comprising a first label of the plurality of labels that identifies a beginning of a repetition, a second label of the plurality of labels that identifies an extremum of the repetition, and a third label of the plurality of labels that identifies an end of the repetition, the first label, second label, and third label being chronologically sequential.

6. The method of claim 1, wherein the low pass filter comprises a cutoff frequency that is greater than an upper bound for a frequency of the plurality of repetitions.

7. The method of claim 1, wherein identifying the plurality of repetitions comprises applying a classification model that has been trained based on features extracted from the labeled sensor signals.

8. The method of claim 7, wherein identifying the plurality of repetitions comprises:

determining, by the classification model, whether a subset of the plurality of labels identifies a repetition of the plurality of repetitions; and

responsive to determining that the subset of the plurality of labels does not identify a repetition of the plurality of repetitions, identifying the subset of labels as identifying noise in the sensor signals.

9. The method of claim 1, wherein the movement measurement device is a wearable fitness device.

10. The method of claim 1, wherein each of the plurality of labels comprises a timestamp.

11. The method of claim 1, further comprising identifying the plurality of repetitions in the sensor signals based on the plurality of labels.

12. A non-transitory computer readable medium comprising instructions that, when executed by a processor, cause the processor to:

receive motion data from a motion measurement device, the motion data comprising a plurality of sensor signals from a plurality of sensors in the motion measurement device, the plurality of sensor signals describing a plurality of repetitions;

generate a combination signal by computing a combination of the plurality of sensor signals;

generate a labeling signal by passing the combination signal through a low pass filter;

generate a first order derivative signal by computing the first order derivative of the labeling signal;

generate a plurality of labels based on the labeling signal and the first order derivative signal, the plurality of labels comprising;

a first label that identifies a beginning of a repetition of the plurality of repetitions,

a second label that identifies an extremum of a repetition of the plurality of repetitions, and

a third label that identifies an end of a repetition of the plurality of repetitions;

label the plurality of sensor signals with the plurality of labels; and

identify the plurality of repetitions in the sensor signals based on the plurality of labels.

13. The computer readable medium of claim 12, wherein the plurality of sensors includes at least one of an accelerometer, a gyroscope, a magnetometer, a GPS module, and an electric compass.

14. The computer readable medium of claim 12, wherein the second label of the plurality of labels is generated by identifying a local extremum of the labeling signal.

15. The computer readable medium of claim 12, wherein the first label and the third label of the plurality of labels are generated by identifying a local extremum of the first order derivative signal.

16. The computer readable medium of claim 12, wherein the instructions for labeling the sensor signals with the plurality of labels further comprise instructions for identifying a plurality of label triplets, each label triplet comprising a first label of the plurality of labels that identifies a beginning of a repetition, a second label of the plurality of labels that identifies an extremum of the repetition, and a third label of the plurality of labels that identifies an end of the repetition, the first label, second label, and third label being chronologically sequential.

17. The computer readable medium of claim 12, wherein the low pass filter comprises a cutoff frequency that is greater than an upper bound for a frequency of the plurality of repetitions.

18. The computer readable medium of claim 12, wherein the instructions for identifying the plurality of repetitions comprises comprise instructions that, when executed, cause the processor to apply a classification model that has been trained based on features extracted from the labeled sensor signals.

19. The computer readable medium of claim 12, wherein the movement measurement device is a wearable fitness device.

20. The computer readable medium of claim 12, further comprising instructions that, when executed, cause the processor to identify the plurality of repetitions in the sensor signals based on the plurality of labels.