Systems, methods and devices for exercise and activity metric computation

1. A method of determining exercise metrics in real-time for a user's body using a wearable device, the method comprising:

receiving a plurality of signals from at least one sensor of the wearable device while the user is performing an exercise activity, wherein the at least one sensor comprises an accelerometer and a plurality of EMG sensors, and wherein the plurality of signals comprises acceleration signals and EMG signals;

pre-processing the plurality of signals to generate a plurality of pre-processed signals;

using a processor, discriminating the plurality of pre-processed signals to generate a plurality of signal segments for each of the plurality of pre-processed signals, wherein generating the plurality of signal segments comprises generating a plurality of feature vectors based on the plurality of pre-processed signals, wherein each feature vector comprises a plurality of feature values, and wherein each feature vector is associated with a time step of the pre-processed signals;

using the processor, temporally correlating a first plurality of signal segments from the plurality of signal segments, the first plurality of signal segments based on acceleration signals, with a second plurality of signal segments from the plurality of signal segments, the second plurality of signal segments based on EMG signals;

computing at least one exercise metric while the user is performing the exercise activity based on the first and second plurality of signal segments as temporally correlated; and

outputting, via at least one of a display of the wearable device and a haptic feedback module of the wearable device, real-time feedback regarding the at least one exercise metric based on the computing.

2. The method of claim 1 , wherein at least two of the EMG sensors are positioned to detect EMG signals from different muscles of the user.

3. The method of claim 2 , wherein the at least one exercise metric comprises a muscle intensity metric.

4. The method of claim 2 , wherein the at least one exercise metric comprises a muscle coordination metric.

5. The method of claim 2 , wherein the at least one exercise metric comprises a muscle ratio metric.

6. The method of claim 2 , wherein the at least one exercise metric comprises a muscle fatigue metric, and wherein the computing comprises computing a shift in a frequency characteristic of at least one EMG signal.

7. The method of claim 2 , wherein generating the plurality of signal segments comprises determining a temporal distance between the plurality of signal segments.

8. The method of claim 1 , wherein the at least one sensor comprises a sensor selected from the group consisting of a gyroscope and a magnetometer.

9. The method of claim 1 , wherein each feature vector comprises time domain and frequency domain data based on the plurality of pre-processed signals.

10. The method of claim 1 , wherein each feature vector is computed over a moving time window.

11. The method of claim 1 , wherein generating the plurality of signal segments further comprises mapping the plurality of feature vectors onto a predetermined feature space.

12. The method of claim 11 , wherein the mapping is predetermined by ANOVA computation over a training data set.

13. The method of claim 11 , wherein the predetermined feature space is predetermined by Principal Component Analysis of a training data set.

14. The method of claim 1 , wherein computing at least one exercise metric comprises performing, using the processor, hierarchical classification to compute a likeliest class for a current time step.

15. A wearable device for determining exercise metrics in real-time for a user's body, the wearable device comprising:

at least one sensor positionable on the user's limb;

an output module comprising at least one of a display and a haptic feedback module; and

a processor operatively coupled to the at least one sensor and the output module, the processor configured to:

receive a plurality of signals from at least one sensor of the wearable device while the user is performing an exercise activity;

pre-process the plurality of signals to generate a plurality of pre-processed signals;

discriminate the plurality of pre-processed signals to generate a plurality of signal segments for each of the plurality of pre-processed signals, wherein generating the plurality of signal segments comprises generating a plurality of feature vectors based on the plurality of pre-processed signals, wherein each feature vector comprises a plurality of feature values, and wherein each feature vector is associated with a time step of the pre-processed signals;

temporally correlate a first plurality of signal segments from the plurality of signal segments, the first plurality of signal segments based on acceleration signals, with a second plurality of signal segments from the plurality of signal segments, the second plurality of signal segments based on EMG signals;

compute at least one exercise metric while the user is performing the exercise activity based on the first and second plurality of signal segments as temporally correlated;

output, via the output module, real-time feedback regarding the at least one exercise metric based on the computing.

16. A non-transitory computer readable medium storing computer-executable instructions, which, when executed by a computer processor, cause the computer processor to carry out a method of determining exercise metrics in real-time for a user's body using a wearable device, the method comprising:

receiving a plurality of signals from at least one sensor of the wearable device while the user is performing an exercise activity;

pre-processing the plurality of signals to generate a plurality of pre-processed signals;

discriminating the plurality of pre-processed signals to generate a plurality of signal segments for each of the plurality of pre-processed signals, wherein generating the plurality of signal segments comprises generating a plurality of feature vectors based on the plurality of pre-processed signals, wherein each feature vector comprises a plurality of feature values, and wherein each feature vector is associated with a time step of the pre-processed signals;

temporally correlating a first plurality of signal segments from the plurality of signal segments, the first plurality of signal segments based on acceleration signals, with a second plurality of signal segments from the plurality of signal segments, the second plurality of signal segments based on EMG signals;

computing at least one exercise metric while the user is performing the exercise activity based on the first and second plurality of signal segments as temporally correlated; and

outputting, via at least one of a display of the wearable device and a haptic feedback module of the wearable device, real-time feedback regarding the at least one exercise metric based on the computing.