Signal quality metric for cardiovascular time series
A method and system for determining a signal quality metric of a cardiovascular time series utilizing a wireless sensor device are disclosed. In a first aspect, the method comprises determining subsequent values of the cardiovascular time series and comparing the determined subsequent values to a threshold value. In a second aspect, a wireless sensor device comprises a processor and a memory device coupled to the processor, wherein the memory device includes an application that, when executed by the processor, causes the processor to determine subsequent values of the cardiovascular time series and compare the determined subsequent values to a threshold value.
1. A method for determining a signal quality metric of a cardiovascular time series utilizing a wireless sensor device, the method comprising:
generating subsequent values of the cardiovascular time series by performing array-based calculations on the cardiovascular time series;
comparing the generated subsequent values to a threshold value and determining whether each individual time series value of the cardiovascular time series is a normal value or an artifact value by performing element-wise division between the cardiovascular time series and the subsequent values of the cardiovascular time series to produce the signal quality metric comprising an array of instantaneous time-series signal quality values;
performing interpolation on the signal quality metric to determine a uniformly sampled signal quality metric; and
relaying, using a transmitter, the uniformly sampled signal quality metric to another user or device.
2. The method of claim 1 , wherein the threshold value is determined from signal variability of the cardiovascular time series.
3. The method of claim 1 , further comprising:
comparing the generated subsequent values to a reference value in addition to the threshold value using a decision tree model to determine whether each individual time series value of the cardiovascular time series is a normal value or an artifact value.
4. The method of claim 3 , wherein the reference value is continuously updated within a predetermined window represented by a number of normal consecutive beats (Nb) using a first set of conditions.
5. The method of claim 4 , further comprising:
flagging at least one individual time series value of the cardiovascular time series if a second set of conditions is satisfied; and
identifying potential artifacts in the cardiovascular time series by comparing the at least one flagged individual time series value to the Nb.
6. The method of claim 5 , wherein if the at least one flagged individual time series value is greater than Nb, then the at least one flagged individual time series value is determined to be a normal value, but if the at least one flagged individual time series value is less than Nb, then the at least one flagged individual time series value is determined to be a potential artifact value.
7. The method of claim 6 , wherein the potential artifact value is confirmed as an artifact value if any of a third set of conditions is satisfied.
8. The method of claim 7 , further comprising:
updating the cardiovascular time series with 0 values for confirmed artifact values and with 1 values for confirmed normal values.
9. The method of claim 8 , further comprising:
formulating the signal quality metric utilizing the updated cardiovascular time series; and
uniformly sampling the signal quality metric using any of nearest neighbor and cubic interpolation.
10. A wireless sensor device for determining a signal quality metric of a cardiovascular time series, comprising:
a processor; and
a memory device coupled to the processor, wherein the memory device includes an application that, when executed by the processor, causes the processor to:
generate subsequent values of the cardiovascular time series by performing array-based calculations on the cardiovascular time series;
compare the generated subsequent values to a threshold value and determine whether each individual time series value of the cardiovascular time series is a normal value or an artifact value by performing element-wise division between the cardiovascular time series and the generated subsequent values of the cardiovascular time series to produce the signal quality metric comprising an array of instantaneous time-series signal quality values;
perform interpolation on the signal quality metric to determine a uniformly sampled signal quality metric; and
a transmitter to relay the uniformly sampled signal quality metric to another user or device.
11. The system of claim 10 , wherein the application, when executed by the processor, further causes the processor to:
compare the generated subsequent values to a reference value in addition to the threshold value using a decision tree model to determine whether each individual time series value of the cardiovascular time series is a normal value or an artifact value.
12. The system of claim 11 , wherein the reference value is continuously updated within a predetermined window represented by a number of normal consecutive beats (Nb) using a first set of conditions.
13. The system of claim 12 , wherein the application, when executed by the processor, further causes the processor to:
flag at least one individual time series value of the cardiovascular time series if a second set of conditions is satisfied; and
identify potential artifacts in the cardiovascular time series by comparing the at least one flagged individual time series value to the Nb.
14. The system of claim 13 , wherein if the at least one flagged individual time series value is greater than Nb, then the at least one flagged individual time series value is determined to be a normal value, but if the at least one flagged individual time series value is less than Nb, then the at least one flagged individual time series value is determined to be a potential artifact value.
15. The system of claim 14 , wherein the potential artifact value is confirmed as an artifact value if any of a third set of conditions is satisfied.
16. The system of claim 15 , wherein the application, when executed by the processor, further causes the processor to:
update the cardiovascular time series with 0 values for confirmed artifact values and with 1 values for confirmed normal values.
17. The system of claim 16 , wherein the application, when executed by the processor, further causes the processor to:
formulate the signal quality metric utilizing the updated cardiovascular time series; and
uniformly sample the signal quality metric using any of nearest neighbor and cubic interpolation.