Robust detection of a change in a physiological metric
Embodiments are directed to systems and methods for determining a change in a physiological metric over a target period of time. A plurality of data sets may be used to generate a plurality of contextual signals. Each of these contextual signals may be associated with a different activity context, and may be used to generate a plurality of instances of the physiological metric. Information about the amount of change of each instance of the physiological metric may be used to estimate an overall amount of change in the physiological metric over the target period of time.
1 . A system for estimating an overall change in a physiological metric over a target period of time, the system comprising:
a sensor unit comprising a set of sensors and configured to output a plurality of data sets associated with the target period of time;
a signal generator configured to generate a plurality of contextual signals using the plurality of data sets, wherein each contextual signal of the plurality of contextual signals is associated with a different corresponding activity context of a plurality of activity contexts;
a metric generator configured to generate a plurality of metric data sets, wherein each metric data set includes multiple values of the physiological metric calculated using a corresponding contextual signal of the plurality of contextual signals; and
a change detection unit configured to generate an estimate of the overall change in the physiological metric over the target period of time using the plurality of metric data sets.
2 . The system of claim 1 , wherein each contextual signal comprises a corresponding first data subset selected from a first data set of the plurality of data sets.
3 . The system of claim 2 , wherein the first data set includes data values of a heart rate data type.
4 . The system of claim 2 , wherein each contextual signal comprises a corresponding second data subset selected from a second data set of the plurality of data sets.
5 . The system of claim 4 , wherein the second data set includes data values of a motion metric data point.
6 . The system of claim 1 , further comprising a normalization unit configured to normalize each of the plurality of contextual signals using a corresponding set of factors, wherein the metric generator receives the plurality of contextual signals after they have been normalized by the normalization unit.
7 . The system of claim 6 , wherein the normalization unit includes a plurality of stages, each of which is configured to perform a different normalization operation on each of the plurality of contextual signals.
8 . The system of claim 1 , wherein the change detection unit comprises a metric change unit configured to generate a plurality of sets of change metrics using the plurality of metric data sets, wherein each set of change metrics is associated with a different corresponding activity context of the plurality of activity contexts.
9 . The system of claim 8 , wherein each set of change metrics of the plurality of sets of change metrics comprises an estimated rate of change of the multiple values of the physiological metric of the corresponding activity context.
10 . The system of claim 8 , wherein the change detection unit further comprises a overall change unit configured to generate the estimate of the overall change of the physiological metric over the target period of time using the plurality sets of change metrics.