Synchronizing sensors using heart rate signals
A method of synchronizing output signals from a plurality of sensors each measuring a heart rate derivable physiological signal of a subject along with a physiological signal of interest includes acquiring first and second output signals using respective first and second sensors, where each sensor includes an independent clock for associating its output signal with a respective time domain. The method further includes determining first and second heart rate sequence signals from the first and second output signals, identifying timings of at least one heart rate event in the respective first and second heart rate sequence signals, and determining a first time domain transformation for the time domain of the second output signal to the time domain of the first output signal using a model fitted to the timings of the heart rate event in the respective first and second heart rate sequence signals.
1 . A method of synchronizing output signals from a plurality of sensors each measuring a heart rate derivable physiological signal of a subject along with a physiological signal of interest, comprising the steps of:
acquiring a first output signal using a first sensor configured to measure a heartbeat derivable physiological signal of a subject along with a first physiological signal of interest;
acquiring a second output signal using a second sensor configured to measure a heartbeat derivable physiological signal of the subject along with a second physiological signal of interest,
wherein each of the first and second sensor comprises an independent clock for associating the signals measured by the first and second sensors with a respective time domain, the method further comprising the steps of:
determining a first heart rate sequence signal from the first output signal;
determining a second heart rate sequence signal from the second output signal;
identifying timings of at least one heart rate event in the respective first and second heart rate sequence signals; and
determining a first time domain transformation for the time domain of the second output signal to the time domain of the first output signal using a model fitted to the timings of the at least one heart rate event in the respective first and second heart rate sequence signals,
capturing a first physiological signal of interest, using the first sensor that is different from the first output signal;
capturing a second physiological signal of interest using the second sensor that is different from the second output signal; and
synchronizing the second physiological signal of interest to the first physiological signal of interest by applying model parameters of the model to the second physiological signal of interest.
2 . The method according to claim 1 , wherein the step of determining a first time domain transformation comprises the steps of identifying a first time difference between the timing of a first heart rate event in the first heart rate sequence signal and the timing of a corresponding first heart rate event in the second heart rate sequence signal, and forming the model based on the first time difference.
3 . The method according to claim 2 , wherein the step of determining a first time domain transformation further comprises identifying a second time difference between the timing of a second heart rate event in the first heart rate sequence signal and the timing of a corresponding second heart rate event in the second heart rate sequence signal, and the step of forming the model comprises a step of forming a linear model based on the first and second time differences.
4 . The method to claim 3 , wherein the linear model is described as A+Bt 2 , wherein A represents an initial time offset between the first and second heart rate sequence signals, B represents a clock drift of the clock of the second sensor based on the first and second time differences, and t 2 is the time instant of the time domain of the second output signal.
5 . The method according to claim 1 , wherein the first heart rate sequence signal is based on a portion of the first output signal and the second heart rate sequence signal is based on a portion of the second output signal.
6 . The method according to claim 1 , wherein the method further comprises a step of resampling the second output signal based on the first time domain transformation.
7 . The method according to claim 1 , wherein at least one of the first and second sensors providing the first and second output signals is a body-worn sensor.
8 . The method according to claim 7 , wherein the body-worn sensor is configured to be worn on the chest, the head, and/or a leg of a subject.
9 . The method according to claim 1 , wherein at least one of the sensors provides an output signal is an ECG-measuring sensor.
10 . The method according to claim 1 , wherein at least one of the sensors provides an output signal is a PPG-measuring sensor.
11 . The method according to claim 1 , further comprising the steps of;
providing a third and fourth output signal each acquired using a respective third and fourth sensor measuring a heart rate derivable physiological signal of a subject; and
determining a second and a third time domain transformation for the time domains of the third and fourth output signals to the time domain of the first output signal using respective models fitted to the timings of heartbeats in the respective first and third heart rate sequence signals and first and fourth heart rate sequence signals, respectively.
12 . The method according to claim 11 , wherein two of the first, second, third, and fourth sensors are ECG-measuring sensors.
13 . The method according to claim 11 , wherein two of the first, second, third, and fourth sensors are PPG-measuring sensors.
14 . The method according to claim 1 , wherein the method is applied to a measurement of a subject during a sleep cycle.
15 . A system for synchronizing sensor output signals comprising:
at least a first and a second sensor, each configured to measure a heartbeat derivable physiological signal of a subject along with a physiological signal of interest and to acquire a first and a second output signal, respectively, wherein each sensor comprises an independent clock for associating the signals measured by the first and second sensors with a respective time domain; and
a processing unit configured to:
receive the first and second output signals;
determine a first heart rate sequence signal from the first output signal;
determine a second heart rate sequence signal from the second output signal;
identify timings of at least one heart rate event in the respective first and second heart rate sequence signals; and
determine a first time domain transformation for the time domain of the second output signal to the time domain of the first output signal using a model fitted to the timings of the at least one heart rate event in the respective first and second heart rate sequence signals;
wherein
the first sensor is configured to capture a first physiological signal of interest that is different from the first output signal,
the second sensor is configured to capture a second physiological signal of interest that is different from the first output signal, and
the processing unit is configured to synchronize the second physiological signal of interest to the first physiological signal of interest by applying model parameters of the model to the second physiological signal of interest.