Method for determining respiratory timing parameters from respiratory monitoring measurements of a subject

Abstract

The present disclosure relates to a method (MO) for providing respiratory timing parameters from respiratory monitoring measurements of a subject, the method comprising the steps of: providing (M1) a respiratory effort signal (E) and a respiratory flow signal (F) of a subject; determining (M2a) an ensemble of peaks (pE) in the respiratory effort signal (E) and determining (M2b) an ensemble of valleys (vE) in the respiratory effort signal (E); and identifying (M3) times associated with the valleys (vE) as preliminary inspiratory onset times (tio); refining (M4) preliminary inspiratory onset times (tio) within respective peak-to-peak time intervals (Tpp) of the respiratory effort signal (E) by: determining (M41) first derivative (F1d) of the respiratory flow signal (F) in a respective peak-to-peak interval (Tpp); determining (M42) local peaks (pF1d) and valleys (vF1d) in the respiratory flow signal first derivative (F1d); determining (M43) a time midpoint (tmid) between the time of a local valley (vF1d) in the respiratory flow signal first derivative (F1d) closest in time to the later endpoint in the respective peak-to-peak time interval (Tpp) and the time of a local peak (pF1d) in the respiratory flow signal derivative (F1d) in the respective peak-to-peak time interval (Tpp); evaluating (M44) whether the respiratory effort signal (E) at the determined time midpoint (tmid) satisfies a predetermined inspiratory onset time condition, and if satisfied, selecting (M45) the determined time midpoint (tmid) as an inspiratory onset time (tio) instead of the preliminary inspiratory onset time (tio) for that peak-to-peak interval (Tpp), whereas if the condition is not satisfied, keep the preliminary inspiratory onset time as the inspiratory onset time for that peak-to-peak interval.

Claims

1. A computer-implemented method for generating respiratory timing parameters from respiratory monitoring measurements of a subject, the method performed by one or more processors and comprising the steps of: receiving a respiratory effort signal of a subject from at least one respiratory effort measurement sensor and a respiratory flow signal of the subject from at least one respiratory flow measurement sensor, processing the respiratory effort signal to identify an ensemble of peaks and an ensemble of valleys in the respiratory effort signal; identifying times associated with the valleys in the respiratory effort signal as preliminary inspiratory onset times; segmenting the respiratory effort signal into peak-to-peak time intervals based on the identified peaks and valleys; and refining the preliminary inspiratory onset times within respective peak-to-peak time intervals of the respiratory effort signal by: calculating a first derivative of the respiratory flow signal in a respective peak-to-peak interval, identifying local peaks and valleys in the respiratory flow signal first derivative, calculating a time midpoint between the time of a local valley in the respiratory flow signal first derivative closest in time to the later endpoint in the respective peak-to-peak time interval and the time of a local peak in the respiratory flow signal first derivative in the respective peak-to-peak time interval, and determining whether the respiratory effort signal at the calculated time midpoint satisfies a predetermined inspiratory onset time condition, and if satisfied, selecting the calculated time midpoint as an inspiratory onset time instead of the preliminary inspiratory onset time for that peak-to-peak interval, whereas if the condition is not satisfied, keeping the preliminary inspiratory onset time as the inspiratory onset time for that peak-to-peak interval.

2. The method according to claim 1, wherein the respiratory effort signal and the respiratory flow signal are obtained via non-invasive respiratory monitoring measurements.

3. The method according to claim 1, wherein the respiratory effort signal and the respiratory flow signal are obtained via bio-impedance signal measurements.

4. The method according to claim 1, wherein the predetermined inspiratory onset time condition is whether an amplitude range of the respiratory effort signal from the calculated time midpoint to the later endpoint in the respective peak-to-peak time interval is at least 75% of a valley-to-peak amplitude range of the respiratory effort signal in the respective peak-to-peak time interval.

5. The method according to claim 1, wherein the step of calculating the first derivative of the respiratory flow signal includes using a Savitsky-Golay derivative kernel incorporating a 2.sup.nd degree polynomial fit.

6. The method according to claim 5, wherein the Savitsky-Golay derivative kernel is characterized by a frame length equal to or less than 150 ms.

7. The method according to claim 1, wherein the step of identifying the ensemble of peaks in the respiratory effort signal includes selecting peaks determined to have a prominence greater than an average prominence determined over a sub segment of the respiratory effort signal.

8. The method according to claim 1, wherein the step of identifying the ensemble of valleys in the respiratory effort signal includes inverting a sub segment of the respiratory effort signal and selecting peaks determined to have a prominence greater than an average prominence determined over said inverted sub segment.

9. The method according to claim 7, wherein a length of the sub segment is between 10 and 15 seconds.

10. The method according to claim 1, further comprising the step of: determining if any identified valleys are missing between any two consecutive peaks in the ensemble of peaks, and if so, inverting the two-consecutive-peak segment missing a determined valley, and determining a location of an undetermined valley by peak detection using half of the two-consecutive-peak segment's dynamic range as a minimum prominence threshold, and if unable to determine a valley still, determining the location of the undetermined valley as the midpoint between the two consecutive peaks.

11. The method according to claim 1, further comprising the step of: determining if any determined peaks are missing between any two consecutive valleys in the ensemble of valleys, and if so, determining a location of an undetermined peak in the two-consecutive-valley segment missing a determined peak by peak detection using half of the two-consecutive-valley segment's dynamic range as a minimum prominence threshold, and if unable to determine a peak still, determining the location of the undetermined peak as the midpoint between the two consecutive valleys.

12. The method according to claim 1, wherein the step of identifying peaks and valleys in the respiratory flow signal first derivative includes using more than a predetermined ratio of a local dynamic range as minimum peak prominence.

13. The method according to claim 1, wherein the step of calculating the time midpoint includes using peaks in the respiratory flow signal first derivative which are the closest, but which are no closer than a predetermined time offset, to the latest later endpoint of the respective peak-to-peak time interval.

14. The method according to claim 1, further comprising the step of: identifying times associated with the peaks of the ensemble of peaks of the respiratory effort signal as expiratory onset times.

15. The method according to claim 6, wherein the Savitsky-Golay derivative kernel is characterized by a frame length less than 120 ms.

16. The method according to claim 12, wherein the predetermined ratio is equal to or more than 5-10%.

17. The method according to claim 1, wherein the step of calculating the time midpoint includes using peaks in the respiratory flow signal first derivative which are the closest, but no closer than a predetermined time offset, to the later endpoint of the respective peak-to-peak time interval, and wherein the predetermined time offset is less than 200 milliseconds.

18. The method according to claim 1, wherein the step of calculating the time midpoint includes using peaks in the respiratory flow signal first derivative which are the closest, but no closer than a predetermined time offset, to the later endpoint of the respective peak-to-peak time interval, and wherein the predetermined time offset is less than 160 milliseconds.

19. The method according to claim 1, further comprising the steps of: identifying times associated with the peaks of the ensemble of peaks of the respiratory effort signal as expiratory onset times, and determining whether the inspiratory onset times and the expiratory onset times are equal in number.

20. A computer program comprising instructions which, when executed by a computing device, cause the computing device to carry out a method for generating respiratory timing parameters from respiratory monitoring measurements of a subject, the method comprising the steps of: receiving a respiratory effort signal of a subject from at least one respiratory effort measurement sensor and a respiratory flow signal of the subject from at least one respiratory flow measurement sensor; processing the respiratory effort signal to identify an ensemble of peaks and an ensemble of valleys in the respiratory effort signal; identifying times associated with the valleys in the respiratory effort signal as preliminary inspiratory onset times; segmenting the respiratory effort signal into peak-to-peak time intervals based on the identified peaks and valleys; and refining the preliminary inspiratory onset times within respective peak-to-peak time intervals of the respiratory effort signal by: calculating a first derivative of the respiratory flow signal in a respective peak-to-peak interval, identifying local peaks and valleys in the respiratory flow signal first derivative, calculating a time midpoint between the time of a local valley in the respiratory flow signal first derivative closest in time to the later endpoint in the respective peak-to-peak time interval and the time of a local peak in the respiratory flow signal first derivative in the respective peak-to-peak time interval, and determining whether the respiratory effort signal at the calculated time midpoint satisfies a predetermined inspiratory onset time condition, and if satisfied, selecting the calculated time midpoint as an inspiratory onset time instead of the preliminary inspiratory onset time for that peak-to-peak interval, whereas if the condition is not satisfied, keeping the preliminary inspiratory onset time as the inspiratory onset time for that peak-to-peak interval.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) The invention will in the following be described in more detail with reference to the enclosed drawings, wherein:

(2) FIG. 1 illustrates a schematic of a method according to one embodiment of the invention; wherein dashed boxes indicates optional embodiments;

(3) FIG. 2 illustrates a sub-segment of a respiratory effort signal be used as input to the method according to one embodiment of the invention;

(4) FIG. 3 illustrates a sub-segment of a respiratory flow signal be used as input to the method according to one embodiment of the invention;

(5) FIG. 4 illustrates a sub-segment of a respiratory effort signal wherein peaks and valleys have been determined according to one embodiment of the invention;

(6) FIG. 5 illustrates a sub-segment of a respiratory flow signal derivative according to one embodiment of the invention;

(7) FIG. 6 illustrates a respiratory effort signal and a respiratory flow signal derivative according to one embodiment of the invention;

(8) FIG. 7 illustrates a selection of determined respiratory parameters according to one embodiment of the invention;

(9) FIG. 8 illustrates prominence determined over the same sub-segment of the respiratory effort signal shown in previous figures, the prominence as determined according to one embodiment of the invention;

(10) FIG. 9 illustrates expiration and inspiration periods determined according to one embodiment of the invention;

(11) FIG. 10 illustrates a schematic of a device according to one embodiment of the invention;

(12) FIG. 11 illustrates a device according to one embodiment of the invention while arranged on a subject.

DESCRIPTION OF EMBODIMENTS

(13) The present invention will be described more fully hereinafter with reference to the accompanying drawings, in which preferred embodiments of the invention are shown. This invention may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art. In the drawings, like numbers refer to like elements.

(14) FIG. 1 illustrates a schematic of a method according to one embodiment of the invention; wherein dashed boxes indicates optional embodiments. The method M0 comprises the step of providing M1 a respiratory effort signal E and a respiratory flow signal F of a subject. FIG. 2 and FIG. 3 respectively show examples of such signals. The method M0 further comprises a step of determining M2a an ensemble of peaks IDE in the respiratory effort signal E and a step of determining M2b an ensemble of valleys vE in the respiratory signal E. This can be seen in FIG. 4 which shows a respiratory effort signal E with peaks IDE and valleys v.sub.E indicated. The method also comprises a step of identifying M3 times associated with the valleys v.sub.E as preliminary inspiratory onset times t.sub.io. The preliminary inspiratory onset times t.sub.io within respective peak-to-peak intervals T.sub.pp of the respiratory effort signal E is then refined in a refinement step M4. This will be explained in the following.

(15) The refinement step M4 comprises a step M41 of determining F.sub.1d of the respiratory flow signal F in a respective peak-to-peak interval T.sub.pp and a step of determining M42 local peaks p.sub.F1d and valleys v.sub.F1d in the respiratory flow signal first derivative F.sub.1d. The respiratory flow signal derivative F.sub.1d is shown in FIG. 5 where peaks p.sub.F1d and valleys v.sub.F1d of the respiratory flow signal first derivative F.sub.1d are indicated. The method further comprises a step of determining M43 a time midpoint t.sub.mid between the time of a local valley v.sub.F1d in the respiratory flow signal first derivative F.sub.1d closest in time to the later endpoint in the respective peak-to-peak time interval T.sub.pp and a peak pdf in the respiratory flow signal first derivative F.sub.1d in the respective peak-to-peak interval T.sub.PP. This is shown in FIG. 5 wherein the time midpoint t.sub.mid of a midpoint m.sub.F1d between a valley v.sub.F1d and a peak p.sub.F1d is indicated.

(16) The method M0 further comprises a step of evaluating M44 whether the respiratory effort signal E at the determined time midpoint t.sub.mid satisfies a predetermined inspiratory onset time condition, and if satisfied, selecting M45 the determined time midpoint t.sub.mid as an inspiratory onset time t.sub.io instead of the preliminary inspiratory onset time t.sub.io for that peak-to-peak interval T.sub.PP. This is indicated in FIG. 6 wherein both timepoints t.sub.mid and t.sub.io are indicated. Depending on whether the respiratory effort signal E at the determined time midpoint t.sub.mid satisfies the predetermined inspiratory onset, either the time midpoint or the preliminary inspiratory onset time will be selected as the inspiratory onset time t.sub.io.

(17) This is done for a plurality of such peak-to-peak time intervals, thereby resulting in an ensemble of inspiratory onset times t.sub.io some of which may have been refined. Such ensemble of inspiratory onset times t.sub.io may then by used in further respiration analysis.

(18) As is explained in the summary of this disclosure, the respiratory effort signal E and the respiratory flow signal F may be provided via a number of devices, each of which are characterized by advantages which may be beneficial to the disclosed method. In a preferred embodiment, the respiratory effort signal E and the respiratory flow signal F are obtained via non-invasive respiratory monitoring measurements. In particular, the respiratory effort signal E and the respiratory flow signal F are obtained via bio-impedance signal measurements.

(19) The predetermined inspiratory onset time condition may be selected differently depending on the situation but according to a preferred embodiment, the predetermined inspiratory onset time condition is whether the amplitude range A.sub.i of the inspiratory effort signal E at the inspiratory onset time t.sub.io to the subsequent expiratory onset time t.sub.eo is at least 75% of the amplitude range A.sub.vp of the respiratory effort signal E as measured from the preliminary inspiratory onset time t.sub.io, i.e. from a valley, to the following expiratory onset time t.sub.eo, i.e. a peak. This is shown in the upper graph of FIG. 6.

(20) Further, the step of determining M41 first derivative of the respiratory flow signal F preferably includes using a Savitsky-Golay derivative kernel incorporating a 2.sup.nd degree polynomial fit. The Savtizky-Golay derivative kernel may for example be characterized by a frame length equal to or less than 150 ms, preferably less than 120 ms. The figures represent data wherein a Savitsky-Golay derivative kernel incorporating a 2.sup.nd degree polynomial fit was used, and also wherein the frame length was set to 110 ms.

(21) According to one embodiment, the step of determining M2a an ensemble of peaks in the respiratory effort signal E includes selecting M2a-1 peaks determined to have a prominence greater than an average prominence determined over a sub-segment of the respiratory effort signal E. The prominence for a sub-segment is shown in FIG. 8 and the average prominence is then determined over this subsegment and only peaks who has a prominence greater than the average prominence is considered. The average prominence may be determined as mentioned in the summary. The valleys v.sub.E of the respiratory effort signal E may be determined in a similar manner; wherein the step of determining M2b an ensemble of valleys in the respiratory effort signal E includes inverting M2b-1 a sub-segment of the respiratory effort signal E and selecting M2b-2 peaks determined to have a prominence greater than an average prominence determined over said inverted sub-segment.

(22) In a preferred embodiment, the method comprises the step of determining M2a-2 if there is missing any determined valleys between any two consecutive peaks in the ensemble of peaks, and if so, inverting M2a-3 the two-consecutive-peak segment missing a determined valley and determining M2a-4 the location of the undetermined valley by peak detection using half of the two-consecutive-peak segment's dynamic range as a minimum prominence threshold, and if unable to determine a valley still, determining M2a-5 the location of the undetermined valley as the midpoint between the two consecutive peaks. This increases the likelihood that there is an equal set of inspiratory onset times and expiratory onset times obtained, and that each onset times are determined in a correct manner.

(23) Likewise, the method may comprise the step of determining M2b-3 if there is missing any determined peaks between any two consecutive valleys in the ensemble of valleys, and if so, determining M2b-4 the location of the undetermined peak in the two-consecutive-valley segment's dynamic range as a minimum prominence threshold, and if unable to determine a peak still, determining M2b-5 the location of the undetermined peak as the midpoint between the two consecutive valleys.

(24) According to one embodiment, the step of determining M42 local peaks p.sub.F1d and valleys v.sub.F1d in the respiratory flow signal first derivative F.sub.1d includes using more than a predetermined ratio of a local dynamic range as minimum peak prominence. The predetermined ratio may be equal to or more than 5-10%. To obtain the data for represented in some figures, a predetermined ratio of 5% or 1/20 was used.

(25) In a preferred embodiment, as indicated in FIG. 1, the step of determining M43 a time midpoint t.sub.mid includes using M43-1 peaks in the respiratory flow signal first derivative F.sub.1d which are closest, but no closer than a predetermined time offset T.sub.offset, to the latest endpoint of the respective peak-to-peak time interval T.sub.PP. The predetermined time offset T.sub.offset may be less than 200 milliseconds, preferably less than 160 milliseconds. This is shown in FIG. 5 wherein the T.sub.offset is indicated. To obtain the data for represented in some figures, a time offset T.sub.offset=156 milliseconds was used.

(26) The method M0 may also comprise the step of identifying M5 times associated with the peaks of the ensemble of peaks p.sub.E of the respiratory effort signal E as expiratory onset times t.sub.eo and determining M6 whether the inspiratory onset times t.sub.io and the expiratory onset times t.sub.eo are qual in number. Expiratory onset times t.sub.eo are indicated in FIG. 2. Both expiratory onset times t.sub.eo and inspiratory onset times t.sub.io are indicated in upper graph of FIG. 6.

(27) The expiratory onset times t.sub.eo and the inspiratory onset times t.sub.io may be used to determine a number of respiratory timing parameters. For example, the method M0 may comprise the step of determining M6 a first respiration parameter being an inspiratory time interval by taking expiratory onset time and subtracting from it the inspiratory onset time. Also, the method may comprise the step of determining M7 a second respiration parameter being an expiratory time interval by taking inspiratory onset time and subtracting from it the expiratory onset time. Using these, further respiratory timing parameters may be determined, thus, the method according to one further embodiment may comprise the step of determining M8 a respiratory rate by taking the reciprocal of the sum of the inspiratory onset time interval and the expiratory time interval, or alternatively, a respiratory rate by taking the reciprocal of the time interval between two adjacent peaks (expiratory onset times) of the respiratory effort signal, or alternatively, a respiratory rate by taking the reciprocal of the time interval between two adjacent valleys (inspiratory onset times) of the respiratory effort signal, or alternatively, a respiratory rate obtained by combining the time-course of reciprocal of the time-interval between two adjacent peaks and the time-course of the reciprocal of the time-interval of two adjacent valleys. Also, the method M0 may comprises the step of determining M9 an inspiratory duty cycle by multiplying the inspiratory time interval with the respiratory rate. The respiratory rate, the inspiratory time interval Ti and the expiratory time interval Te, and the duty cycle are each shown in FIG. 7.

(28) With such respiratory timing parameters, it may thus be possible to indicate expiration and inspiration phases in the respiratory effort signal E and the respiratory flow signal F as done in FIG. 9. Here, inspiration is indicated by grey backgrounds.

(29) The method M0 may be provided as instructions in a computer program which, when executed by a computing device, cause the computing device to carry out the method according to first aspect or any embodiments thereof. The computing device may be a portable computing device such as a smartphone, a smartwatch, a tablet, or a laptop. The computing device may alternatively be a workstation or a server. In case of a server, the program code may be controlled from an interface running on a remote computing device. The program code may be executed by means of cloud computing.

(30) FIG. 10 shows a device for providing respiratory parameters from respiration monitoring measurements. As shown in FIG. 1, the device 100 comprises a current signal injection module 112. The current signal injection module 112 may be configured to generate and output the current signal S1, which is to be applied to the subject. The current signal injection module 112 may comprise a current source for generating a current signal S1. The current signal injection module 112 may be configured to output an AC current signal. The device 100 further comprises a bioimpedance measurement sensor 110. The bioimpedance measurement sensor 110 may be configured to receive voltage input signals representing a voltage generated by the current signal S1 applied to the subject. The bioimpedance measurement sensor 110 may be configured to extract a measured bioimpedance signal S2 from the received voltage input signals. The bioimpedance measurement sensor 110 may be configured to process the received voltage input signals, e.g. by filtering the input signals, in order to extract relevant information.

(31) The bioimpedance measurement sensor 110 may comprise two or more electrodes 114, which may be arranged to be in contact with skin of the subject. The electrodes 114 may be connected to the current signal injection module 112 to receive the current signal S1 and provide the current signal through tissue of the subject. The electrodes 114 may also be connected to the bioimpedance measurement sensor 110 for providing voltage input signals that may be used for measuring the bioimpedance signal S2. The electrodes 114 may be arranged in a bipolar arrangement, wherein the same electrodes 114 are used for providing the current signal S1 to the subject and for acquiring the voltage input signals. However, the electrodes 114 may alternatively be arranged in a tetrapolar arrangement, wherein two electrodes are used for providing the current signal S1 to the subject and two other electrodes are used for acquiring the voltage input signals.

(32) More than two (or four) electrodes 114 may be provided, which may allow selection of which electrodes 114 to be used in a measurement, so that electrodes 114 providing highest quality bioimpedance signal S2 may be selected. The selection of which electrodes 114 to be used may be performed in set-up of the device 100 or may be dynamically changed during signal acquisition e.g. when conditions for acquiring the bioimpedance signal change. The bioimpedance measurement sensor 110 with electrodes 114 may be configured to be attached on a thorax region of the subject. The bioimpedance measurement sensor 110 may be arranged on a patch device 116 configured for being arranged on a thorax region of the subject, wherein the electrodes 114 may be mounted to be exposed on the patch device 116, such that the electrodes 114 may be arranged in contact with the skin of the subject. The patch device 116 may for instance comprise an adhesive patch, a textile/garment being worn by the subject, or a belt, which may be configured to be attached around the torso of the subject.

(33) When a bioimpedance measurement is performed based on electrodes 114 arranged on the thorax of a subject, chest expansion may cause a change in a current path between the electrodes 114, such that the bioimpedance is changed in relation to a respiratory effort. Also, air has a different impedance than tissue. As an amount of air present in the lungs varies during a respiratory cycle, the bioimpedance is also changed in relation to respiratory airflow. Thus, the bioimpedance measurement sensor 110 may be configured for acquisition of a bioimpedance signal S2 which holds information of both respiratory effort and respiratory airflow. The processing unit 120 may be configured to receive the bioimpedance signal S2 from the bioimpedance measurement sensor 110. The processing unit may be further configured to process the bio-impedance signal S2 to extract therefrom a respiratory effort signal E and a respiratory flow signal F. These may then be used by a processing unit configured to execute the method herein disclosed. The processing unit may be the same processing unit extracting the respiratory effort signal E and the respiratory flow signal F.

(34) In an optional embodiment, the processing unit 120 may further be configured to receive a reference signal S3 from a reference measurement sensor 130. The reference signal S3 may be acquired so as to isolate respiratory effort from respiratory airflow, e.g. by using a sensor which is placed or configured for acquiring a signal which is only affected by either respiratory effort or respiratory airflow. Hence, the reference signal S3 may represent respiratory effort or respiratory airflow.

(35) The processing unit 120 may be implemented in hardware, or as any combination of software and hardware. The processing unit 120 may, for instance, be implemented as software being executed on a general-purpose computer. The device 100 may thus comprise one or more physical processors, such as a central processing unit (CPU), which may execute the instructions of one or more computer programs in order to implement functionality of the processing unit 120. Thus, the device 120 may comprise a single processing unit, which may provide a plurality of functionalities e.g. as separate threads within the processing unit 120. The processing unit 120 may alternatively be implemented as firmware arranged e.g. in an embedded device, or as a specifically designed processing unit, such as an Application-Specific Integrated Circuit (ASIC) or a Field-Programmable Gate Array (FPGA). The reference measurement sensor 130 may be part of and may be delivered with the device 100. The device 100 may thus be set-up for communication between the reference measurement sensor 130 and the processing unit 120. However, the reference measurement sensor 130 may alternatively be separately delivered, e.g. by a different vendor than the vendor providing the device 100. A user may thus connect the reference measurement sensor 130 to the processing unit 120, e.g. by attaching a wire between the reference measurement sensor 130 and a port in a housing in which the processing unit 120 is arranged, whereby the processing unit 120 and the reference measurement sensor 130 may then exchange set-up messages for automatically setting up communication between each other. Alternatively, a user may initiate a discovery procedure for allowing a wireless communication between the reference measurement sensor 130 and the processing unit 120 to be established and again for automatically setting up communication between the reference measurement sensor 130 and the processing unit 120. In a further alternative, the reference measurement sensor 130 and the bioimpedance measurement sensor 110 may be configured to separately communicate the reference signal S3 and the bioimpedance signal S2 to a remotely arranged processing unit 120, e.g. a processing unit 120 arranged in the cloud. The signals may be communicated after an entire period of gathering the signals, such as signals acquired during a night's sleep of the subject. The processing unit 120 may then synchronize the signals before processing.

(36) A reference measurement sensor 130 configured to acquire a reference signal representing a respiratory effort may be any sensor which may be configured to acquire a representation of the respiratory effort. For instance, the reference measurement sensor 130 may include an esophageal manometer, a respiratory inductance plethysmography (RIP) belt, a thoracoabdominal polyvinylene fluoride (PVDF) belt, an accelerometer, or an electromyograph (EMG) sensor.

(37) A reference measurement sensor 130 configured to acquire a reference signal representing a respiratory airflow may be any sensor which may be configured to acquire a representation of the respiratory airflow. For instance, reference measurement sensor may include an oro-nasal thermal sensor, such as a thermistor, a polyvinylene fluoride sensor, or a thermocouple, a nasal pressure transducer, a pneumotachograph sensor, or a spirometer. The processing unit 120 may be configured to receive reference signals S3 from a plurality of reference measurement sensors 130. The plurality of reference measurement sensors 130 may comprise only sensors configured to acquire a reference signal S3 representing respiratory effort, only sensors configured to acquire a reference signal S3 representing respiratory airflow, or one or more sensors configured to acquire a reference signal S3 representing respiratory effort combined with one or more sensors configured to acquire a reference signal S3 representing respiratory airflow. To illustrate these options, reference measurement sensors 130 are indicated by dashed lines in FIG. 1.

(38) The device 100 may comprise one or more housings, in which the bioimpedance measurement sensor 110, the processing unit 120 and the reference measurement sensor 130 may be arranged. The housings may be connected by wires for allowing communication between the sensors and the processing unit 120. Alternatively, one or more of the sensors 110, 130 and the processing unit 120 may be set up for wireless communication. The device 100 may thus be delivered to be ready to use, e.g. in a single package with all parts of the device 100 already set up to communicate with each other.

(39) The processing unit 120 may be arranged in a housing on the patch device 116. The reference measurement sensor 130 may also be arranged on the same patch device 116. However, in an alternative embodiment, the processing unit 120 may be arranged in a central housing, which may be separate from the patch device 116. The central housing may further comprise an output port for connection to an external unit, which may receive the respiratory effort signal E and the respiratory flow signal F for further processing of the components. Alternatively, or additionally, the central housing may comprise a communication unit for wireless communication of the respiratory effort signal E and the respiratory flow signal F to the external unit. The central housing may also be connected to a display for enabling the respiratory effort signal E and the respiratory flow signal F to be output on the display. Also, the reference signal may be output on the display S3. This may allow a physician, nurse or any other person, to manually inspect signals representing respiration of the subject, e.g. for manual analysis of the respiration.

(40) In the drawings and specification, there have been disclosed preferred embodiments and examples of the invention and, although specific terms are employed, they are used in a generic and descriptive sense only and not for the purpose of limitation, the scope of the invention being set forth in the following claims.