Sensor system and method for determining a breathing type

Abstract

A sensor system is for determining a breathing type of a subject. It has first and second electromagnetic radiation (e.g. optical) sensors, for example PPG sensors. The sensors are for application to the skin of the subject at different sensor locations. The signals from the first and second sensors are analyzed to determine if there is predominantly a first breathing type (e.g. chest breathing) or predominantly a second breathing type (e.g. belly breathing), or a mixture of the first and second breathing types.

Claims

1. A sensor system for determining a breathing type of a subject, comprising: a first electromagnetic radiation sensor having a first detector for association with the skin of the subject at a first sensor location; a second electromagnetic radiation sensor having a second detector for association with the skin of the subject at a second sensor location; and a controller which is adapted to: analyze the signals from the first and second detectors, wherein analyzing comprises: determining a first measure from the first detector signal; determining a second measure from the second detector signal; determining a relationship between the first and second measures; wherein the first and second measures each comprise: (i) a measure of respiratory induced intensity variation relative to the amplitude at the pulse frequency; or (ii) a measure of respiratory induced pulse-amplitude variation relative to the amplitude at the pulse frequency; or (iii) a measure of respiratory induced intensity or pulse-amplitude variation relative to the DC-level at that detector; and determine from the analysis if there is predominantly a first breathing type or predominantly a second breathing type, or a mixture of the first and second breathing types.

2. A system as claimed in claim 1, wherein the first and second electromagnetic radiation sensors operate in the wavelength interval between 400 nm and 1000 nm.

3. A system as claimed in claim 1, wherein the first and second electromagnetic radiation sensors each comprise a PPG sensor.

4. A system as claimed in claim 1, wherein the first breathing type comprises chest breathing and the second breathing type comprises belly breathing.

5. A method for determining a breathing type of a subject, comprising: associating a first electromagnetic radiation sensor having a first detector with the skin of the subject at a first sensor location; associating a second electromagnetic radiation sensor having a second detector with the skin of the subject at a second sensor location; analyzing the signals from the first and second detectors, wherein analyzing comprises: determining a first measure from the first detector signal; determining a second measure from the second detector signal; and determining a relationship between the first and second measures, wherein the first and second measures each comprise: (i) a measure of respiratory induced intensity variation relative to the amplitude at the pulse frequency; or (ii) a measure of respiratory induced pulse-amplitude variation relative to the amplitude at the pulse frequency; or (iii) a measure of respiratory induced intensity or pulse-amplitude variation relative to the DC-level at that detector; and determining from the analysis if there is predominantly a first breathing type or predominantly a second breathing type, or a mixture of the first and second breathing types.

6. A method as claimed in claim 5, comprising providing outputs from the first and second sources of electromagnetic radiation in the wavelength interval between 400 nm and 1000 nm.

7. A method as claimed in claim 5, wherein the first breathing type comprises chest breathing and the second breathing type comprises belly breathing.

8. A method as claimed in claim 7, comprising providing a first sensor location associated with the Superior Vena Cava and providing a second sensor location associated with the Inferior Vena Cava.

9. A method as claimed in claimed in claim 8, wherein the first sensor location is the wrist and the second sensor location is the ankle.

10. A computer program product comprising a non-transitory computer readable medium, wherein the computer program product comprises computer readable code which is adapted, when run on a computer, to receive signals from a first electromagnetic radiation sensor associated with the skin of the subject at a first location and a second electromagnetic radiation sensor associated with the skin of the subject at a second location and to perform the steps of analyzing and determining of the method of claim 5.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) Embodiments of the invention will be described, by way of example only, with reference to the drawings, in which:

(2) FIG. 1 shows a known PPG sensor;

(3) FIG. 2 shows a first example of system in accordance with the invention for determining different breathing types;

(4) FIG. 3 shows a PPG signal, a PPG amplitude waveform and the RIIV waveform; and

(5) FIG. 4 shows a method for determining different breathing types.

DETAILED DESCRIPTION OF THE EMBODIMENTS

(6) The invention provides a sensor system for determining a breathing type of a subject. It has first and second electromagnetic radiation (e.g. optical) sensors, for example PPG sensors. The sensors are for application to the skin of the subject at different sensor locations. The signals from the first and second sensors are analyzed to determine if there is predominantly a first breathing type (e.g. chest breathing) or predominantly a second breathing type (e.g. belly breathing), or a mixture of the first and second breathing types.

(7) In one implementation, the system makes use of two PPG sensors.

(8) FIG. 1 shows a known PPG sensor.

(9) It comprises a light source 10 and an optical detector 12. The light source may comprise a single multi-wavelength light source, or multiple light units with different output wavelengths. The sensor has a processor 14 and it generates an output 16 in the form of a PPG signal which varies over time. The output may be provided in real time and/or it may be logged and stored as a trace over time during a monitoring period. The stored trace may then be analyzed after the monitoring period is over.

(10) The PPG sensor for example comprises a pulse oximeter. The output 16 may be interpreted to provide a pulse rate, a respiration rate and an arterial oxygen saturation level, in known manner.

(11) In one example, the invention is based on the combination of two PPG sensors which may be of the general type shown in FIG. 1.

(12) FIG. 2 shows a first example of system in accordance with the invention being used to monitor a subject 18 in a hospital bed 19.

(13) The system has a first sensor 20 of the same type as shown in FIG. 1, thus having a first source of electromagnetic radiation and a first detector for application to the skin of the subject at a first sensor location. This first location is the wrist in the example shown.

(14) A second sensor 22 is also of the same type as shown in FIG. 1, having a second source of electromagnetic radiation and a second detector for application to the skin of the subject at a second sensor location. The second location is the ankle in the example shown.

(15) A controller 24 analyzes the signals from the first and second sensors and determines from the analysis if there is predominantly a first breathing type (chest breathing) or predominantly a second breathing type (belly breathing), or a mixture of the first and second breathing types.

(16) Some body sites show clearer breathing signals relating to chest breathing whereas other body sites show stronger breathing signals relating to belly breathing. In particular, sensor signals measured from skin in lower body parts, particularly the lower limbs are affected more by the pressure variations in the inferior vena cava (IVC), whereas the upper body parts in particular the upper limbs show sensor signals that are affected more by the pressure variations in the superior vena cava (SVC).

(17) By analyzing the two sensor signals, for example the relative signal strengths of the two sensor signals after each has already been processed to represent the respiration signal, the contributions of the two different breathing types may be determined.

(18) In particular, the analysis may comprise measuring the strength of the respiratory signal relative to the strength of the pulse signal at the two body locations, and comparing these two strength signals.

(19) Chest breathing is typically encountered during stressful situations and belly breathing is generally encountered during a relaxed state of the subject. The controller may in its simplest form provide an output 26 which is a binary indicator representing either chest or belly breathing. In a more detailed implementation, the degrees of belly breathing versus chest breathing may be indicated (e.g. a percentage).

(20) The respiration signals received from the two body locations are preferably normalized. This normalization may comprise dividing the variations in the received signals at the respiratory rate by the amplitude of the variations at the pulse-rate. Other normalization approaches are possible, however, like dividing by the average value (DC) of the received signal at the detector. One benefit of the normalization is to obtain a fixed gain of the respiratory signal irrespective of the local perfusion of the skin. The normalization process is described in further detail below.

(21) The sources of electromagnetic radiation for example operate in the wavelength interval between 400 nm and 1000 nm, where particularly a wavelength around 550 nm is included as the PPG signal is relatively strongest around this wavelength. More than one wavelength is preferably used to improve motion robustness of the device.

(22) The sensors are in the example shown contact sensors, with the detector part in contact with the skin of the subject, while the skin region may be illuminated by an electromagnetic radiator which only needs to be mounted in close proximity to the skin.

(23) The detectors of the sensors may instead form an image sensor array that may remotely detect the reflected/transmitted light from/through the skin of the subject at the two (or more) locations. Typically a camera remotely images the two different skin regions of the subject using either ambient or dedicated illumination.

(24) The respiration signal is typically obtained as the low frequency component of the received detector signal, below the pulse frequency (e.g. 1 Hz). Alternatively, the respiration signal may be obtained the detector signals by extracting the relative strength of the amplitude modulation of the pulse signal at the respiratory rate.

(25) It is for example disclosed in W. Karlen, A. Garde, D. Myers, C. Scheffer, J. Ansermino, and G. Dumont, Estimation of respiratory rate from photoplethysmographic imaging videos compared to pulse oximetry, Biomedical and Health Informatics, IEEE Journal of, 19, 1331-1338 (2015) that the respiratory signal and the breathing rate (or breathing frequency) can be obtained from photoplethysmography (PPG) via three parameters: inter-beat intervals: inhalation increases heart rate, while after exhalation the frequency quickly drops, i.e. the pulse is FM-modulated with the respiratory signal. the DC part of the PPG signal: the DC component increases at inhalation and decreases at exhalation. This effect may be defined as baseline modulation, or respiratory induced intensity variation, RIIV. the AC part of the PPG signal: the AC component increases at inhalation and decreases at exhalation, i.e. the pulse-signal is amplitude modulated with the respiratory signal. This is often referred to as respiratory induced amplitude variation, RIAV.

(26) The RIIV waveform has been found to be of particular interest for this application.

(27) FIG. 3 shows a PPG signal, a PPG amplitude waveform and the RIIV waveform, which is a low pass filtered version of the PPG amplitude waveform. The PPG amplitude waveform is obtained by tracking the peaks of the PPG waveform and creating a linearly interpolated waveform.

(28) The strength of the baseline variation as expressed by the RIIV signal has been found to depend strongly on the body location.

(29) The PPG amplitude (AC amplitude) as well as DC values (area under the curve of the AC signal) varies depending on body site, which is why the normalization mentioned above is desired to compensate for this. The respiratory signal is for example normalized by the amplitude of the pulse signal. In this way, the baseline variation is expressed in relation to the magnitude of the pulse signal as seen in the PPG signal at that particular body location. The purpose is for the baseline variations as measured at each location to have equal significance in that a similar baseline variation magnitude results from a respiration signal. Other normalization approaches to achieve this aim may of course be considered. For example, an alternative normalization may use the DC-value of the reflected light. A signal may in effect be obtained relative to a DC level by taking the logarithm of the sensor signals. In this case, the resulting AC amplitudes are essentially the same as the amplitudes relative to the DC (normalized amplitudes).

(30) When combining the two signals, a weighting may be applied, if needed. Thus one option is simply to measure a ratio of the normalized RIIV signals at the two locations. An alternative is to perform a scaling of one of the normalized signals first. The difference between the signals (or different between scaled/weighted signals) rather than a ratio may instead be used. The most appropriate signal combination of the two normalized RIIV signals may be obtained by previous experimentation.

(31) More generally, any combination of signals (not necessarily RIIV signals) from the two body locations may be used as long as the combined signal has a strong correlation with the type of breathing. Thus regression analysis of experimental data may be used to determine a suitable signal combination which correlates with the breathing type. The example below is only one possible example.

(32) The invention has been tested to provide proof-of-concept, and the results are presented below.

(33) To verify the feasibility of the concept, PPG data obtained from a subject was analyzed, with the subject following a particular breathing protocol. The protocol prescribes breathing at a constant rate of 10 breaths/min with three types of breathing exercised sequentially: belly, chest and a combination of both. The combination is referred to as normal.

(34) Two identical PPG sensors were attached to the wrist and ankle locations of the subject and the PPG-signals were synchronously sampled with a sampling frequency of 128 Hz.

(35) For each breathing modality, respiratory features were extracted from the PPG waveform in the form of the DC variation (respiratory induced intensity variation, RIIV).

(36) An absolute RIIV variation was measured, determined by the amplitude at the breathing frequency in the frequency spectrum of the normalized PPG signal.

(37) A relative variation was also measured, based on the absolute RIIV definition above, but scaled with respect to the amplitude at the pulse frequency as explained above.

(38) For these features, the ratio between the signals gathered the wrist and ankle locations were then calculated, i.e.
RIIV_absolute_ratio=RIIV_absolute(wrist)/RIIV_absolute(ankle).
RIIV_relative_ratio=RIIV_relative(wrist)/RIIV_relative(ankle).

(39) These values are presented in the table below. The ratio is expected to be to be higher the more chest breathing dominates.

(40) TABLE-US-00001 Type of Breathing RIIV_absolute_ratio RIIV_relative_ratio Normal 0.70 0.55 Belly 0.35 0.30 Chest 0.61 0.64

(41) These results show that the relative RIIV measure provides a signal for which chest breathing gives the highest ratio, belly breathing the lowest ratio, while normal breathing shows up in between these two. It also shows how the normalization improves the results.

(42) An alternative measure is the AC variation (respiratory induced amplitude variation, RIAV). Again, an absolute value may be determined, for example as the standard deviation of the derivative of the peak-valley pairs. A relative value is similar to the absolute RIAV definition, but again scaled with respect to the cardiac pulse amplitude to correct for local variations in PPG strength.

(43) The invention is of interest for both home healthcare as well as within the clinical environment. It can be used to help guide a user to achieve relaxed breathing, for example in the event of an asthma attack or other breathing difficulties.

(44) The example above is based on the use of two contact sensors. Recently, non-contact, remote PPG (rPPG) devices (also called camera rPPG devices) for unobtrusive measurements have been introduced. Remote PPG utilizes light sources or, in general radiation sources, disposed remotely from the subject of interest. Similarly, also a detector, e.g., a camera or a photo detector, can be disposed remotely from the subject of interest. Therefore, remote photoplethysmographic systems and devices are considered unobtrusive and well suited for medical as well as non-medical everyday applications. This type of sensing may be used by the system of the invention, although remote PPG devices typically achieve a lower signal-to-noise ratio.

(45) In this case the PPG sensors may for example make use of a camera flash as the optical source, and a camera image sensor as the sensor device. In this way, with a suitable application running on a mobile phone, the system can be implemented.

(46) There may be more than two sensors, to provide additional information to be processed in order to determine the breathing type with greater accuracy.

(47) FIG. 4 shows a method for determining a breathing type of a subject.

(48) In step 30 a first sensor having a first source of electromagnetic radiation and a first detector is associated with the skin of the subject at a first sensor location. It may be applied to the skin or it may be arranged for remotely monitoring the body at that location.

(49) In step 32, a second sensor having a second source of electromagnetic radiation and a second detector is associated with the skin of the subject at a second sensor location.

(50) In step 33 the first second signal S1 is obtained. It is converted to a measure of the respiration trace in step 35 for example by low pass filtering or other signal processing to obtain a first measure M1 of the respiration. This measure is normalized in step 37 to provide a relative value R1.

(51) In step 34 the second signal S2 is obtained. It is converted to a measure of the respiration trace in step 36 for example by low pass filtering or other signal processing to obtain a second measure M2 of the respiration. This may be the respiratory induced intensity variation (RIIV). This measure is normalized in step 38 to provide a relative value R2.

(52) The two relative values are processed in step 40, for example to find a ratio, or a weighted ratio, or other linear or non-linear combination of the signals (also may be referred to as a relationship). Form this processing, it is determined in step 42 if there is predominantly a first breathing type or predominantly a second breathing type, or a mixture of the first and second breathing types.

(53) An output is provided in step 44.

(54) The first breathing type comprises chest breathing and the second breathing type comprises belly breathing.

(55) The signal processing may be implemented by a computer program which forms or is part of the controller 24. The controller can be implemented in numerous ways, with software and/or hardware, to perform the various functions required. A processor is one example of a controller which employs one or more microprocessors that may be programmed using software (e.g., microcode) to perform the required functions. A controller may however be implemented with or without employing a processor, and also may be implemented as a combination of dedicated hardware to perform some functions and a processor (e.g., one or more programmed microprocessors and associated circuitry) to perform other functions.

(56) Examples of controller components that may be employed in various embodiments of the present disclosure include, but are not limited to, conventional microprocessors, application specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs).

(57) In various implementations, a processor or controller may be associated with one or more storage media such as volatile and non-volatile computer memory such as RAM, PROM, EPROM, and EEPROM. The storage media may be encoded with one or more programs that, when executed on one or more processors and/or controllers, perform at the required functions. Various storage media may be fixed within a processor or controller or may be transportable, such that the one or more programs stored thereon can be loaded into a processor or controller.

(58) The examples above make use of PPG measurements. The electromagnetic sensors are used to obtain a respiration trace, and do not need to output photoplethysmography signals. Thus, other reflectance or transmittance sensors may be used.

(59) Two breathing types are discussed above. Thoracic (chest) and diaphragmatic (belly) breathing are the two most common types of breathing. There is also clavicular breathing, which is breathing into the top third of the lungs and no deeper. Clavicular breathing is accomplished by raising the collarbone (clavicle) and shoulders during the inhalation and keeping the rest of the torso motionless. Clavicular breathing is the most shallow type of breathing. Clavicular breathing may also be distinguishable, or else it may be identified as chest breathing.

(60) While the invention has been illustrated and described in detail in the drawings and foregoing description, such illustration and description are to be considered illustrative or exemplary and not restrictive; the invention is not limited to the disclosed embodiments.

(61) Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word comprising does not exclude other elements or steps, and the indefinite article a or an does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measured cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.