Improvements in or relating to ear apparatus

Abstract

An ear apparatus (1), locatable at, or at least partly within, an ear of a human or other animal, for detecting change in inspiratory duration and / or expiratory duration of said human or other animal, the apparatus (1) comprises: one or more, or a plurality of, sensor means (6), configured to detect respiratory data comprising: onset of inhalation; and transition to exhalation, and means for analysing and / or processing (7), or means for communicating (8) for subsequent analysing and / or processing, the respiratory data, and deriving therefrom a plurality of respiratory data points for one or more respiratory cycles of said human or other animal from which change in inspiratory duration and / or expiratory duration is deriveable.

Description

[0001] Improvements in or relating to Ear Apparatus

[0002] The present invention relates to an ear apparatus, locatable at, or at least partly within, an ear of a human or other animal, for detecting change in inspiratory duration and / or expiratory duration of said human or other animal, and an associated method.

[0003] As asthma worsens, airways narrow owing to inflammation and smooth muscle spasm in the airways. When a person breaths-in (inspiratory phase), the airways are naturally opened by the negative pressure within the chest cavity - the volume of which is expanded by the surrounding respiratory muscles - raising the ribs and lowering the diaphragm. However, on breathing out (expiratory phase), the chest cavity volume is reduced, increasing pressure, and this pressure effect constricts the already tightened airways of the asthma sufferer. As the asthma worsens, with increasing inflammation, the length of time that it takes for the person to breath-out noticeably prolongs, compared with the time taken to breath in. There is, therefore, a decrease in the inspiratory duration : expiratory duration ratio (the l:E ratio), or the duty cycle (the ratio of the inspiratory duration to the total respiratory cycle duration).

[0004] During some conditions, such as asthma, where airway obstruction is reversible, the degree of obstruction may vary with certain factors including: time of day (typically worse in the morning than mid-afternoon); degree of inflammation; and response to treatment. In other conditions such as COPD (chronic obstructive pulmonary disease), airway obstruction is considered irreversible, i.e. will not substantially alter with the above factors.

[0005] As asthma airway obstruction typically varies throughout the day, real-time, real-World monitoring (i.e. during normal daily activities) would offer the ability to detect those changes and provide information on the degree of obstruction and the degree of variability. However, this not achievable with present apparatus which are unwieldly, or simply inaccurate.

[0006] Peak expiratory flow rate (PEFR) monitors are devices into which the subject forcibly blows and measures the peak flow rate in expiration. As airways narrow - when affected by, for example, asthma - the PEFR reduces. A change in PEFR may, therefore, be indicative of asthma inflammation. However, this is an inconvenient measure owing to the apparatus required and / or the method itself - it is often poorly performed, and patients I subjects do not readily co-operate with the method. This is, therefore, not a widely used and performed measure, and is not feasible with young children. Although asthma monitoring may be achieved through detecting PEFR and variations in time, which may identify prolonged expiration, it suffers from the same disadvantages.

[0007] Respiratory rate (RR - number of breaths per minute) is an important measure for such conditions as sepsis, pneumonia and other respiratory conditions, heart failure and acute severe asthma. However, raised respiratory rate may be a very late sign, for example in acute severe asthma. There is, therefore, a need for earlier detection of change in respiratory conditions.

[0008] A known apparatus for breathing rate monitoring for conditions such as asthma consists of a tightly fitting belt to be placed around the chest, encompassing accelerometers that detect breathing movements. These are not easily worn and, so, not suitable for everyday life, are poorly accepted by patients and are, therefore, not widely used.

[0009] Respiratory rate may be monitored from some medical devices using photoplethysmography (PPG). PPG is an optical technique well-known in pulse oximeters, registering the volume of blood in an area of target tissue. This volume of blood changes with the transmitted pulse pressure with each heart-beat, providing time series data representing a pulse waveform. Autonomic nervous system

[0010] 1

[0011] P24365WOOO changes and intra thoracic pressure changes during the breathing cycle are detectable from some finger based PPG I pulse oximeters and cause respiratory- dependent changes in the pulse waveform, including interbeat variability of heart rate (the interbeat interval reduces with inspiration and increases during expiration), amplitude variation of the pulsatile signal of the PPG (reducing during inspiration and increasing during expiration), and DC variability (or baseline wander).

[0012] All three aspects of respiratory related variation have been analysed to assess respiratory rate. However, this has shown low quality signal owing to the low number of data points per respiratory cycle (e.g. only 4 points for a respiratory rate of 15 breaths / m inute and pulse rate of 60bpm), as the respiratory data points occur with each peak-trough pulsatile pulse signal, when using interbeat interval and / or pulse amplitude variation. Similarly, respiratory data relating to expiratory and inspiratory changes are present in an ECG (electrocardiography) and / or MCG (magnetocardiography) signal. These also include interbeat timing differences and signal amplitude variations.

[0013] A number of different sensor types have been previously proposed to record respiratory rate or pattern from an ear of a subject; however, each has its own unique disadvantages or failings.

[0014] Inertial measurement units (IMlls) have been proposed for detecting: respiratory rate; head related respiratory movements; and / or pulse I heart-beat related respiratory patterns, using ballistocardiography (BCG) for detecting head movement with pulse pressure waves.

[0015] PPG has been proposed for detecting: respiratory rate, including pulse related respiratory patterns; and duty cycle I inspiratory: expiratory ratio, including pulse related respiratory patterns in COPD categorisation.

[0016] Microphones have been proposed for detecting: duty cycle, inspiratory : expiratory ratio, including audible breathing sounds in COPD categorisation; and audible pulse detection of interbeat variability (from a microphone in a sealed ear-canal).

[0017] Experiments with IMlls have attempted to record respiratory rate by detecting head related respiratory movements; however, at high error rates and with poor accuracy.

[0018] Experiments with PPG have attempted to measure respiratory rate using a PPG sensor at the external ear (pinna). Duty cycle monitoring from the outer-ear PPG has been attempted from COPD patients, but found to be inaccurate. Further, it was found that PPG for respiratory induced intensity variations (DC baseline wander of PPG signal) from an ear PPG was inaccurate compared with measuring chest movements. It should be noted that PPG data provides, effectively, a single respiratory data point per heart beat (similar to ECG, MCG and BCG). Accordingly, this mode of detection provides very few data points and, certainly, not enough to provide a measure of duty cycle.

[0019] Experiments with accelerometers have attempted to measure respiration with an in-ear accelerometer, with the aim of detecting heart rate respiratory timing variability by detecting pulsatile head movement related to pressure waves transmitted to the head with each heart-beat, i.e. ballistocardiography. However, as this analysis is similar to respiratory rate detection from PPG, it suffers from the same low quality signal owing to the low number of data points per respiratory cycle.

[0020] Experiments with microphones have attempted to measure inspiratory : expiratory ratios from a chest wall using a digital stethoscope. However the stethoscope is not an ear apparatus, not wearable, and was placed in a number of different sites on the chest. Further, it has been proposed to detect respiratory rate from respiratory noises from microphones located in or near to the ear-canal; however, this provides results of poor quality with breath sounds often being

[0021] 2

[0022] P24365WOOO practicably imperceptible, and the method cannot distinguish different parts of a full breathing cycle. Accordingly, directly estimating respiratory rate using breathing sounds from ear sensors is unreliable from frequency analysis techniques.

[0023] There is, therefore, a need for an improved apparatus and method for monitoring respiratory waveform timings, having greater accuracy and more reliability of detection than is presently known in the art. There is, also, a need for such an apparatus which is wearable during normal daily activities.

[0024] According to a first aspect, the present invention provides an ear apparatus, locatable at, or at least partly within, an ear of a human or other animal, for detecting change in inspiratory duration and / or expiratory duration of said human or other animal, the apparatus comprises: one or more, or a plurality of, sensor means, configured to detect respiratory data, the respiratory data comprising: onset of inhalation; and transition to exhalation, and means for analysing and / or processing, or means for communicating for subsequent analysing and / or processing, the respiratory data, and deriving therefrom a plurality of respiratory data points for one or more, or a plurality of, respiratory cycles of said human or other animal from which change in inspiratory duration and / or expiratory duration is deriveable, wherein the means for analysing and / or processing is configured to identify a subset of respiratory data points surrounding onset of inhalation and surrounding transition to exhalation, for more accurately determining change in inspiratory duration and / or expiratory duration.

[0025] Preferably, the ear apparatus is a wearable ear apparatus.

[0026] Preferably, inspiratory duration may stay static whilst expiratory duration may prolong. Alternatively, both inspiratory duration and expiratory duration may prolong.

[0027] Preferably, the ear apparatus is easily worn during daily activities, making such daily activities unimpeded by the ear apparatus.

[0028] Preferably, the one or more, or a plurality of, sensor means, comprise: movement detection sensor(s), which are configured to detect respiratory data from movement of the ear apparatus and / or from movement of a head of said human or other animal; non-movement detection sensor(s), which do not rely upon movement for detection of respiratory data; and / or combinations of movement and non-movement sensor(s).

[0029] Preferably, the one or more, or a plurality of, sensor means are configured to detect movement comprising: a roll backwards of a head of said human or other animal during inhalation; and a roll forwards of said head during exhalation.

[0030] Preferably, the one or more, or a plurality of, sensor means, are configured to detect movement of a / said head of said human or other animal comprising any one or more of the group comprising: a) rotation of said head around an axis approximately, or substantially, parallel to at least part of one or both ear-canals of said human or other animal; b) movement of said head along an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or c) movement of said head along an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.

[0031] Preferably, detecting movement of a / the head of said human or other animal comprising rotation of the head around an axis approximately, or substantially, parallel to one or both ear-canals at the or their respective external opening(s).

[0032] 3

[0033] P24365WOOO Preferably, the one or more, or a plurality of, sensor means, are configured to detect movement of a / said head of said human or other animal comprising any one or more of the group comprising: d) movement of said head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of said human or other animal; e) rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or f) rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.

[0034] Preferably, the one or more, or a plurality of, sensor means, are configured to detect movement of a / said head of said human or other animal comprising any vectoral movement being a fusion of movements along or around any one or more axes.

[0035] Preferably, configuring detection of an accelerometer sensor and / or a gyroscope sensor along and / or around an axis of detection approximately, or substantially, parallel to at least part of an ear-canal of said human or other animal comprises detecting movement along and / or around a transverse axis of the head of the human or other animal.

[0036] Preferably, configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to detect of movement comprising: rotation of said head around and / or movement of said head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of said human or other animal; movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.

[0037] Preferably, the means for analysing and / or processing is configured to determine respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of a / said head comprising: rotation of said head around and / or movement of said head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of said human or other animal; movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.

[0038] Preferably, the means for analysing and / or processing comprises an algorithm for determining respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of said head.

[0039] Preferably, the algorithm provides an analysis configuration, determining which one or combination of the sensor inputs provides the best respiratory data.

[0040] Preferably, configuring at least one detection axis of an inertial measurement unit (IMU) to provide sensor data from detection of movement along and / or around an axis of detection approximately, or substantially, parallel to an ear-canal of said human or other animal.

[0041] 4

[0042] P24365WOOO Preferably, configuring a detection axis, or detection axes, of an accelerometer sensor, gyroscope sensor and / or magnetometer sensor to provide sensor data from detection of movement of a / said head: along and / or around an ear-canal axis (i.e. medial to lateral direction), or vice versa, with respect to said human or other animal; along and / or around a vertex-caudal axis (i.e. head to toe direction), or vice versa, with respect to said human or other animal; and / or along and / or around an anterior-posterior axis (i.e. front to back direction), or vice versa, with respect to said human or other animal.

[0043] Preferably, the means for analysing and / or processing is configured to receive sensor data from: a) one or more axes of detection of an accelerometer sensor; and / or b) one or more axes of detection of a gyroscope sensor, and further configured to detect movement based upon one or more axes of detection having greater signal to noise ratio for respiratory, inspiratory, and / or expiratory timing data.

[0044] Preferably, at least part of the ear apparatus comprises a sensor housing, locatable in or around an ear of said user, so as to provide one or more, or a plurality of, sensor means, in a condition where the one or more sensors means provide sensor data from detection of movement: a) along and / or around an axis of detection substantially parallel to an earcanal (i.e. a direction along or around the ear-canal) of said human or other animal; b) along and / or around a vertex-caudal axis (i.e. head to toe direction), or vice versa, with respect to said human or other animal; and / or c) along and / or around an anterior-posterior axis (i.e. front to back direction), or vice versa, with respect to said human or other animal.

[0045] Preferably, the curve of data points plotted against time is analysed to detect aspects including: area under the curve for inspiratory and / or expiratory phases; slope; and amplitude. Analysis further includes: comparison between inspiratory and expiratory phases, including by means of shapes, curves, gradients, and timing of first, second or higher order derivative analyses; confluence, smoothness or other features of the relevant plotted curves; and / or analysis by machine learning and computer vision processes.

[0046] Preferably, the above analytic outputs are compared between: various inspiratory and expiratory periods; time periods including hours or days for an individual; and / or the individual and data from larger populations, for example of health volunteers of the same or similar age.

[0047] Preferably, the apparatus is configured to provide signal fusion and / or signal averaging, including ensemble averaging, of PPG, audio, IMU and / or other sensor data encompassing respiratory data from the one or more, or a plurality of, sensor means.

[0048] Preferably, the apparatus is configured to provide signal fusion and / or signal averaging, including ensemble averaging, of PPG, audio, IMU and / or other sensor data encompassing one or more sets of respiratory data from the same sensor means.

[0049] Preferably, the one or more, or a plurality of, sensor means, is / are configured to detect respiratory data comprising: exhalation end point; any inhalation pause; and / or any exhalation pause, in the one or more respiratory cycles.

[0050] 5

[0051] P24365WOOO Preferably, the means for analysing and / or processing is configured to identify a plurality of respiratory data points: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause, for more accurately determining change in inspiratory duration and / or expiratory duration.

[0052] Preferably, the means for analysing and / or processing excludes respiratory data points being a predetermined time period outside of: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause.

[0053] Preferably, the one or more, or a plurality of, sensor means, is / are configured to detect a predetermined number of respiratory data points per respiratory cycle.

[0054] Preferably, the one or more, or a plurality of, sensor means, is / are configured to adapt the number of respiratory data points detected and / or analysed, based upon an input sensor type.

[0055] Preferably, the one or more, or a plurality of, sensor means, is / are configured to detect 10 to 100, or 20 to 80, or 30 to 70, or 40 to 60, respiratory data points per respiratory cycle for a PPG input signal.

[0056] Preferably, the means for analysing and / or processing is configured to identify a predetermined number of respiratory data points per respiratory cycle.

[0057] Preferably, the means for analysing and / or processing is configured to adapt the number of respiratory data points detected and / or analysed, based upon an input sensor type.

[0058] Preferably, the means for analysing and / or processing is configured to identify 10 to 100, or 20 to 80, or 30 to 70, or 40 to 60, respiratory data points per respiratory cycle for a PPG input signal.

[0059] Preferably, the means for analysing and / or processing is configured to amalgamate a plurality of individual sensor readings to derive data at a single respiratory timing point.

[0060] Preferably, the means for analysing and / or processing is configured to align signals from different sensors, and / or different respiratory cycles, at a respiratory timing point and / or at a defined point in a / the respiratory cycle, enhancing the signal and, thereby, accuracy of data at a / the respiratory timing point, whilst averaging out the signal noise.

[0061] Preferably, enhancing the signal and, thereby, accuracy of onset of inhalation, transition to exhalation, exhalation end point, any inhalation pause, and / or any exhalation pause. Preferably, the apparatus is configured to detect change in inspiratory duration and / or expiratory duration over a predetermined detection period. Further preferably, the apparatus is configured to detect change in exhalation end point, any inhalation pause, and / or any exhalation pause.

[0062] Preferably, the apparatus is configured to detect change in inspiratory duration and / or expiratory duration over a detection duration of: a) up to 30 seconds; b) up to 20 respiratory cycles; or c) a single respiratory cycle.

[0063] Preferably, the means for analysing, processing and / or communicating is configured to derive a plurality of data points during an inspiratory phase of said respiratory cycle.

[0064] 6

[0065] P24365WOOO Preferably, the means for analysing, processing and / or communicating is configured to derive a plurality of data points during an expiratory phase of said respiratory cycle.

[0066] Preferably, the means for analysing, processing and / or communicating is configured to derive absolute and / or relative duration data for inspiratory and / or expiratory duration.

[0067] Preferably, the means for analysing and / or processing the respiratory data is configured to isolate onset of inhalation and transition to exhalation data from other data detectable at an ear of said individual.

[0068] Preferably, the means for analysing and / or processing the respiratory data is additionally configured to isolate exhalation end point; and / orany inhalation I exhalation pause data from other data detectable at an ear of said individual.

[0069] Preferably, the means for analysing and / or processing the respiratory data comprises an algorithm. Preferably, an output of the algorithm comprises: onset of inhalation and transition to exhalation and, further preferably, end of exhalation and / or any inhalation I exhalation pause.

[0070] Preferably, the one or more, or a plurality of, sensor means comprise one or more of the following: a motion detector; an optical detector; an inertial measurement unit; a three- six-, or more, axis inertial measurement unit; a photoplethysmogram (PPG) sensor, being a contact or non-contact sensor; a ballistocardiogram (BCG) sensor; an electrode for electrocardiographic or electroenchephalographic monitoring; an ultrasound transducer; a radio frequency detector; a magnetometer or electromagnetic sensor; and / or a microphone.

[0071] Preferably, the one or more, or a plurality of, sensor means is / are located in: an in-ear portion, shaped to be received in a concha and / or an ear-canal of said human or other animal; an outer-ear portion, shaped to be received in or on an outer pinna of said human or other animal; a behind the ear portion or over the ear portion, shaped to be received behind an ear of said human or other animal and / or over an ear of said human or other animal; and / or an associated apparatus, being located adjacent an ear of said human or other animal or in connection with wearable ear apparatus.

[0072] Preferably, the apparatus is an audio earphone, audio earbud, sleep bud, earplug, hearing aid or other ear device.

[0073] Preferably, the means for communicating the respiratory data comprises: wireless communications apparatus, or a wired communications apparatus, for communicating with: remote means for analysing and / or processing the respiratory data; or remote further means for analysing and / or processing the respiratory data.

[0074] Preferably, the apparatus further comprises, preferably remote, storage means and / or, preferably remote, display means for storing and / or displaying the respiratory data.

[0075] Preferably, the means for analysing and / or processing is configured to analyse and compare independent accelerometer, gyroscope and / or magnetometer channels so as to dynamically choose the channel providing the highest quality respiratory data.

[0076] 7

[0077] P24365WOOO Preferably, the apparatus dynamically chooses which channel of a / the inertial measurement unit provides the highest quality respiratory data.

[0078] Preferably, the means for analysing and / or processing is configured to fuse and / or average respiratory cycle data to increase the number of data points.

[0079] Preferably, aligning several respiratory waveforms to align the shape to determine the average. Accordingly, noise that is out of synch is, effectively cancelled out whilst the desired signal is reinforced.

[0080] Preferably, the means for processing and / or analysing proactively selects data from a multitude of sensors dependent on quality of signal to combine, to determine a respiratory pattern suitable to detect inspiratory duration and expiratory duration. For instance, one sensor may detect onset of inhalation but then be overcome with signal noise, and other sensors may contribute to the detecting transition to exhalation and / or finish of exhalation.

[0081] Preferably, analysis by machine learning techniques provides an output related to inspiratory duration and expiratory duration from inputs from a combination of sensor types.

[0082] Preferably, sensors located in both ears provide data to improve quality either by synchronous I reciprocal or other relationship between sensors in different ears, being either different or of the same type.

[0083] Preferably, the invention may benefit from additional information from external sensors. For instance, externally facing microphones or other IMlls on the ear device, or on other devices such as watches, mobile phones, other wearables, jewellery, etc.

[0084] Preferably, an IMU, or other sensor, detects onset of respiration, providing time of onset of respiration, or detects a further or different defined point, to allow synchronisation and summation I fusion of data from other sensors.

[0085] According to a second aspect, the invention provides a method for detecting change in inspiratory duration and / or expiratory duration during one or more, or a plurality of, respiratory cycles of a human or other animal, the method comprising: locating an ear apparatus at, or at least partly within, an ear of the human or other animal, the ear apparatus comprising one or more, or a plurality of, sensor means; detecting respiratory data, using the one or more, or plurality of sensor means, to detect respiratory data comprising: onset of inhalation; and transition to exhalation, and analysing and / or processing, or communicating for subsequent analysing and / or processing, the respiratory data so as to derive therefrom a plurality of respiratory data points for the one or more, or plurality of, respiratory cycles and deriving from the plurality of respiratory data points change in inspiratory duration and / or expiratory duration of the human or other animal, wherein identifying a subset of respiratory data points surrounding onset of inhalation and surrounding transition to exhalation, to more accurately determine change in inspiratory duration and / or expiratory duration.

[0086] Preferably, wearing the ear apparatus during daily activities, without impeding such daily activities. Preferably, monitoring the constriction of airways.

[0087] Preferably, detecting respiratory data comprises: detecting respiratory data from movement of the ear apparatus and / or from movement of a head of said human or other animal; detecting respiratory data without, or which does not rely upon, movement for detection; and / or detecting data from combinations of movement and non-movement sensor(s).

[0088] 8

[0089] P24365WOOO Preferably, the one or more, or a plurality of, sensor means detecting movement comprising: a roll backwards of a head of the human or other animal during inhalation; and a roll forwards of the head during exhalation.

[0090] Preferably, the one or more, or a plurality of, sensor means, detecting movement of a / the head of said human or other animal comprising any one or more of the group comprising: a) rotation of the head around an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal; b) movement of the head along an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or c) movement of the head along an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.

[0091] Preferably, detecting movement of a / the head of said human or other animal comprising rotation of the head around an axis approximately, or substantially, parallel to one or both ear-canals at the or their respective external opening(s).

[0092] Preferably, the one or more, or a plurality of, sensor means, detecting movement of a / the head of the human or other animal comprising any one or more of the group comprising: d) movement of the head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal; e) rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or f) rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.

[0093] Preferably, the one or more, or a plurality of, sensor means, detecting movement of a / the head of the human or other animal comprising any vectoral movement being a fusion of movements along or around any one or more axes.

[0094] Preferably, configuring detection of an accelerometer sensor and / or a gyroscope sensor along and / or around an axis of detection approximately, or substantially, parallel to at least part of an ear-canal of said human or other animal comprises detecting movement along and / or around a transverse axis of the head of the human or other animal.

[0095] Preferably, configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to detect of movement comprising: rotation of the head around and / or movement of the head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of the human or other animal; movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.

[0096] Preferably, configuring the means for analysing and / or processing to determine respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of a / the head comprising: rotation of the head around and / or movement of the head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of the human or other animal; movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or

[0097] 9

[0098] P24365WOOO movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.

[0099] Preferably, utilising an algorithm for determining respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of said head.

[0100] Preferably, configuring at least one detection axis of an inertial measurement unit (IMU) to provide sensor data from detection of movement along and / or around an axis of detection approximately, or substantially, parallel to an ear-canal of said human or other animal.

[0101] Preferably, configuring a detection axis, or detection axes, of an accelerometer sensor, gyroscope sensor and / or magnetometer sensor to provide sensor data from detection of movement of a / the head: along and / or around an ear-canal axis (i.e. medial to lateral axis), or vice versa, with respect to the human or other animal; along and / or around a vertex-caudal axis (i.e. head to toe axis), or vice versa, with respect to the human or other animal; and / or along and / or around an anterior-posterior axis (i.e. front to back axis), or vice versa, with respect to the human or other animal.

[0102] Preferably, receiving sensor data from: a) one or more axes of detection of an accelerometer sensor; and / or b) one or more axes of detection of a gyroscope sensor, and detecting movement based upon one or more axes of detection having greater signal to noise ratio for respiratory, inspiratory, and / or expiratory timing data.

[0103] Preferably, wherein at least part of the ear apparatus comprises a sensor housing, being located in or around an ear of the user, the sensor housing locating one or more, or a plurality of, sensor means, in a defined condition where the one or more sensor means detect sensor data from movement: a) along and / or around an axis of detection substantially parallel to an earcanal (i.e. a direction along or around the ear-canal) of the human or other animal; b) along and / or around a vertex-caudal axis (i.e. head to toe direction), or vice versa, with respect to the human or other animal; and / or c) along and / or around an anterior-posterior axis (i.e. front to back direction), or vice versa, with respect to the human or other animal.

[0104] Preferably, analysing the curve of data points plotted against time to detect aspects including: area under the curve for inspiratory and / or expiratory phases; slope; and amplitude.

[0105] Preferably, analysis further includes: comparison between inspiratory and expiratory phases, including by means of shapes, curves, gradients, and timing of first, second or higher order derivative analyses; confluence, smoothness or other features of the relevant plotted curves; and / or analysis by machine learning and computer vision processes.

[0106] Preferably, comparing the above analytic outputs between: various inspiratory and expiratory periods; time periods including hours or days for an individual; and / or the individual and data from larger populations, for example of health volunteers of the same or similar age.

[0107] Preferably, the method comprising fusing and / or averaging respiratory data from one or more, or a plurality of, sensor means.

[0108] Preferably, the method comprising fusing and / or averaging one or more sets of respiratory data from the same sensor means. Preferably, detecting respiratory data comprising: exhalation end point; any inhalation pause; and / or

[0109] 10

[0110] P24365WOOO any exhalation pause, in the one or more respiratory cycles.

[0111] Preferably, identifying a subset of respiratory data points surrounding onset of inhalation and surrounding transition to exhalation, for more accurately determining change in inspiratory duration and / or expiratory duration.

[0112] Preferably, identifying a plurality of respiratory data points: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause, for more accurately determining change in inspiratory duration and / or expiratory duration.

[0113] Preferably, excluding respiratory data points being a predetermined time period outside of: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause.

[0114] Preferably, detecting or identifying a predetermined number of respiratory data points per respiratory cycle.

[0115] Preferably, adapting the number of respiratory data points detected and / or analysed, based upon an input sensor type.

[0116] Preferably, detecting 10 to 100, or 20 to 80, or 30 to 70, or 40 to 60, respiratory data points per respiratory cycle for a PPG input signal.

[0117] Preferably, identifying 10 to 100, or 20 to 80, or 30 to 70, or 40 to 60, respiratory data points per respiratory cycle for a PPG input signal.

[0118] Preferably, amalgamating a plurality of individual sensor readings to derive data at a single respiratory timing point.

[0119] Preferably, aligning signals from different sensors, and / or different respiratory cycles, at a respiratory timing point and / or at a defined point in a / the respiratory cycle, enhancing the signal and, thereby, accuracy of data at a / the respiratory timing point, whilst averaging out the signal noise.

[0120] Preferably, enhancing the signal and, thereby, accuracy of onset of inhalation, transition to exhalation, exhalation end point, any inhalation pause, and / or any exhalation pause.

[0121] Preferably, detecting change in inspiratory duration and / or expiratory duration over a predetermined detection period. Further preferably, detecting change in exhalation end point, any inhalation pause, and / or any exhalation pause.

[0122] Preferably, detecting change in inspiratory duration and / or expiratory duration over a detection duration of: a) up to 30 seconds; b) up to 20 respiratory cycles; or c) a single respiratory cycle.

[0123] Preferably, deriving a plurality of data points during an inspiratory phase of said respiratory cycle.

[0124] Preferably, deriving a plurality of data points during an expiratory phase of said respiratory cycle.

[0125] Preferably, deriving absolute and / or relative duration data for inspiratory and / or expiratory duration.

[0126] Preferably, isolating onset of inhalation and transition to exhalation data from other data detectable at an ear of the individual.

[0127] Preferably, further isolating exhalation end point; and / or any inhalation I exhalation pause data from other data detectable at an ear of the individual.

[0128] 11

[0129] P24365WOOO Preferably, analysing the sensor data with an algorithm. Further preferably, the algorithm outputting: onset of inhalation and transition to exhalation and, further preferably, end of exhalation and / or any inhalation I exhalation pause.

[0130] Preferably, locating the one or more, or a plurality of, sensor means in: an in-ear portion, shaped to be received in a concha and / or an ear-canal of the human or other animal; an outer-ear portion, shaped to be received in or on an outer pinna of the human or other animal; a behind the ear portion or over the ear portion, shaped to be received behind an ear of the human or other animal and / or over an ear of the human or other animal; and / or an associated apparatus, being located adjacent an ear of the human or other animal or in connection with wearable ear apparatus.

[0131] Preferably, comparing independent accelerometer, gyroscope and / or magnetometer channels so as to dynamically choose the channel providing the highest quality respiratory data.

[0132] Preferably, dynamically choosing which channel of a / the inertial measurement unit provides the highest quality respiratory data.

[0133] Preferably, fusing and / or averaging respiratory cycle data to increase the number of data points.

[0134] Preferably, aligning several respiratory waveforms to align the shape to determine the average.

[0135] Preferably, proactively selecting data from a multitude of sensors dependent on quality of signal and combining to determine a respiratory pattern suitable to detect inspiratory duration and expiratory duration.

[0136] Preferably, analysing using machine learning techniques to provide an output related to inspiratory duration and expiratory duration from inputs from a combination of sensor types.

[0137] Preferably, locating sensor at or in both ears to provide data to improve quality either by synchronous I reciprocal or other relationship between sensors in different ears, being either different or of the same type.

[0138] Preferably, analysing additional data from external sensors, such as externally facing microphones or other IMlls on the ear device, or on other devices such as watches, mobile phones, other wearables, jewellery, etc.

[0139] Preferably, an IMU, or other sensor, detecting onset of respiration, providing time of onset of respiration, and synchronising and summation I fusing of data from other sensors.

[0140] Preferably, the ear apparatus is according to the first aspect.

[0141] The present invention may also relate to an ear apparatus, locatable at, or at least partly within, an ear of a human or other animal, for detecting change in inspiratory duration and / or expiratory duration of said human or other animal, the apparatus comprises: one or more, or a plurality of, sensor means, configured to detect respiratory data, the respiratory data comprising: onset of inhalation; and transition to exhalation, and means for analysing and / or processing, or means for communicating for subsequent analysing and / or processing, the respiratory data, and deriving therefrom a plurality of respiratory data points for one or more, or a plurality of, respiratory cycles of said human or other animal from which change in inspiratory duration and / or expiratory duration is deriveable. Preferably, comprising any one or more features of the first aspect.

[0142] 12

[0143] P24365WOOO The present invention may also relate to a method for detecting change in inspiratory duration and / or expiratory duration during one or more, or a plurality of, respiratory cycles of a human or other animal, the method comprising: locating an ear apparatus at, or at least partly within, an ear of the human or other animal, the ear apparatus comprising one or more, or a plurality of, sensor means; detecting respiratory data, using the one or more, or plurality of sensor means, to detect respiratory data comprising: onset of inhalation; and transition to exhalation, and analysing and / or processing, or communicating for subsequent analysing and / or processing, the respiratory data so as to derive therefrom a plurality of respiratory data points for the one or more, or plurality of, respiratory cycles and deriving from the plurality of respiratory data points change in inspiratory duration and / or expiratory duration of the human or other animal. Preferably, comprising any one or more features of the second aspect.

[0144] The present invention may relate to a method in which a sensor, worn by a human or other mammal within or attached to audio earphone, audio earbud, sleep bud, earplugs, hearing aids or other ear devices, detects data incorporating physiological respiratory characteristics detectable at the ear, outputs the data to a processor, which may be connected by wire, or remote and wirelessly connected, an algorithm of the processor analyses times series or aggregated data incorporating, preferably 40 to 60 time series signal data points per output respiratory waveform, and provides an output characterising the absolute and / or relative duration of physiological data related separately to both inspiratory and expiratory duration.

[0145] The following paragraphs provide general preferred aspects of the present invention.

[0146] The number of respiratory cycles to monitor may depend upon the quality of the signal and / or quality of the data perse. For instance, a good quality signal having clearly defined data points may require only a single respiratory cycle for deriving duty cycle, whereas a less good quality signal may require many more, such as 6 or more cycles.

[0147] Sensors of the present invention may be as follows: one or more PPG sensors (pinna and / or in-ear; contact based and / or standoff); one or more IMlls, including accelerometers and / or gyroscopes (in ear-canal, in concha bowl, on outer pinna, behind ear, on earphone wire or lead, or device attached to wire or lead); and / or one or more microphones (in-ear, in ear-canal, in concha bowl and / or behind the ear).

[0148] Such sensors may have the following characteristics:

[0149] PPG, representing pulse timing, AC I DC variations by averaging cycles to increase number of data points; ballistocardiogram from IMU, representing pulse timing changes with respiration, AC I DC variations:

[0150] IMU >200Hz, high resolution movement data; one or more accelerometers; and / or one or more gyroscopes; accelerometers and / or gyroscopes detecting respiratory related head movements; and / or

[0151] 13

[0152] P24365WOOO microphones, representing respiratory sounds, interbeat variability, or other audible pulse signal characteristics, by detecting audible pulse signals in an occluded ear canal.

[0153] An algorithm of the present invention, preferably by way of it being part of the means for processing and / or analysing, may receive sensor data from the one or more PPG sensors, IMU and / or microphones, and operate as follows.

[0154] PPG detection at, around or within, the ear is more reliable than non-ear PPG sensors because the PPG source is more central, being closer to the brain and heart, as compared to a wrist or finger. Accordingly, the PPG source is less likely to be affected by peripheral modulation (e.g. through vasoconstriction, poor circulation, hypotension, etc.) and limb movement factors.

[0155] BCG detection in a vertical orientation (e.g. opposite to a gravitational acceleration signal) is determined from outputs of 6-axis IMU. This may also be used for detecting and / or averaging BCG and other signals. The algorithm will analyse the signal and utilise particular channel(s) from the IMU which provide the best respiratory signal, or respiratory related signal, - preferably, one of gyroscope or accelerometer channels.

[0156] The processor determines the optimum signal, e.g. the IMU channel with cleanest respiratory signal, or respiratory related signal, and / or amalgamates signals from different sensors of the 6 axes dynamically throughout the respiratory cycle, to use the most prominent at each phase in the cycle. The IMU data may also provide discrete timing points (such as start and / or end) between individual breath cycles. By contrast, PPG signals show continuous variability without a discrete differential between two breathing cycles, making PPG, effectively, blind to a pause in movement which an IMU can detect - the pause being a period at the end of expiration and / or at the end of inspiration where the subject is not inhaling I exhaling. PPG is also unable to accurately time the onset of inspiration, the transition from inspiration to expiration and / or the end of respiration.

[0157] Following analysis of the sensor data, the output of the algorithm is, preferably, inspiratory duration and expiratory duration. This may include a trace of inspiratory and expiratory data showing the difference(s) in characteristics. The output of the algorithm may include: end of respiratory cycle; relative amplitude of signal for inspiration to expiration; volume variability or effort; and / or determination of a bigger or deeper breath.

[0158] A typical respiratory duty cycle for a healthy person may be 40%, i.e. approximately 40% of the duration of total inspiration I expiration cycle being taken up by inspiration, and 60% taken up by expiration.

[0159] In conditions such as COPD, the relative duration of expiratory phase extends, altering the percentage weightings to, for example, 30% in inspiration and 70% in expiration. Similar changes happen dynamically with active asthma and during an asthma exacerbation, extending the duty cycle.

[0160] Determining a change in duty cycle from, for example, 40% to 30% during active asthma or an exacerbation would be very advantageous. Assuming that this reduction in duty cycle is linked to a prolongation of the overall duration of the respiratory cycle, this would represent a prolongation of the expiratory phase. By way of example, if a baseline total cycle length is 5 seconds (e.g. 12 breaths per minute) at 40% duty cycle, the expiratory phase lasts 3 seconds. With a reduction of the duty cycle (assuming prolongation of overall cycle) to 30%, then the expiratory phase extends to 4.65 seconds with Inspiratory phase stable at 2 seconds.

[0161] In the context of asthma, a 15% reversibility or 25% change has been indicated to be significant. So, ideally, one needs to be able to detect a 15% stepwise change in the total reduction of duty cycle by 10%, and requires far more respiratory data points than is achievable through standard PPG and BCG detectors

[0162] 14

[0163] P24365WOOO I analysis. The waveform data would need to demonstrate a reduction of 1 .5%, from 40% to 38.5%. For a respiratory rate of 12 breaths per minute (5 second cycle), this represents a prolongation of the expiratory phase from 3 seconds to 3.19 seconds. To detect this change, the Nyquist frequency equation predicts a minimal sample rate at double the frequency, i.e. 2 x 1 / 0.19 = 10Hz.

[0164] However, respiratory rate can vary from 6 to 60 breaths per minute or more and, so, the sampling frequency needs to encompass 60 breaths per minute, i.e. 1 second per cycle. A 40% duty cycle represents an expiratory phase of 0.6 seconds and, when reducing to 38.5% duty cycle, the expiratory phase would be approximately 0.64 seconds of a total breathing cycle duration to 1.039 seconds. So, the required frequency would be 2 x 1 / 0.04 = 50Hz.

[0165] Accordingly, at 12 breaths per minute, 5 seconds for each cycle, 10Hz sampling provides 50 data points per respiratory cycle. Conversely, at 60 breaths per minute, 1 second for each cycle, 50 Hz sampling provides 50 data points per respiratory cycle.

[0166] It should be noted that, as standard PPG and BCG detection provides only one data point per pulse wave cycle this would only provide 5 signal data points per breath cycle at a respiratory rate of 12 / min and pulse rate 60 / min, and 17 signal data points per breath cycle at respiratory rate of 12 / min and pulse rate 200 / min, being numbers too low for the accuracy required.

[0167] The algorithm and sensor of the present invention may provide at least 25 respiratory data points per respiratory cycle - preferably, at least 40 to 60.

[0168] For example, a 6-axis IMU, with accelerometer and / or gyroscope, operating at 300 Hz, the processor may select the best signal from its accelerometers and / or gyroscopes to provide high-accuracy and high-frequency inspiratory and expiratory data. Detection axes - represented by being capable of detecting movement around and / or along an axis of detection approximately, or substantially, parallel to an earcanal of the test subject - obtains data from movement of the head around and / or along a direction of the ear-canal. At a respiratory rate of 60 / minute, i.e. one breath per second, the invention provides 300 signal data points per respiratory cycle. In addition, the invention may include a method of active comparison between IMU channels to dynamically chose the best IMU sensor output.

[0169] By contrast, standard IMUs and microphones operating at higher frequencies do not provide good quality data. For instance, a standard IMU operating at 1000Hz for detecting head related respiratory cycle movements. Although it may sample much faster, the data provided includes fewer good quality data points, and is more inaccurate. Further, although a standard microphone may sample at a high frequency, it also exhibits poor signal quality, providing fewer good quality data points and is more inaccurate.

[0170] The present invention provides, preferably, at least 10 to 60 good quality signal data points per respiratory cycle which represent a true respiratory signal of inspiratory and expiratory activity. This enables accurate detection of inspiratory: expiratory pattern and / or duty cycle by the following methods i) to iii). i) Sampling a reliable single respiratory waveform at, preferably, 40 to 60 times within the waveform: a) using accelerometer and / or gyroscope detection around an axis of detection parallel to the ear-canal, analysing and / or adjusting the signal to detect a slight movement up and back with inhale of the subject, and corresponding down and forward with exhale (although this movement assumes a subject is sitting) and provides 300Hz high quality data; b) actively selecting a best waveform from IMU data:

[0171] 15

[0172] P24365WOOO 1 . looking and matching a known (or expected) waveform; and / or

[0173] 2. assessing if a signal in one channel is different to all (or a majority) of others; c) using accelerometer and / or gyroscope detection around an axis different to a) to detect a slight movement up and back with inhale of the subject, and corresponding down and forward with exhale to provide 300 Hz high quality data; ii) Sampling multiple waveforms to aggregate, preferably, the 40 to 60 good data points by: c) amalgamating PPG and / or BCG waveforms; d) amalgamating IMU signals; e) amalgamating audio I microphone signals for either or both of respiratory sounds and / or audible pulse detection; f) using timing for aggregating average data when sampling onset of inhalation and / or transition to exhalation, by combining PPG, IMU and / or sound data to improve collection of, preferably, 40 to 60 good data points.

[0174] The present invention may be a standalone apparatus or be integrated into an ear device, or non-ear device, having additional functionality.

[0175] The present invention provides an ear apparatus, preferably wearable ear apparatus, for monitoring changes of airflow dynamics.

[0176] Advantageously, the invention proposes using inspiratory : expiratory ratio and / or duty cycle to monitor constriction of the airways. The invention also proposes such a use in the real-World, during daily activities.

[0177] Advantageously, detection of movement around and / or along an axis of detection approximately, or substantially, parallel to an ear-canal of said human or other animal provides an improved respiratory signal, from which improved respiratory data points may be derived.

[0178] Advantageously, if a best axis of detection is unknown, or the one or more sensor means is / are located in a sub-optimal condition, or having sub-optimal alignment, the present invention may choose the sensor axis, or sensor axes, providing the best respiratory signal, and derive its plurality of respiratory data points therefrom.

[0179] Advantageously, an in-ear aspect of the present invention: ensures detection and handling of good quality data; and / or reduces signal noise from talking and chewing.

[0180] Advantageously, algorithm(s) of the present invention: calibrate the IMU; identify which channel of 6-axis IMU (or more axes) has best respiratory data; remove other (unwanted) signals, such as gross head movement and / or BCG signal; and / or consider bilateral measurement, such as by allowing isolation of synchronous movement only and / or removing a rotational I asymmetrical movement artefact (i.e. when walking).

[0181] Advantageously, analysing and / or processing data relating to gyroscope movement around an axis of detection being parallel to the ear-canal provides good quality signals suitable for deriving inspiratory : expiratory ratio and duty cycle. For asthma, at rest signals may be sufficient and / or suitable. However, the multiple-axis IMU provides the ability to remove movement detected by the accelerometer and, thereby, remove movement artefact noise from the gyroscope and / or IMU data.

[0182] Advantageously, the present invention provides an apparatus and method for real-time asthma monitoring for home use during normal daily activities. The apparatus may be earbuds, hearing aids, earplugs or other ear devices, locatable at

[0183] 16

[0184] P24365WOOO or around an ear of the subject. In addition, the present invention provides an apparatus and method which assists with monitoring such conditions as asthma and obstructive disorders, such as COPD (Chronic obstructive pulmonary disease), without being used in an excluded method of treatment. In an alternative the ear devices may be a dedicated medical monitoring apparatus.

[0185] Advantageously, the invention can derive patterns and / or rhythms of breathing which may be indicative of a change in health status of the subject. For example, gaps in breathing can be related to transient upper airways obstruction in obstructive sleep apnoea.

[0186] Advantageously, the addition and interpretation of data from audio earphones, audio earbuds, sleep buds, earbuds, hearing aids or other ear devices provide the ability to monitor for changes in breathing patterns, including inspiratory and expiratory ratios and rhythms, airway obstruction, etc. during usual daily activities, including at night, or during specific challenges, such as exercise or challenge tests.

[0187] Microphone data is cleaned I improved by fusion and / or averaging of data. PPG data is cleaned I improved by fusion and / or averaging of data.

[0188] Preferably, providing signal fusion and / or signal averaging, including ensemble averaging, of PPG, audio, IMU and / or other sensor data encompassing respiratory data from the one or more, or a plurality of, sensor means.

[0189] Preferably, providing signal fusion and / or signal averaging, including ensemble averaging, of PPG, audio, IMU and / or other sensor data encompassing one or more sets of respiratory data from the same sensor means.

[0190] Advantageously, the invention does not require a sealed ear-canal, allowing a subject to hear normally and the ear-canal to be ventilated.

[0191] In a preferred aspect, asthma monitoring may be conducted from earbuds, hearing aids, earplugs, or other ear devices, such as dedicated medical monitors.

[0192] Advantageously, the invention utilises signal averaging of PPG, audio, IMU and / or other sensor data so as to remove noise from a signal, enabling identification of, at least: onset of inhalation; and / or transition to exhalation. Preferably, enabling identification of any one or more of: exhalation end point; any inhalation pause; and / or any exhalation pause.

[0193] Although aspects of the invention may have been described as beneficial to identifying an onset of asthma, the invention is not a method of treatment of the human or animal body since it does not provide direct diagnosis of asthma, or any other medical condition. The method, perse, may be used to monitor a medical condition, or may be used as part of a suite of diagnosis steps or tests; however, it is maintained that this is not excluded subject-matter. Should it be required, the method may be restricted to ‘non-diagnosis’, ‘non-therapeutic’, etc.

[0194] Those skilled in the art will understand that an inhalation pause is a pause in breathing after completion of exhalation and before inhalation commences. Further, those skilled in the art will understand that an exhalation pause is a pause in breathing after completion of inhalation and before exhalation commences.

[0195] The term ‘individual’ herein is used to refer to a single human or other animal test subject.

[0196] Those skilled in the art will know that signal fusion I sensor fusion I data fusion is a process by which data from multiple sources is combined to provide more accurate, consistent and / or useful data.

[0197] Those skilled in the art will know that signal averaging is a technique by which an average of corresponding data is computed in a data series. It is often used to extract a signal waveform from background noise through evoking the signal

[0198] 17

[0199] P24365WOOO repeatedly whilst cancelling out any random noise. Consequently, if the input respiratory data is PPG, audio and / or IMU data, it improves the signal-to-noise ratio and can obtain a clearer picture of the PPG signal, audio and / or IMU signal.

[0200] Those skilled in the art will know that ensemble averaging is useful for reinforcing consistent and repeatable data in a signal, whilst cancelling or removing any noise, including random noise, that may vary during such repetitions. A specific point for timing (fiducial point) is defined, about which an ensemble of time varying signals are created, and averaging is conducted at intervals. Consequently, if the input respiratory data is PPG, audio, and / or IMU, which data is often hidden by noise, ensemble averaging provides a clearer picture of the PPG signal, audio signal and / or IMU signal.

[0201] The term ‘means’ may be replaced by the term 'device(s)’ or ‘apparatus’ depending upon context.

[0202] The invention will now be disclosed, by way of example only, with reference to the following drawings, in which:

[0203] Figure 1 is a schematic drawing of a first embodiment of ear apparatus;

[0204] Figure 2 is a schematic drawing of a second embodiment of ear apparatus; Figure 3 is a schematic drawing of a third embodiment of ear apparatus;

[0205] Figures 4a and 4b are schematic drawings showing relevant axes of detection from an ear apparatus of the invention, with respect to a whole head of a test subject; and

[0206] Figures 5a and 5b are plots of respiratory data obtained from an ear apparatus of the present invention.

[0207] Figure 1 shows a wearable ear apparatus, generally identified by reference 1 . The ear apparatus 1 includes an ear-canal portion 2, for inserting into an ear-canal of an individual, and an exterior ear portion 3, for resting in a pinna and / or concha region of an ear of the individual. As shown in Figure 1 , the ear-canal portion 2 is positioned within an ear-canal 4 and the exterior ear portion 3 is positioned in a concha region 5 of a human test subject.

[0208] The ear-canal portion 2 includes a six-axis IMU 6, being located along a central axis of the ear-canal portion 2, and the IMU 6 is configured within the earcanal portion 2 to detect various movements of the head of the test subject. For the avoidance of doubt, it should be noted that configured is a term which may mean an analysis configuration and / or a physical configuration, the latter being a defined physical orientation of one or more axes of detection of an IMU, or other sensor(s). In the present example, the IMU 6 is physically oriented. First and second axes (A and B) of the IMU are concerned with detecting rotation of the head around and / or movement of the head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the test subject. The IMU 6 is, therefore, located so as to be capable of detecting lateral movements of the head and / or when a head may roll forwards and backwards during beathing. In addition, third and fourth axes (C and D) of the IMU 6 are concerned with detecting movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the test subject. Further, fifth and sixth axes (E and F) of the IMU 6 are concerned with detecting movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the test subject.

[0209] The exterior ear portion 3 includes a processor 7, with an associated algorithm, for analysing sensor data received from the IMU 6 (and any other sensors). The processor 7 is for deriving a plurality of respiratory data points from the sensor data and determining change in inspiratory duration and / or expiratory

[0210] 18

[0211] P24365WOOO duration and, thereby, change in respiratory duty cycle. The output of the algorithm or processor 7 is communicated, by communicator 8, to an external database I display apparatus 9, provided for storing data and outputting respiratory data as and when required. The external database I display apparatus 9 could be a mobile phone, computer, or the like.

[0212] Figure 1 shows all six sensor / detection geometries, being shown by arrows having references A through to F. Arrows A and B identify the first and second axes of detection of the IMU; Arrows B and C identify the third and fourth axes of detection of IMU 6; Arrows E and F identify the fifth and sixth axes of detection of IMU 6 respectively. Although six axes of detection are defined, the apparatus 1 need not use all six if obtaining good data from a lower number. Accordingly, the apparatus may use any number of axes of detection, and may alter that according to different circumstances. According to this example, Arrow A, being the first axis of detection, provides good respiratory data capturing movement when a head rolls forwards and backwards during inhalation I exhalation.

[0213] Figure 4a also shows the defined detection axes with respect to a subject’s head, again represented by Arrows A through to F. Although just the first detection axis provides good data, further detection axes I further detection geometries may enhance the respiratory data, by taking additional sensor data into account when providing a plurality a respiratory data points. For instance, Arrow C, corresponding to the third axis of detection, detects a nodding movement up and / or down of the head during beathing. Further, Arrow E, corresponding to the fifth axis of detection, detects a forwards and backwards movement of the head during beathing

[0214] The apparatus 1 is supplied with a power source and appropriate circuitry (not shown).

[0215] By way of an alternative, it should be noted that processor 7 may send raw sensor data to the communicator 8, which may communicate that raw sensor data directly to an external processor 7’ (and associated algorithm) located in the external database I display apparatus 9. As such, processors 7; 7’ may be provided in both the ear apparatus 1 and the external database I display apparatus 9, providing analysis of the data through respective algorithms in one location, the other or both locations.

[0216] Figure 2 shows a wearable ear apparatus, generally identified by reference 21 . The ear apparatus 21 includes an ear-canal portion 22, for inserting into an earcanal of an individual, and an exterior ear portion 23, for resting in a pinna and / or concha region of an ear of the individual. The ear-canal portion 22 is intended to be positioned within an ear-canal (not shown) and the exterior ear portion 23 is intended to be positioned in a concha region (not shown) of a human or other animal test subject.

[0217] The exterior ear portion 23 includes a six-axis IMU 26, being located in the environs of the concha region and / or pinna region of the human subject. The IMU 6 is configured to have a physical orientation the same as the IMU of Figure 1 , having the same first to sixth detection axes I geometries, again represented by Arrows A to F. Accordingly, that information will not be repeated but is, of course relevant to this example. The exterior ear portion 23 also includes a processor 27, with an associated algorithm, for analysing sensor data received from the IMU 26 (and any other sensors). The processor 27 is for deriving a plurality of respiratory data points from the sensor data and determining change in inspiratory duration and / or expiratory duration and, thereby, change in respiratory duty cycle. The output of the algorithm is communicated, by communicator 28, to an external database I display apparatus 29, provided for storing data and outputting respiratory data as and when required.

[0218] 19

[0219] P24365WOOO The ear-canal portion 22 may, optionally, include one or more additional sensors 26’, being additional to the IMU 26.

[0220] The apparatus 21 is supplied with a power source and appropriate circuitry (not shown).

[0221] By way of an alternative, it should be noted that communicator 28 may communicate raw sensor data directly to an external processor 27’ located in the external database I display apparatus 29. In a further alternative, processors 27; 27’ may be provided in both the ear apparatus 21 and the external database I display apparatus 29.

[0222] In a variant of Figure 2, the exterior ear portion may be extended so as to locate the IMU behind an ear of the test subject. In a further variant, the exterior ear portion may be a standalone apparatus, having no physical connection to an earcanal portion, or have, indeed, no ear-canal portion. The IMU could be located simply near an ear of the test subject.

[0223] Figure 3 shows a wearable ear apparatus, generally identified by reference 31 . The apparatus 31 includes an ear-canal portion 32, being anatomically shaped to map an ear-canal of the subject. The apparatus 31 also includes an exterior ear portion 33, which includes a wing portion 33a, to aid securely fastening the ear apparatus 31 to a test subject.

[0224] An IMU 36 is located within the ear-canal portion 32, along a limb 32a thereof. Unlike Figures 1 and 2 where a physical orientation of the IMU is required, this example detects in all potential first to sixth axes, A to F’, and configuration involves comparing and / or analysing the data from the specific sensors of IMU 36 to determine which one, or combination, provides the best respiratory data. The limb 32a may, optionally, include one or more additional sensors 36’, being additional to the IMU 36.

[0225] Apparatus 31 includes a processor 37 and communicator 38. The processor 37 is intended to compare and / or analyse the data from the specific sensors of IMU 36 to determine which one, or combination, provides the best respiratory data, and the communicator 38 communicates such data to an external database I display means (not shown).

[0226] Alternatively, the communicator 38 may communicate raw sensor data to a database I display means (not shown) which, optionally, includes a processor or further processor (not shown), in a similar manner to that described above for Figures 1 and 2.

[0227] Operation and use of the ear apparatus 1 ; 21 ; 31 , according to any one of those three embodiments, is further described in the following example.

[0228] Example 1

[0229] In accordance with a first example of the invention, an ear apparatus 1 ; 21 ; 31 is located in an ear of a human test subject.

[0230] Readings were taken from the sensor means, which was configured to have detection axes A to F for accelerometer sensors and a gyroscope sensors of the IMU, to at least provide sensor data relating to detection of movement of the head of the test subject, as defined by those axes, whilst the test subject breathed in and out normally.

[0231] Respiratory data was created from the output of the sensor(s), which was processed and analysed to provide a plurality of respiratory data points, leading to a plot of respiratory movement 50 over time, as shown in Figure 5a. In Figure 5a, time is represented as the x-axis and the y-axis may be considered acceleration, although one may choose a different y-axis.

[0232] 20

[0233] P24365WOOO In Figure 5a, inspiratory data is, generally, shown above the time axis and expiratory data, generally, shown below the time axis. However, in reality, it is more complex than that, and Figure 5b is intended to assist understanding of the data provided. Figure 5b provides the same plot as Figure 5a; however, it is annotated (as per the lettering below) so as to define various regions or points in the plot corresponding to, at least, the following: a) onset of inhalation; b) transition to exhalation; c) exhalation end point; d) any inhalation I inspiratory pause; and / or e) preparatory pause.

[0234] Respiratory duty cycle is a calculation that shows the ratio of time spent breathing in, Ti, to the total time of a breath, Ttot (Ti plus time spent exhaling, Te), which is given by the ratio Ti / Ttot. As an individual’s breathing pattern changes, or as a temporary obstruction of an airway develops, the duty cycle will change as a result.

[0235] Detection of change in the duty cycle of the individual identifies changes in the inspiratory duration and expiratory duration, which can be used to detect changes in the control or state of asthma, or the onset of asthma exacerbation, commonly referred to as an asthma attack. This is not considered a method of treatment by diagnosis and, therefore, not considered excluded subject-matter, since change in duty cycle perse does not diagnose asthma, there being many extraneous factors to consider on top of any change in duty cycle.

[0236] Example 2

[0237] In accordance with a second example of the invention, an ear apparatus 1 ; 21 ; 31 is located in an ear of a human test subject.

[0238] Ear apparatus 1 ; 21 ; 31 includes a PPG sensor which detects PPG waveforms from the human test subject. The processor, and its associated algorithm, is configured to conduct signal averaging, specifically ensemble averaging, to remove, cancel or mute signal noise.

[0239] The processor analyses and process the respiratory data to combine, for example five, PPG waveforms from each respiratory cycle (i.e. a full cycle being the respiratory cycle and not a cardiac pulse cycle) by aligning them to a peak (for example peak inspiration gyroscopic acceleration or other identifiable feature) and averages them in intervals.

[0240] In a further example for a gyroscopic IMU signal, the peak chosen may be the peak of acceleration at the approximate mid-point of inspiration (the peak approximately mid-way between a) and b) in Figure 5b). However, it should be noted that a different peak or reference timing point may be chosen and, for PPG, this may include a point being at the transition between inspiration and expiration, or at the end of expiration.

[0241] From the respiratory data created by signal averaging, one can derive, at least, one or more or a combination of the following: onset of inhalation; transition to exhalation; exhalation end point; any inhalation I inspiratory pause; and / or any preparatory pause.

[0242] Accordingly, from the respiratory data obtained by any of the processes previously described, information relevant to the physiological dynamics of airflow for the individual may be derived. For example in 5b) demonstrating gyroscopic IMU data, it can be seen that peak acceleration for inspiration occurs approximately midpoint during and is symmetrical for inspiration; however, for expiration the peak acceleration is skewed to the earlier portion of expiration and is asymmetrical.

[0243] 21

[0244] P24365WOOO

Claims

Claims:1 .) An ear apparatus, locatable at, or at least partly within, an ear of a human or other animal, for detecting change in inspiratory duration and / or expiratory duration of said human or other animal, the apparatus comprises: one or more, or a plurality of, sensor means, configured to detect respiratory data, the respiratory data comprising: onset of inhalation; and transition to exhalation, and means for analysing and / or processing, or means for communicating for subsequent analysing and / or processing, the respiratory data, and deriving therefrom a plurality of respiratory data points for one or more, or a plurality of, respiratory cycles of said human or other animal from which change in inspiratory duration and / or expiratory duration is deriveable, wherein the means for analysing and / or processing is configured to identify a subset of respiratory data points surrounding onset of inhalation and surrounding transition to exhalation, for more accurately determining change in inspiratory duration and / or expiratory duration.2.) An ear apparatus as claimed in claim 1 , wherein the one or more, or a plurality of, sensor means, comprise: movement detection sensor(s), which are configured to detect respiratory data from movement of the ear apparatus and / or from movement of a head of said human or other animal; non-movement detection sensor(s), which do not rely upon movement for detection of respiratory data; and / or combinations of movement and non-movement sensor(s).3.) An ear apparatus as claimed in claim 1 or claim 2, wherein the one or more, or a plurality of, sensor means, are configured to detect movement of a / said head of said human or other animal comprising any one or more of the group comprising: a) rotation of said head around an axis approximately, or substantially, parallel to at least part of one or both ear-canals of said human or other animal; b) movement of said head along an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or c) movement of said head along an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.4.) An ear apparatus as claimed in any preceding claim, wherein the one or more, or a plurality of, sensor means, are configured to detect movement of a / said head of said human or other animal comprising any one or more of the group comprising: d) movement of said head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of said human or other animal. e) rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or f) rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.5.) An ear apparatus as claimed in any preceding claim, wherein configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to detect of movement comprising:22P24365WOOOrotation of said head around and / or movement of said head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of said human or other animal; movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.6.) An ear apparatus as claimed in any preceding claim, wherein the means for analysing and / or processing is configured to determine respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of a / said head comprising: rotation of said head around and / or movement of said head along an axis approximately, or substantially, parallel to at least part of one or both earcanals of said human or other animal; movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to a caudal-vertex axis of said human or other animal; and / or movement of said head along and / or rotation of said head around an axis approximately, or substantially, parallel to an anterior-posterior axis of said human or other animal.7.) An ear apparatus as claimed in any preceding claim, wherein configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to provide sensor data from detection of movement of a / said head: along and / or around an ear-canal axis (i.e. medial to lateral axis), or vice versa, with respect to said human or other animal; along and / or around a vertex-caudal axis (i.e. head to toe axis), or vice versa, with respect to said human or other animal; and / or along and / or around an anterior-posterior axis (i.e. front to back axis), or vice versa, with respect to said human or other animal.8.) An ear apparatus as claimed in any preceding claim, wherein the means for analysing and / or processing is configured to receive sensor data from: a) one or more axes of detection of an accelerometer sensor; and / or b) one or more axes of detection of a gyroscope sensor, and further configured to detect movement based upon one or more axes of detection having greater signal to noise ratio for respiratory, inspiratory, and / or expiratory timing data.9.) An ear apparatus as claimed in any preceding claim, wherein at least part of the ear apparatus comprises a sensor housing, locatable in or around an ear of said user, so as to provide one or more, or a plurality of, sensor means, in a condition where the one or more sensors means provide sensor data from detection of movement: a) along and / or around an axis of detection substantially parallel to an earcanal of said human or other animal; b) along and / or around a vertex-caudal axis, or vice versa, with respect to said human or other animal; and / or c) along and / or around an anterior-posterior axis, or vice versa, with respect to said human or other animal.23P24365WOOO10.) An ear apparatus as claimed in any preceding claim, wherein the one or more, or a plurality of, sensor means, is / are configured to detect respiratory data comprising: exhalation end point; any inhalation pause; and / or any exhalation pause, in the one or more respiratory cycles.11.) An ear apparatus as claimed in any preceding claim, wherein the means for analysing and / or processing is configured to: identify a plurality of respiratory data points: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause, for more accurately determining change in inspiratory duration and / or expiratory duration; and / or exclude respiratory data points being a predetermined time period outside of: before and after onset of inhalation; before and after transition to exhalation; before and after exhalation end point; and / or before and after any inhalation pause and / or any exhalation pause.12.) An ear apparatus as claimed in any preceding claim, wherein the means for analysing and / or processing is configured to: amalgamate a plurality of individual sensor readings to derive data at a single respiratory timing point; and / or align signals from different sensors, and / or different respiratory cycles, at a respiratory timing point and / or at a defined point in a / the respiratory cycle, enhancing the signal and, thereby, accuracy of data at a / the respiratory timing point, whilst averaging out the signal noise.13.) An ear apparatus as claimed in any preceding claim, wherein the one or more, or a plurality of, sensor means comprise one or combination of the following: a motion detector; an optical detector; an inertial measurement unit; a three- six-, or more, axis inertial measurement unit; a photoplethysmogram (PPG) sensor, being a contact or non-contact sensor; a ballistocardiogram (BCG) sensor; an electrode for electrocardiographic or electroenchephalographic monitoring; an ultrasound transducer; a radio frequency detector; a magnetometer; electromagnetic sensor means; capacitive sensor means; and / or a microphone.14.) An ear apparatus as claimed in any preceding claim, wherein the one or more, or a plurality of, sensor means is / are located in: an in-ear portion, shaped to be received in a concha and / or an ear-canal of said human or other animal;24P24365WOOOan outer-ear portion, shaped to be received in or on an outer pinna of said human or other animal; a behind the ear portion or over the ear portion, shaped to be received behind an ear of said human or other animal and / or over an ear of said human or other animal; and / or an associated apparatus, being located adjacent an ear of said human or other animal or in connection with wearable ear apparatus.15.) An ear apparatus as claimed in any preceding claim, wherein an IMU, or other sensor, detects onset of respiration, providing time of onset of respiration, or detects any other defined point, to allow synchronisation and summation I fusion of data from other sensors.16.) A method for detecting change in inspiratory duration and / or expiratory duration during one or more, or a plurality of, respiratory cycles of a human or other animal, the method comprising: locating an ear apparatus at, or at least partly within, an ear of the human or other animal, the ear apparatus comprising one or more, or a plurality of, sensor means; detecting respiratory data, using the one or more, or plurality of sensor means, to detect respiratory data comprising: onset of inhalation; and transition to exhalation, and analysing and / or processing, or communicating for subsequent analysing and / or processing, the respiratory data so as to derive therefrom a plurality of respiratory data points for the one or more, or plurality of, respiratory cycles and deriving from the plurality of respiratory data points change in inspiratory duration and / or expiratory duration of the human or other animal, wherein identifying a subset of respiratory data points surrounding onset of inhalation and surrounding transition to exhalation, to more accurately determine change in inspiratory duration and / or expiratory duration.17.) A method as claimed in claim 16 comprising signal fusion, signal averaging and / or ensemble averaging of PPG, audio, IMU and / or other sensor data encompassing respiratory data from the one or more, or a plurality of, sensor means.18.) A method as claimed in claim 16 or claim 17, wherein the one or more, or a plurality of, sensor means, detecting movement of a / the head of said human or other animal comprising any one or more of the group comprising: a) rotation of the head around an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal; b) movement of the head along an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or c) movement of the head along an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.19.) A method as claimed in any one of claims 16 to 18, wherein the one or more, or a plurality of, sensor means, detecting movement of a / the head of the human or other animal comprising any one or more of the group comprising: d) movement of the head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal;25P24365WOOOe) rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or f) rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal.20.) A method as claimed in any one of claims 16 to 19 comprising any one or combination of i) to iii) below: i) configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to detect of movement comprising: rotation of the head around and / or movement of the head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal; movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal; ii) configuring the means for analysing and / or processing determine respiratory data from an accelerometer sensor and / or a gyroscope sensor detecting movement of a / the head comprising: rotation of the head around and / or movement of the head along an axis approximately, or substantially, parallel to at least part of one or both ear-canals of the human or other animal; movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to a caudal-vertex axis of the human or other animal; and / or movement of the head along and / or rotation of the head around an axis approximately, or substantially, parallel to an anterior-posterior axis of the human or other animal; and / or iii) configuring a detection axis, or detection axes, of an accelerometer sensor and / or a gyroscope sensor to provide sensor data from detection of movement of a / the head: along and / or around an ear-canal axis, or vice versa, with respect to the human or other animal; along and / or around a vertex-caudal axis, or vice versa, with respect to the human or other animal; and / or along and / or around an anterior-posterior axis, or vice versa, with respect to the human or other animal.21 .) A method as claimed in any one of claims 16 to 20 comprising receiving sensor data from: d) one or more axes of detection of an accelerometer sensor; and / or e) one or more axes of detection of a gyroscope sensor, and detecting movement based upon one or more axes of detection having greater signal to noise ratio for respiratory, inspiratory, and / or expiratory timing data.22.) A method as claimed in any one of claims 16 to 21 comprising: analysing the curve of data points plotted against time to detect aspects including: area under the curve for inspiratory and / or expiratory phases; slope; and / or amplitude;26P24365WOOOfurther analysing comprising: comparison between inspiratory and expiratory phases; confluence, smoothness or other features of the relevant plotted curves; and / or analysis by machine learning and computer vision processes.23.) A method as claimed in any one of claims 16 to 22 comprising any one or combination of the following i) to x): i) comparing analytic outputs between: various inspiratory and expiratory periods; time periods including hours or days for an individual; and / or the individual and data from larger populations, for example of healthy volunteers of the same or similar age; ii) amalgamating a plurality of individual sensor readings to derive data at a single respiratory timing point; iii) aligning signals from different sensors, and / or different respiratory cycles, at a respiratory timing point and / or at a defined point in a / the respiratory cycle, enhancing the signal and, thereby, accuracy of data at a / the respiratory timing point, whilst averaging out the signal noise; iv) enhancing the signal and, thereby, accuracy of onset of inhalation, transition to exhalation, exhalation end point, any inhalation pause, and / or any exhalation pause; v) comparing independent accelerometer, gyroscope and / or magnetometer channels so as to dynamically choose the channel providing the highest quality respiratory data; vi) comparing independent accelerometer, gyroscope and / or magnetometer channels so as to dynamically choose the channel providing the highest quality respiratory data; vii) dynamically choosing which channel of a / the inertial measurement unit provides the highest quality respiratory data; viii) fusing and / or averaging respiratory cycle data to increase the number of data points; ix) aligning several respiratory waveforms to align the shape to determine the average; and / or x) proactively selecting data from a multitude of sensors dependent on quality of signal and combining to determine a respiratory pattern suitable to detect inspiratory duration and / or expiratory duration.

24. A method as claimed in any one of claims 16 to 23 comprising detecting respiratory data comprising: detecting respiratory data from movement of the ear apparatus and / or from movement of a head of said human or other animal; detecting respiratory data without, or which does not rely upon, movement for detection; and / or detecting data from combinations of movement and non-movement sensor(s).

25. A method as claimed in any one of claims 16 to 24 comprising detecting respiratory data comprising: exhalation end point; any inhalation pause; and / or any exhalation pause, in the one or more respiratory cycles.27P24365WOOO

Images