The invention relates to a physiological monitoring device and method, for monitoring physiological signals of a subject in conjunction with monitoring motion of the subject.

Various physiological monitoring devices which are wearable by a subject and which monitor physiological signals are known, for example these may monitor ECG and/or respiration signals of the subject. ECG signals may be monitored to determine if an arrhythmia of the subject's heart has occurred. This involves analysis of the ECG signals, and signal to noise ratio is an issue particularly in the mobile environment. If the signal to noise ratio is low, then the analysis of the ECG signals may lead to detection of false arrhythmias. Similarly, respiration signals may be monitored to determine if the respiration rate of the subject is within an acceptable range. This involves analysis of the respiration signals, and signal to noise ratio is again an issue particularly when the subject is mobile. If the signal to noise ratio is low, then the analysis of the respiration signals may lead to detection of false respiration rates. Various sources of noise in ECG and respiration signafs are known, for example noise caused by motion of the subject. To enhance analysis of ECG and respiration signals it is desirable to monitor sources of noise, e.g. subject movement.

According to a first aspect of the invention there is provided a physiological monitoring device adapted to be worn by a subject, comprising

one or more electrodes adapted to be placed on the subject and to detect physiological signals from the subject,

a physiological signal detector which receives the physiological signals from the electrodes,

a motion detector which detects motion of the subject,

a microprocessor which receives physiological signals from the physiological signal detector and motion signals from the motion detector, and correlates the motion signals with analysis of the physiological signals, such that at any point of time when motion of the subject is equal to or greater than a motion threshold, analysis of the physiological signals is disregarded. The motion detector may comprise an accelerometer. The accelerometer may be a single- axis accelerometer. The accelerometer may be a multi-axis accelerometer. The accelerometer may be a three-axis accelerometer which measures the acceleration of movement of the subject along three orthogonal axes. The accelerometer may be configured to detect only accelerations, and therefore subject movement, within a predetermined range. The accelerometer may provide information concerning its, and therefore the subject's, orientation e.g. upright, prone.

The microprocessor may comprise embedded motion signal analysis software which it uses to analyse motion signals received from the motion detector.

When the motion detector comprises an accelerometer the microprocessor may analyse acceleration signals from the accelerometer. When the motion detector comprises a multi- axis accelerometer the microprocessor may analyse acceleration signals along each axis.

The microprocessor may analyse the acceleration signals along each axis separately. For each axis, the microprocessor may analyse the acceleration signals along the axis by comparing a measured acceleration signal with a previously-measured acceleration signal to determine a change in acceleration along the axis. For each axis, the microprocessor may analyse the acceleration signals along the axis by comparing a measured acceleration signal with an immediatly-preceding, previously-measured acceleration signal to determine a change in acceleration along the axis. The motion threshold of the subject may be determined as an acceleration change threshold for each axis. The motion threshold of the subject may be determined as an acceleration change threshold of typically O.lg for a sagittal axis, typically O.lg for a transverse axis and typically 0.6g for a frontal axis.

The microprocessor may sum the acceleration signals along each axis and analyse the summed acceleration signals. The microprocessor may analyse the summed acceleration signals by comparing a summed acceleration signal with a previous summed acceleration signal to determine a change in summed acceleration. The motion threshold of the subject may be determined as a summed acceleration change threshold. The motion threshold of the subject may be determined as a summed acceleration change threshold above lg that is considered acceptable for physiological signal detection and reporting.

The motion threshold may be determined from a study of motion and physiological signal detection in one or more normal subjects. Physiological signals and motion signals may be collected for the or each subject as the subject undertakes various activities, each with a different type/level of movement. The activities may comprise, for example, resting in various positions e.g. sitting, standing, lying on back, front, left and right sides, walking on the level and on stairs, running at various rates, cycling, getting dressed, cooking, housework. For the or each subject and each activity, the physiological signals and motion signals are analysed to determine the motion threshold, i.e. a threshold level for the motion signals at or above which it is decided that noise in the physiological signals due to motion of the subject is unacceptable.

The physiological signals detected from the subject may comprise ECG signals detected from the subject's heart.

An ECG motion threshold may be determined from a study of motion and arrhythmia detection in one or more normal subjects i.e. with normal heart function and therefore no arrhythmias. ECG signals and acceleration signals may be collected for the or each subject as the subject undertakes various activities, each with a different type/level of movement. For the or each subject and each activity, the ECG signals and motion signals are analysed to determine the ECG motion threshold, i.e. a threshold level for the motion signals at or above which it is decided that noise in the ECG signals due to motion of the subject is too great to be confident that analysis of the ECG signals will detect a real arrhythmia.

An ECG motion threshold may be determined by analysis of the ECG signals to measure the level of noise in these signals, and setting a value of the ECG motion threshold where the noise level of the ECG signals is considered to be unacceptable.

The microprocessor may comprise embedded ECG analysis software which analyses the ECG signals. The microprocessor may analyse the ECG signals to monitor physiological parameters such as the subject's heart beat rate and ECG features such as P wave shape, QRS width. The microprocessor may use monitoring of these parameters to determine if the subject experiences a heart arrhythmia. The microprocessor may record data concerning an arrhythmia such as the time that it occurs, the duration of the arrhythmia and the shape of the arrhythmia and can output data concerning the arrhythmia. The microprocessor may receive motion signals and correlate these with the analysis of the ECG signals. At points of time when the subject's motion is equal to or greater than the ECG motion threshold, analysis of arrhythmia in the ECG signals may be disregarded, e.g. the ECG analysis software may cease to output arrhythmia data.

The physiological signals detected from the subject may comprise respiration signals detected from the subject.

A respiration motion threshold may be determined from a study of motion and respiration in one or more subjects having normal respiration. Respiration signals and acceleration signals may be collected for the or each subject as the subject undertakes various activities, each with a different type/level of movement. For the or each subject and each activity, the respiration signals and motion signals are analysed to determine the respiration motion threshold, i.e. a threshold level for the motion signals at or above which it is decided that noise in the respiration signals due to motion of the subject is too great to be confident that analysis of the respiration signals will result in a true measure of the subject's respiration.

A respiration motion threshold may be determined by analysis of the respiration signals to measure the level of noise in these signals, and setting a value of the respiration motion threshold where the noise level of the respiration signals is considered to be unacceptable.

The microprocessor may comprise embedded respiration analysis software which analyses the respiration signals. The microprocessor may analyse the respiration signals to monitor physiological parameters such as the subject's respiration rate. The microprocessor may use monitoring of these parameters to determine if the subject experiences abnormal respiration. The microprocessor may record data concerning abnormal respiration such as the time that it occurs, the duration, and can output data concerning the abnormal respiration. The microprocessor may receive motion signals and correlate these with the analysis of the respiration signals. At points of time when the subject's motion is equal to or greater than the respiration motion threshold, analysis of the respiration signals may be disregarded, e.g. the respiration analysis software may cease to output respiration data.

The microprocessor may comprise an analogue to digital converter (ADC). The microprocessor may receive analogue physiological signals and use the ADC to convert these to digital physiological signals for analysis.

The electrodes may be any known electrodes which, when placed on a subject, are able to detect ECG signals from the subject's heart. An ECG signal comprising a difference between the voltage signal between first and second electrodes may be fed to the ECG signal detector. An ECG signal comprising a difference between the voltage signal between first and third electrodes may be fed to the ECG signal detector. These two signals may be used in the assessment of physiological parameters of the subject.

The electrodes may be any known electrodes which, when placed on a subject, are able to detect respiration signals from the subject. The electrodes may measure fluctuations in impedance signals from the subject's lungs during breathing. The impedance signals may be used in the assessment of the respiration of the subject.

The physiological monitoring device may further comprise a temperature sensor for measuring the temperature of the subject's skin.

According to a second aspect of the invention there is provided a method of monitoring physiological signals of a subject in conjunction with monitoring motion of the subject comprising

using one or more electrodes to detect physiological signals from the subject, using a motion detector to detect motion of the subject,

using a microprocessor to receive the physiological signals and motion signals and correlate the motion signals with analysis of the physiological signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of the physiological signals is disregarded. β

According to a third aspect of the invention there is provided a physiological monitoring device adapted to be worn by a subject, comprising

one or more electrodes adapted to be placed on the subject and to detect ECG signals from the subject's heart,

an ECG signal detector which receives the ECG signals from the electrodes, a motion detector which detects motion of the subject,

a microprocessor which receives ECG signals from the ECG signal detector and motion signals from the motion detector, and correlates the motion signals with analysis of the ECG signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of arrhythmia in the ECG signals is disregarded.

According to a fourth aspect of the invention there is provided a method of monitoring ECG signals of a subject in conjunction with monitoring motion of the subject comprising

using one or more electrodes to detect ECG signals from the subject's heart, using a motion detector to detect motion of the subject,

using a microprocessor to receive the ECG signals and motion signals and correlate the motion signals with analysis of the ECG signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of arrhythmia in the ECG signals is disregarded.

In this way, detection of 'false' arrhythmias may be reduced.

According to a fifth aspect of the invention there is provided a physiological monitoring device adapted to be worn by a subject, comprising

one or more electrodes adapted to be placed on the subject and to detect respiration signals from the subject,

a respiration signal detector which receives the respiration signals from the electrodes,

a motion detector which detects motion of the subject,

a microprocessor which receives respiration signals from the respiration signal detector and motion signals from the motion detector, and correlates the motion signals with analysis of the respiration signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of the respiration signals is disregarded.

According to a sixth aspect of the invention there is provided a method of monitoring respiration signals of a subject in conjunction with monitoring motion of the subject comprising

using one or more electrodes to detect respiration signals from the subject, using a motion detector to detect motion of the subject,

using a microprocessor to receive the respiration signals and motion signals and correlate the motion signals with analysis of the respiration signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of the respiration signals is disregarded.

Embodiments of the invention will now be described by way of example only, with reference to the accompanying drawings, in which

Figure 1 is a schematic representation of a physiological monitoring device according to the first aspect of the invention, and

Figure 2 is a flowchart representing the method of the second aspect of the invention.

Referring to Figure 1, the physiological monitoring device 1 comprises three electrodes 3a, 3b, 3c for measuring ECG signals, two electrodes 3d, 3e for measuring respiration signals, a temperature sensor 5, an ECG signal detector 7, a respiration detector 9, a motion sensor 11 a microprocessor 13 and a transceiver 19.

In this embodiment, the electrodes 3a to 3e and the temperature sensor 5 are housed in a flexible patch (not shown) which is attached to the subject by an adhesive layer. The flexible patch is made of a biocompatible material. The ECG signal detector 7, the respiration detector 9, the motion sensor 11 and the microprocessor 13 are contained within a device (not shown) which removably attaches to the flexible patch by clips. Five clips are provided on the device which attach to corresponding clips on the flexible patch. a

When connected, three of the clips each provide a signal from one of electrodes 3a, 3b, 3c in the flexible patch to the ECG signal detector 7 in the device, and the remaining two clips each provide a signal from one of electrodes 3d, 3e in the flexible patch to the respiration signal detector 9 in the device. It will be appreciated that other arrangements of the components of the physiological monitoring device may be employed. The device further contains a battery (not shown) for providing power to the other components contained in the device. The device comprises charger connections (not shown) for connection of the device battery to a charger. The device further comprises a battery power monitoring device (not shown). When the battery power approaches a preset level, a warning light provided on the device (not shown) is lit to indicate to the subject that the battery of the device should be charged.

The electrodes 3a, 3b, 3c may comprise any known electrode which, when placed on a subject, are able to detect ECG signals from the subject's heart. The electrodes may, for example, be placed on the subject's torso or on the subject's chest. The first electrode 3a is closest to the subject's left leg, a second electrode 3b is closest to the subject's right arm and a third electrode 3c is closest to the subject's left arm. It will be appreciated, however, that other electrode positions may be used. Each of the electrodes 3a, 3b, 3c is connected by a lead and clips to the ECG signal detector 7. An ECG signal comprising a difference in voltage between the signal of electrode 3b and the signal of electrode 3a is fed to the ECG signal detector 7. An ECG signal comprising a difference in voltage between the signal of electrode 3c and the signal of electrode 3a is fed to the ECG signal detector 7. These two signals are used in the assessment of the ECG physiological parameters of the subject. A signal is also sent to the ECG signal detector 7 from electrode 3a, this signal is used as a reference signal. In this embodiment of the invention the physiological monitoring device comprises three electrodes for ECG measurement, but it will be appreciated that other embodiments may comprise more or less electrodes.

The electrodes 3d, 3e may comprise any known electrode which, when placed on a subject, are able to detect respiration signals from the subject. The electrodes may, for example, be placed on the subject's torso or on the subject's chest. A known sinusoida l signal at a specific frequency (57kHz) is injected into the electrodes 3d and 3e, and impedance changes caused by fluctuations of the volume of air in the subject's lungs are measured by the electrodes. The impedance signals are fed by a lead to the respiration signal detector 9.

The temperature sensor 5 comprises a thermistor. This measures resistance signals of the subject's skin and passes these signals via a lead to the microprocessor 13 where they are used as an indication of the temperature of the subject's skin.

The ECG signal detector 7 comprises analogue hardware and is able to receive continuous signals from the electrodes 3a, 3b, 3c. In this embodiment the ECG signal detector 7 comprises a noise filter 15, having a passband of approximately 0.5 Hz to 25 Hz. The noise filter 15 attempts to filter noise from the signals received from the electrodes caused by, for example, mobile phone interference or movement by the subject. The efficacy of the noise filter 15 is limited as it has to have a certain passband to achieve adequate detection of the ECG signals. The size of the passband, however, allows some motion artefacts to be detected which results in noise in the ECG signals. The ECG signal detector 7 is connected to the microprocessor 13 and sends filtered ECG signals to the microprocessor 13.

The respiration signal detector 9 also comprises analogue hardware and receives impedance signals from the electrodes 3d and 3e. Measurement of the impedance signals is highly susceptible to motion artefacts, and the respiration signal detector 9 comprises a noise filter. The noise filter attempts to filter noise due to motion from the impedance signals received from the electrodes. The respiration signal detector 9 is connected to the microprocessor 13 and sends filtered respiration signals to the microprocessor 13.

The motion detector 11 of the physiological monitoring device 1 comprises a three-axis accelerometer for measurement of three axis motion and orientation of the subject. The accelerometer comprises a digital accelerometer micro electro mechanical system and measures the acceleration of motion of the subject along three orthogonal axes. The accelerometer may be configured to detect only accelerations, and therefore subject movement, within a pre-determined range. The accelerometer is connected to the microprocessor 13. Signals comprising the acceleration along each axis are sent to the microprocessor 13. The microprocessor 13 comprises an analogue to digital converter (ADC) 17. This receives the filtered ECG signals from the ECG signal detector 5 and the filtered respiration signals from the respiration signal detector 7 and samples these signals, e.g. at a rate of 360 samples per second to create digital ECG signals and at a rate of 120 samples per second to create digital respiration signals.

The microprocessor 13 comprises embedded acceleration analysis software which it uses to analyse the three acceleration signals from the motion detector 11. The acceleration analysis software analyses the acceleration signals received from each axis of the accelerometer separately. For each axis, the acceleration analysis software analyses the acceleration signals along the axis by comparing a measured acceleration signal with a previously-measured acceleration signal to determine a change in acceleration along the axis. The acceleration analysis software sets a motion threshold of the subject as an acceleration change threshold for each axis, for example O.lg for a sagittal axis, O.lg for a transverse axis and 0.6g for a frontal axis.

The microprocessor 13 comprises embedded ECG analysis software and uses this to analyse the ECG signals received from the ECG signal detector 7. The ECG analysis software periodically, e.g. every 15 minutes, outputs details of the subject's heart rate to the transceiver 19. The ECG analysis software is remotely configurable to allow a clinician to set the periodic reporting frequency appropriate for the subject. The respiration analysis software is also remotely configurable to allow a clinician to set a heart rate limit which is deemed acceptable for the subject. The ECG analysis software analyses the ECG signals to monitor the subject's heart beat rate. If this exceeds the set heart rate limit, the microprocessor 13 outputs a heart rate alarm and data concerning the heart rate to the transceiver 19. The ECG analysis software further analyses the ECG signals to monitor ECG features such as P wave shape, QRS width. The ECG analysis software periodically outputs such details of the subject's ECG signals to the transceiver 19. The ECG analysis software uses monitoring of these ECG parameters to determine if the subject experiences a heart arrhythmia, for example any of asystole, ventricular fibrillation, ventricular tachycardia, bradycardia, tachycardia, atrial fibrillation, supra-ventricular fibrillation. ECG arrhythmia detection and analysis is based on the Pan-Tompkins method and uses analysis of the slope, amplitude and width of QRS complexes to detect an arrhythmia. The arrhythmia detection and analysis algorithm used includes a series of steps which include lowpass and highpass filtering, derivative, squaring, and integration functions, and adaptive thresholding and search procedures. If an arrhythmia occurs the ECG analysis software records data concerning the arrhythmia such as the time that it occurs, the duration of the arrhythmia and the shape of the arrhythmia and outputs an arrhythmia alarm and data concerning the arrhythmia to the transceiver 19. During analysis of the ECG signals, the embedded ECG analysis software receives the measured change of acceleration for each axis of the accelerometer from the acceleration analysis software and correlates this with the analysis of the ECG signals. At points of time when the measured change of acceleration signal along an axis is equal to or greater than the acceleration change threshold (motion threshold) for that axis, and for a certain time (e.g. 30 seconds) thereafter, the ECG analysis software ceases to output information, particularly arrhythmia data.

The microprocessor 13 comprises embedded respiration analysis software and uses this to analyse the respiration signals received from the respiration signal detector 9. The respiration analysis software analyses the impedance signals from the detector 9 to calculate an average respiration rate over a set period of time, e.g. 30 seconds. The respiration analysis software periodically, e.g. every 15 minutes, outputs details of the subject's respiration rate to the transceiver 19. The respiration analysis software is remotely configurable to allow a clinician to set the periodic reporting frequency appropriate for the subject. The respiration analysis software is also remotely configurable to allow a clinician to set a respiration rate limit which is deemed acceptable for the subject. The respiration analysis software analyses the respiration signals to monitor the subject's respiration rate. If this exceeds the set respiration rate limit, the microprocessor 13 outputs a respiration rate alarm and data concerning the respiration rate to the transceiver 19. During analysis of the respiration signals, the embedded respiration analysis software receives the measured change of acceleration for each axis of the accelerometer from the acceleration analysis software and correlates this with the analysis of the respiration signals. At points of time when the measured change of acceleration signal along an axis is equal to or greater than the acceleration change threshold (i.e. motion threshold) for that axis, and for a certain time (e.g. 30 seconds) thereafter, the respiration analysis software ceases to output data.

The transceiver 19 outputs data from the physiological monitoring device 1 by a wired or wireless (e.g. WiFi) communication link. The transceiver 19 is able to detect if it is out of range of wireless coverage and generate an alarm for the subject.

The physiological monitoring device 1 may comprise a visual and/or audio warning device to indicate various alarms to the subject or a third party (e.g. a nurse), e.g. that electrodes/leads have become disconnected, the power level of the battery is low, that the transceiver is out of wireless range, that abnormal heart or respiration rate has occurred, that an arrhythmia has occurred. The transceiver 19 is also be used to receive commands from the subject's clinician via the server.

The output of the physiological monitoring device 1 is received, for example, by a server (not shown) used by a medical practitioner e.g. a clinician monitoring the condition of the subject. The server provides access to the physiological monitoring device 1 by clinicians and other medical staff. Configuration data for the physiological monitoring device, such as heart and respiration limits, is stored in the server and input to the physiological monitoring device 1 from the server. The server receives information from the physiological monitoring device 1, stores this data and displays this information for use by a clinician. The server will display various alarms to the clinician, e.g. that electrodes/leads have become disconnected, the power level of the battery is low, that the transceiver is out of wireless range, that abnormal heart or respiration rate has occurred, that an arrhythmia has occurred. The server may also collate the received information and provide trends in physiological data of the subject to the clinician. The server may also perform further analysis of the subject physiological data. The server may comprise a web browser for the input and output of data.

In the embodiment of the invention described above, components of the device are connected to each other by leads. It will be appreciated, however, that the one or more of the components can be connected to one or more other components by means of a wireless connection. The motion threshold may be determined by a number of methods, as described above.

In an alternative embodiment of the physiological device of the invention, the motion threshold may not be pre-set but may be specifically determined for the subject using the device by any of the methods detailed above.

It will be appreciated that physiological monitoring devices according to the invention may further comprise components for monitoring other physiological parameters such as subject blood oxygen levels.

Referring to Figure 2, the method of monitoring physiological signals of a subject in conjunction with monitoring motion of the subject comprises using one or more electrodes to detect physiological signals from the subject, using a motion detector to detect motion of the subject, and using a microprocessor to receive the physiological signals and motion signals from the motion detector, and correlate the motion signals with analysis of the physiological signals, such that at any point of time when the motion of the subject is equal to or greater than a motion threshold, analysis of the physiological signals is disregarded.

The motion t reshold for use in the method may be determined by any of the methods detailed above.