Phonocardiogram sensing device

Abstract

A phonocardiogram (PCG) sensing device is disclosed. The phonocardiogram sensing device comprises a body audio sensing device, and the body audio sensing device comprises a body audio sensing transducer. The body audio sensing transducer is arranged to sense animal body audio when the body audio sensing device is disposed adjacent the animal body at the body audio sensing location, and an acoustic path of the body audio is defined between the animal body and the body audio sensing transducer. The PCG sensing device further comprises an environment audio sensing device. The ambient audio sensing device comprises an ambient audio sensing transducer arranged to sense ambient audio present in an environment in the vicinity of the phonocardiogram sensing device when the body audio sensing device is disposed in the body audio sensing position. The body audio sensing device generates a body audio signal indicative of animal body audio, and the ambient audio sensing device generates an ambient audio signal indicative of ambient audio, the ambient audio signal being used to increase a signal-to-noise ratio of the body audio signal.

Description

Technical Field

[0001] The present invention relates to a phonocardiogram sensing device, a monitoring system for monitoring body-generated data, and a wearable component for a monitoring system. Background Art

[0002] Cardiovascular disease (CVD) is the leading cause of death worldwide. CVD usually develops gradually, generally over many years, with very small changes occurring each year, and symptoms may not become noticeable until years into the disease.

[0003] The first step in evaluating a patient for the presence of CVD in their heart involves listening to the heart sounds – called cardiac auscultation – because these sounds provide important initial clues in evaluating the patient and guide further diagnostic testing.

[0004] Heart sounds are produced by the opening and closing of the heart valves and the flow of blood through the heart and adjacent blood vessels. The main normal heart sounds, known as S1 and S2, are the sounds that correspond to the closing of specific heart valves. Abnormalities in the S1 and S2 sounds can indicate a heart abnormality. In addition, the presence of S3 sounds can indicate abnormalities caused by disease, while S4 sounds can indicate a pathological condition.

[0005] However, heart sounds are difficult to interpret by clinicians, especially those who are not experienced in cardiac auscultation. Although experienced clinicians can determine some cardiac conditions based on the intensity, frequency, location, and timing of the cardiac cycle, this has significant limitations because successful detection of abnormalities depends on the clinician's skill and the frequencies of many of the relevant heart sounds are below the human hearing range.

[0006] Based on the patient's examination, the clinician decides whether further testing is needed. If risk factors or indicators identified during the examination suggest that a heart-related abnormality may be present, an electrocardiogram (ECG) may be done to obtain more information about heart function.

[0007] An electrocardiogram (ECG) records the electrical activity associated with blood flow through the heart, lungs, and other organs during the cardiac cycle. Using electrodes placed on the skin, small electrical signals due to the depolarization and repolarization of the heart muscle can be detected. During the cardiac cycle, a normal ECG pattern consists of several parts - a P wave representing atrial depolarization, a QRS complex representing ventricular depolarization, and a T wave representing ventricular repolarization.

[0008] A healthy heart produces a generally consistent and repeatable pattern during the cardiac cycle, and ECG recordings provide information about heart rate and rhythm, including information that may indicate cardiac abnormalities such as arrhythmias, heart enlargement due to hypertension, and myocardial infarction.

[0009] Other diseases can also be detected by auscultation. For example, the sounds produced by the lungs can indicate lung abnormalities, including diseases such as pneumonia, emphysema, asthma, bronchitis, and cancer. Such sounds can include:

[0010] Rhonchi – low-pitched crackles that may be caused by bronchi containing fluid or mucus;

[0011] Wheezing – a high-pitched whistling sound that indicates a partial airway obstruction; and

[0012] ·Wheezes – high-pitched turbulent sounds that generally indicate upper airway obstruction or narrowing.

[0013] However, like cardiac auscultation, pulmonary auscultation has significant limitations, as successful detection of abnormalities requires hearing the relevant indicator sounds, and accurate diagnosis depends on the skill of the clinician. Summary of the invention

[0014] According to a first aspect of the present invention, a phonocardiogram (PCG) sensing device is provided, the device comprising: a body audio sensing device, the body audio sensing device comprising a body audio sensing transducer, the body audio sensing transducer being arranged to sense animal body audio when the body audio sensing device is set at a body audio sensing position adjacent to the animal body, wherein when the body audio sensing device is set at the body audio sensing position, the acoustic path of the body audio is limited between the animal body and the body audio sensing transducer; and an environmental audio sensing device, the environmental audio sensing device comprising an environmental audio sensing transducer, the environmental audio sensing transducer being arranged to sense environmental audio present in the environment near the phonocardiogram sensing device when the body audio sensing device is set at the body audio sensing position; the body audio sensing device generates a body audio signal for indicating the animal body audio, and the environmental audio sensing device generates an environmental audio signal for indicating the environmental audio, and the environmental audio signal can be used to increase the signal-to-noise ratio of the body audio signal.

[0015] In one embodiment, the PCG sensing device includes a diaphragm that defines an audio cavity when the body audio sensing device is positioned adjacent to the animal's body, the diaphragm enhancing body sounds emanating from the animal's body.

[0016] In one embodiment, the body audio sensing device is connected to the ambient audio sensing device using a vibration damping connection.The vibration damping connection can include a resilient O-ring received in opposing circumferential grooves when the body audio sensing device is connected to the ambient audio sensing device.

[0017] In one embodiment, the body audio sensing transducer comprises a MEMS microphone.

[0018] In one embodiment, the body audio sensing arrangement comprises a PCG filtering component arranged to filter the body audio signal generated by the body audio sensing transducer.

[0019] In one embodiment, the body audio sensing device comprises a PCG amplification component arranged to amplify the body audio signal generated by the body audio sensing transducer. The PCG amplification component can have an associated gain of approximately five.

[0020] In one embodiment, the ambient audio sensing transducer comprises a MEMS microphone.

[0021] In one embodiment, the ambient audio sensing device comprises an ambient filtering component arranged to filter the ambient audio signal generated by the ambient audio sensing transducer.

[0022] In one embodiment, the ambient audio sensing device comprises an ambient amplification component arranged to amplify the ambient audio signal generated by the ambient audio sensing transducer. The ambient amplification component can have an associated gain of approximately 20.

[0023] According to a second aspect of the present invention, there is provided a wearable garment comprising at least one PCG sensing device according to the first aspect of the present invention.

[0024] In one embodiment, the wearable garment includes a plurality of PCG sensing devices disposed at selected locations to optimize acquisition of relevant body audio signals.

[0025] In one embodiment, the multiple PCG sensing devices include a front multiple PCG devices, and the front multiple PCG devices include a PCG device that is set on the opposite side of the sternum of the animal's body and aligned with the pulmonary artery when in use, a PCG device that is set on the opposite side of the sternum of the animal's body and aligned with the tricuspid valve when in use, a PCG device that is set on the mid-clavicular line adjacent to the mitral valve area when in use, and a PCG device that is set on the mid-axillary line adjacent to the mid-axillary area when in use.

[0026] In one embodiment, the multiple PCG sensing devices include a plurality of posterior PCG devices, including a PCG device disposed on opposite sides of a vertical centerline directly below a shoulder blade of the animal's body, a PCG device disposed on opposite sides of the vertical centerline at a middle portion of the back of the animal's body, and a PCG device disposed on opposite sides of the vertical centerline adjacent to a lower lobe of a lung of the animal's body.

[0027] In one embodiment, the plurality of PCG sensing devices includes at least one cervical PCG device disposed at a neck region of the patient to sense audio frequencies from a carotid artery.

[0028] In one embodiment, the wearable garment includes a plurality of ECG electrodes disposed at selected locations to optimize acquisition of relevant body electrical signals.

[0029] In one embodiment, the plurality of ECG electrodes includes a RAECG electrode disposed on a right side of an atrium of the animal's body and a LAECG electrode disposed on a left side of the atrium.

[0030] In one embodiment, the plurality of ECG electrodes includes RLD ECG electrodes disposed vertically at a location below the heart of the animal's body and aligned with the RA ECG electrodes.

[0031] In one embodiment, the wearable garment includes a right leg drive (RLD) amplifier, which uses signals obtained from the RA electrode and the LA electrode to drive a feedback electrical signal into the animal's body at the RLD electrode, and the feedback signal is used to improve common mode rejection by canceling the common mode signal that would otherwise exist in the RA electrode and the LA electrode.

[0032] In one embodiment, the wearable garment includes at least one ultrasonic sensor.

[0033] In one embodiment, information from the at least one ultrasonic sensor is used to acquire respiratory and cardiac cycle data.

[0034] In one embodiment, the at least one ultrasound sensor is disposed on the wearable device such that, in use, the at least one ultrasound sensor is disposed adjacent to the patient's heart or chest.

[0035] In one embodiment, the wearable garment includes at least one photoplethysmography (PPG) sensor.

[0036] In one embodiment, data from the at least one photoplethysmography (PPG) sensor is used to monitor changes in blood volume and blood oxygenation.

[0037] In one embodiment, data from the at least one photoplethysmography (PPG) sensor is used to derive physiological parameters associated with heart rate variability, blood pressure, ankle-brachial pressure, cardiovascular disease, aging, neurological disorders, lung disease, and / or respiratory rate.

[0038] In one embodiment, data acquired from the at least one ultrasound sensor and data derived from the at least one PPG sensor are used to detect the presence of fluid in the lungs of the patient.

[0039] In one embodiment, the wearable garment comprises a data acquisition device arranged to synchronously receive signals from the PCG device and the ECG electrodes.

[0040] In one embodiment, the data acquisition device is arranged to synchronously receive signals from at least one ultrasound device, the PCG device, and the ECG electrodes.

[0041] In one embodiment, the data acquisition device is arranged to synchronously receive signals from at least one PPG device, the at least one ultrasound device, the PCG device, and the ECG electrodes.

[0042] In one embodiment, the data acquisition device comprises a PCG signal filtering stage arranged to filter the PCG signal received from the PCG device.

[0043] In one embodiment, the PCG signal filtering stage comprises an AC coupling filter arranged to remove a DC offset voltage in the PCG signal.

[0044] In one embodiment, the data acquisition device includes an instrumentation amplifier that receives ECG signals from the RA ECG electrode and the LA ECG electrode and amplifies a difference between the RA ECG signal and the LA ECG signal.

[0045] In one embodiment, the data acquisition device comprises a DC blocking amplifier arranged to remove a DC component in the amplified signal produced by the instrumentation amplifier.

[0046] In one embodiment, the data acquisition device comprises a notch filter arranged to suppress disturbances of approximately 50 Hz.

[0047] In one embodiment, the data acquisition device comprises an A / D converter arranged to convert analog PCG signals and analog ECG signals into digital PCG signals and digital ECG signals.The A / D converter can be a sigma-delta A / D converter.

[0048] In one embodiment, the PCG signal filtering stage comprises an anti-aliasing filter arranged to ensure that substantially no aliasing occurs in the A / D converter.

[0049] In one embodiment, the wearable garment comprises a memory arranged to store data indicative of a PCG signal and an ECG signal.

[0050] In one embodiment, the wearable garment comprises a wireless transceiver arranged to facilitate transmission of data indicative of the PCG signal and the ECG signal from the wearable device.

[0051] According to a third aspect of the present invention, a system for monitoring body-generated data is provided, the system comprising: a wearable device according to the second aspect of the present invention; a noise cancellation component, the noise cancellation component being arranged to remove environmentally-related noise from a body audio signal using an environmental audio signal, thereby generating a noise-processed body audio signal, the noise-processed body audio signal having an improved signal-to-noise ratio; and a feature extractor, the feature extractor being arranged to extract a plurality of features from the noise-processed body audio signal and generate at least one feature vector comprising the plurality of features; and a data analysis component, the data analysis component being arranged to predict at least one medical abnormality associated with an animal's body using the at least one feature vector.

[0052] In one embodiment, the data analysis component uses at least one machine learning component that is trained to learn the relationship between the feature vectors and the medical diagnosis.

[0053] In one embodiment, the extracted features are associated with an ECG signal, and the R peak, P peak, Q point, S point, and T peak of the ECG signal are used to derive the extracted features, and the time intervals between R peaks, the difference between consecutive R peak time intervals, and / or the time between P peaks and R peaks are used to derive the extracted features.

[0054] The acquisition of the extracted features associated with the ECG signal can use: frequency domain characteristics, which can include mean, variance, and higher-order moments; skewness; kurtosis; and / or time domain and time-frequency domain characteristics, which can be acquired using wavelets or wavelet packets (WP) and / or synchronized wavelet packets (SWP) of the ECG signal.

[0055] In one embodiment, the extracted features associated with the PCG signal are derived using the S1 , systolic, S2, and diastolic states of the PCG cycle.

[0056] In one embodiment, the extracted features associated with the PCG signal are obtained by using time domain characteristics, frequency domain characteristics, energy characteristics, entropy characteristics and / or kurtosis characteristics.

[0057] In one embodiment, the extracted features include features corresponding to changes in frequency power of a specific frequency band of a specific segment of the cardiac cycle.

[0058] In one embodiment, the machine learning component includes a classifier, which can be implemented using a k-nearest neighbor algorithm (KNN), a support vector machine (SVM), and / or an artificial neural network (ANN) which can include a convolutional neural network (CNN).

[0059] According to a fourth aspect of the present invention, a method for monitoring medical abnormalities in an animal's body is provided, the method comprising: providing a plurality of phonocardiogram sensing devices, each phonocardiogram sensing device comprising: a body audio sensing device, the body audio sensing device comprising a body audio sensing transducer, the body audio sensing transducer being arranged to sense the animal's body audio when the body audio sensing device is adjacent to the animal's body at a body audio sensing position, wherein when the body audio sensing device is set at the body audio sensing position, the acoustic path of the body audio is limited between the animal's body and the body audio sensing transducer; and an environmental audio sensing device, the environmental audio sensing device comprising an environmental audio sensing transducer, the environmental audio sensing transducer being arranged to sense the animal's body audio when the body audio sensing device is adjacent to the animal's body at the body audio sensing position. When the audio sensing position is set, the ambient audio present in the environment near the phonocardiogram sensing device is sensed; the body audio sensing device is set at a limited position on the animal's body; the body audio sensing device is used to generate a body audio signal indicating the animal's body audio and an ambient audio signal indicating the ambient audio present in the environment near each body audio sensing device; the ambient audio signal is used to remove environment-related noise from the body audio signal to generate a noise-processed body audio signal, and the noise-processed body audio signal has an improved signal-to-noise ratio; a plurality of features are extracted from the noise-processed body audio signal and at least one feature vector including the plurality of features is generated; and the at least one feature vector is used to predict at least one medical abnormality associated with the animal's body. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] The present invention will now be described by way of example only with reference to the accompanying drawings, in which:

[0061] Figure 1 is a schematic perspective view of a phonocardiographic apparatus according to an embodiment of the present invention;

[0062] Figure 2 yes Figure 1 A schematic perspective view of the diaphragm of the phonocardiographic device is shown;

[0063] Figure 3 yes Figure 1 A schematic cross-sectional view of a phonocardiographic device is shown;

[0064] Figure 4 yes Figure 1 A schematic cross-sectional view of a vibration-reducing connection of a phonocardiographic device is shown;

[0065] Figure 5 yes Figure 3 A schematic cross-sectional view of a portion of the phonocardiographic device is shown;

[0066] Figure 6 is a block diagram of the functional components of a phonocardiogram device;

[0067] Figure 7 is a block diagram of functional components of a monitoring system according to an embodiment of the present invention;

[0068] Figure 8 is a diagram of the front portion of a wearable garment according to an embodiment of the present invention, the wearable garment comprising a plurality of Figures 1 to 6 The phonocardiogram device and the plurality of ECG electrodes are shown;

[0069] Fig. 9 yes Figure 8 a diagram of a rear portion of the wearable garment shown;

[0070] Fig.10 yes Figure 8 and Fig. 9 an illustration of a neck portion of the wearable garment as shown;

[0071] Fig.11 yes Figure 7 A block diagram of the components of the filter stage of the monitoring system shown;

[0072] Fig.12 yes Figure 7 A block diagram of components of a signal conditioner of a monitoring system is shown;

[0073] Fig.13 yes Figure 7 A block diagram of a data processor filtering stage of a monitoring system shown;

[0074] Fig.14 is a flow chart showing the steps of a method for monitoring an animal's body according to an embodiment of the present invention;

[0075] Fig.15 are graphs of phonocardiographic signals of normal subjects and subjects with cardiovascular disease (CVD); and

[0076] Fig.16 is a block diagram of functional components of an alternative monitoring system according to another embodiment of the present invention. Specific embodiments

[0077] Although the present embodiments are described with respect to the human patient body, it should be understood that the present invention is equally applicable to other animal bodies.

[0078] It should also be understood that although the current embodiments are described with respect to cardiac and pulmonary abnormalities, the present invention is equally applicable to other medical abnormalities, the important thing being that the medical abnormalities are at least related to, and detectable based on, the audio signals generated by the body.

[0079] With reference to the accompanying drawings, Figure 1 and Figure 2 An exemplary phonocardiogram (PCG) apparatus 10 is shown in accordance with an embodiment of the present invention.

[0080] The PCG device 10 includes a body audio sensing device 12 and an ambient audio sensing device 14 connected to the body audio sensing device 12. When in use, the PCG device 10 is placed relative to the patient so that the body audio sensing device 12 contacts the patient's skin. Figure 2 As shown, the body audio sensing device 12 includes a diaphragm 16, such as formed of a flexible plastic material, which defines an audio cavity when the PCG device 10 is placed adjacent to the patient's skin in use. The diaphragm 16 is used to enhance body sounds emanating from the patient.

[0081] like Figure 3 As shown more specifically in FIG. 1 , in this example, the body audio sensing device 12 is connected to the ambient audio sensing device 14 using a vibration-damping connection 18. In one example, as Figure 4 As shown, the vibration-reducing connection 18 includes an elastic O-ring 20 that is received in opposing circumferential grooves 21 when the body audio sensing device 12 is connected to the ambient audio sensing device 14. Thus, vibration transfer between the body audio sensing device 12 and the ambient audio sensing device 14 is reduced.

[0082] The body audio sensing device 12 includes a body audio sensing assembly 22 attached to a body audio wall 24, such as using an adhesive layer 26. A body audio hole 28 is formed in the body audio wall 24 so that, when in use, audio associated with the patient's body can be transmitted from the cavity defined by the diaphragm 16 to the body audio sensing assembly 22. In this example, the diaphragm 16 is provided with an internal threaded portion 17 to facilitate securely connecting the diaphragm 16 to the body audio wall 24.

[0083] Similarly, ambient audio sensing device 14 includes ambient audio sensing assembly 30 attached to ambient audio wall 32, such as using adhesive layer 34. Ambient audio aperture 36 is formed in ambient audio wall 32 to enable ambient audio to pass to ambient audio sensing assembly 30 from the surrounding environment.

[0084] Figure 5 A portion of the phonocardiographic device 12 is shown, particularly showing component 21 of the body audio sensing component 22 and component 23 of the ambient audio sensing component 30 .

[0085] As shown, the body audio sensing assembly 22 includes a body audio sensing transducer 38 and an audio processing device 40 mounted to a circuit board 42. In this example, the body audio sensing transducer 38 is a micro-electromechanical system (MEMS) microphone, but it is understood that any suitable electromechanical transducer is contemplated. The circuit board 42 is provided with a hole 39 aligned with the body audio hole 28 and the sensing portion of the body audio sensing transducer 38.

[0086] Similarly, the ambient audio sensing assembly 30 includes an ambient audio sensing transducer 41 and an audio processing device 43 mounted to a circuit board 45. In this example, the ambient audio sensing transducer 41 is a micro-electromechanical system (MEMS) microphone, but it will be appreciated that any suitable electromechanical transducer is contemplated. The circuit board 45 is provided with a hole 47 aligned with the ambient audio hole 36 and the sensing portion of the ambient audio sensing transducer 41.

[0087] In this example, the circuit boards 42 , 45 of the body audio sensing assembly 22 and the ambient audio sensing assembly 30 are electrically connected together using a flexible ribbon connector 49 , thereby providing a common communication path 51 for the body audio sensing assembly 22 and the ambient audio sensing assembly 30 .

[0088] Figure 6 An example functional component 50 of the PCG device 10 is shown. The functional component 50 includes a body audio functional component 51 and an ambient audio functional component 53.

[0089] The body audio functional component 51 comprises a body audio sensing transducer 38, which in this example is a MEMS microphone, and a high pass filter 54, which is arranged to filter out low frequency signals from the transducer signal generated by the MEMS microphone 38. In this example, the high pass filter 54 is a single pole RC high pass filter with a 3dB cutoff at 8.5Hz, but it will be appreciated that any other suitable high pass filter is conceivable.

[0090] The body audio functional component 51 also includes a pair of single-pole common-mode low-pass filters 56, 58, which are arranged at the respective inputs of a low-pass differential filter 60, which is used to minimize the amplification of common-mode signals. The output of the low-pass differential filter 60 is connected to the input of a low-noise differential instrument amplifier 62, which provides a high common-mode rejection ratio and a very high differential input impedance. The common-mode low-pass filters 56, 58 are used to reduce RF rectification in the low-noise instrument amplifier 62, and in this example, each common-mode low-pass filter 56, 58 is arranged to provide a 3dB cutoff at 40kHz. In this example, the low-pass differential filter 60 is arranged to provide a 3db cutoff at 1.9kHz. However, it is understood that any other suitable common-mode low-pass filter is conceivable, and any other suitable low-pass differential filter is conceivable. In this example, the low-noise instrument amplifier 62 provides a gain of approximately 5, but it is understood that any other suitable gain is conceivable.

[0091] The ambient audio functional component 53 of the ambient audio sensing device 14 is similar to the body audio functional component 51 of the body audio sensing device 12 in that the ambient audio functional component 53 includes an ambient audio sensing transducer 64, which is a MEMS microphone in this example, a high pass filter 66 having a configuration similar to the high pass filter 54, a pair of single-pole common mode low pass filters 68, 70 having a configuration similar to the pair of single-pole common mode low pass filters 56, 58, a low pass differential filter 72 having a configuration similar to the low pass differential filter 60, and a low noise instrument amplifier 74 having a configuration similar to the low noise instrument amplifier 62, except that the gain of the instrument amplifier 74 is approximately 20 instead of 5.

[0092] The body audio sensing device 12 generates a body audio signal 63 indicating the sensed body audio, and the ambient audio sensing device 14 generates an ambient audio signal 76 indicating the sensed ambient audio. The ambient audio signal 76 is used to remove the environment-related noise from the body audio signal 63, thereby increasing the signal-to-noise ratio (SNR) of the body audio signal 63.

[0093] Figure 7 An example monitoring system 80 for monitoring body generated signals, particularly body generated audio signals and body generated electrical signals, is shown. In this embodiment, the monitoring system 80 includes a wearable garment 82 in the form of a vest in this example, a number of PCG devices 10, which are 13 PCG devices 10 in this example, and a number of ECG electrodes, which are right atrium (RA) electrodes 86, left atrium (LA) electrodes 88, and right leg drive (RLD) electrodes 90 in this example.

[0094] In the present example, the monitoring system 80 is arranged to monitor audio and electrical signals associated with the heart and audio signals associated with the lungs, but it will be appreciated that other types of body-generated signals may additionally or alternatively be monitored.

[0095] The PCG device 10 and ECG electrodes 86, 88, 90 are mounted on the vest 82 at selected locations to optimize the collection of relevant body-generated signals. Figures 8 to 10 An example wearable garment 82 is shown in FIG. It should be appreciated that the vest 82 ensures accurate positioning of the PCG device 10 and ECG electrodes 86, 88, 90 relative to the patient's body, and also ensures that there is uniform, consistent pressure between the patient's body and the PCG device 10 and ECG electrodes 86, 88, 90.

[0096] Figure 8 The front of the vest 82 is shown, indicating the locations of the PCG devices 10 and ECG electrodes 86, 88, 90. In this example, 6 PCG devices are positioned on the front, with 2 PCG devices 10 positioned on opposite sides of the patient's sternum and aligned with the patient's pulmonary artery, 2 PCG devices 10 positioned on opposite sides of the patient's sternum and aligned with the patient's tricuspid valve, 1 PCG device 10 positioned on the midclavicular line adjacent to the mitral valve region, and 1 PCG device 10 positioned on the midaxillary line adjacent to the midaxillary region.

[0097] In this example, the RA ECG electrode 86 is disposed on the right side of the patient's atrium and the LA ECG electrode 88 is disposed on the left side of the patient's atrium. The RLD ECG electrode 90 is disposed vertically at a position well below the patient's heart, aligned with the RA ECG electrode 86.

[0098] Fig. 9 The rear portion of the vest 82 is shown, indicating the locations of the PCG devices 10. In this example, six PCG devices 10 are disposed at the rear portion, with two PCG devices 10 disposed on opposite sides of the vertical centerline just below the patient's shoulder blades, two PCG devices 10 disposed on opposite sides of the vertical centerline at the mid-portion of the patient's back, and two PCG devices 10 disposed on opposite sides of the vertical centerline adjacent to the patient's lower lung lobe. The six PCG devices 10 disposed at the rear portion of the vest 82 are arranged to monitor sounds produced by the patient's lungs.

[0099] like Fig.10 As shown, the vest 82 also includes a cervical PCG device 10 disposed at the patient's cervical region. The cervical PCG device 10 is disposed at a location where audio from the carotid artery can be sensed.

[0100] The PCG device 10 and ECG electrodes 86, 88, 90 together provide body audio and body electrical signals that can be used to detect and characterize body abnormalities, which in this example are cardiac and pulmonary abnormalities.

[0101] like Figure 7 As shown, in this example, the vest 82 also includes a data acquisition device 92 that synchronously receives signals from the PCG device 10 and the ECG electrodes 86, 88, 90. Synchronously receiving the data enables the data processing component to map temporal and spatial dependencies.

[0102] The data acquisition device 92 includes a filter stage 94, a signal conditioner 96, and an analog to digital (A / D) converter 98. A rechargeable power storage device (not shown) may also be provided for supplying power to the operating components of the vest 82.

[0103] The filter stage 94 receives data from the PCG device 10. The functional components of the filter stage 94 are Fig.11 Shown in.

[0104] The filter stage 94 includes an AC coupling filter 124 arranged to receive the body audio signal 122 generated by the PCG device 10 and remove any DC offset voltage that may have been generated by the instrumentation amplifier 62. In the present example, the AC coupling filter 124 is a high pass single pole RC filter with a 3 dB cutoff at 34 Hz, but it will be appreciated that suitable variations are contemplated.

[0105] The filter stage 94 also includes an anti-aliasing filter 126 which ensures that no significant aliasing occurs in the A / D converter 98. In the present example, the anti-aliasing filter 126 is a low pass single pole filter with a 3dB cut-off at 8kHz, but it will be appreciated that any suitable variation is contemplated.

[0106] In the present example, the A / D converter 98 is a delta-sigma 24-bit type A / D converter which samples all input signals simultaneously at a clock rate of approximately 8 MHz, and includes a digital decimation filter arranged to operate as a linear phase, finite impulse response third order low pass filter with a 3 dB cutoff at 4 kHz, but it will be appreciated that any suitable variation is contemplated.

[0107] The signal conditioner 96 receives data from the ECG electrodes 86, 88, 90. The functional components of the signal conditioner 96 are Fig.12 Shown in.

[0108] The signal conditioner 96 includes an instrumentation amplifier 130 that receives ECG signals from the RA and LA ECG electrodes 86, 88 and amplifies the difference between the ECG signals. In this example, the instrumentation amplifier 130 is a high gain differential instrumentation amplifier having a gain of approximately 100, but it will be appreciated that any suitable variation is contemplated.

[0109] The signal conditioner 96 also includes a DC blocking amplifier 132 arranged to remove a DC component in the amplified signal produced by the instrumentation amplifier 130 and a notch filter 134 arranged in this example to suppress interference of approximately 50 Hz.

[0110] The signal conditioner 96 also includes a right leg drive (RLD) amplifier 136 that extracts a common mode signal from the signals associated with the RA and LA electrodes 86, 88, inverts the common mode signal, and uses the inverted signal to drive a feedback electrical signal into the patient at the RLD electrode 90. Thus, common mode rejection is improved because the injected signal cancels out the common mode signal that would otherwise be present in the RA and LA electrodes 86, 88.

[0111] The digitized signals indicative of the sensed body audio and electrical signals generated by the A / D converter 98 are stored in the memory 100 and transmitted from the memory 100 or transmitted in substantially real time to a data processing device 104 using a wireless transceiver 102. In this example, the digitized signals are transmitted wirelessly using a Bluetooth or WiFi protocol, but it will be appreciated that any suitable communication protocol may be used.

[0112] The data processing device 104 performs further signal processing, noise removal processing, and analysis of the body audio and electrical signals to extract features that will be used to determine if the patient has any abnormalities, which in this example are heart and lung abnormalities. The data processing device 104 is also used to display the results of the audio and electrical signal analysis to the user.

[0113] In this example, the data processing apparatus 104 comprises a data processor 106 and a display driver 108 arranged to drive a display 109. In this example, the data processing apparatus 104 is implemented using any suitable computing device, which may be a desktop computer, a laptop, a tablet computer or a smartphone.

[0114] The data processing device 104 also includes suitable software to implement a user interface for the monitoring system 80, in particular to enable a user to control the operation of the monitoring system 80, including capturing audio and electrical body signals through the PCG device 10 and the ECG electrodes 86, 88, 90, initiating data analysis of the captured signals, viewing and analyzing diagnostic results generated by the data analyzer 112, and analyzing historical results including disease progression.

[0115] The functional components 138 of the data processor 106 are Fig.13 As shown, the functional components 138 include a noise canceller 140 that uses the ambient audio signal 76 to remove the noise signal from the body audio signal 63 to improve the signal-to-noise ratio (SNR) of the body audio signal.

[0116] The PCG device 10 and associated monitoring system 80 are capable of detecting audio over a wide frequency range, which in this example is 10 Hz to 5 kHz, and it will be appreciated that by improving the SNR of the body audio signal 63, the operating frequency range of the PCG monitoring system 80 can be achieved.

[0117] The inventors recognize that there are more than 200 different features for detecting various cardiac conditions, but it is impractical to use all of these features to determine whether a medical abnormality exists. To this end, the functional component 138 also includes a feature extractor 142, which is arranged to extract features from the processed PCG signal and ECG signal and produce a feature vector that can be used as an input to a machine learning algorithm, which is trained to use the feature vector to predict patient abnormalities.

[0118] The functional component 138 further comprises an error determiner 144 arranged to remove PCG signals and / or ECG signals considered to be abnormal.

[0119] The feature vector may include features derived from one or both of the ECG signal and the PCG signal simultaneously.

[0120] This arrangement enables each feature vector to be mapped to a specific medical abnormality, such as a specific type of cardiovascular disease (CVD), and features included in each feature vector are selected using a statistical gain ranking method, such as the information gain ratio (IGR). Features used to predict patient abnormalities are typically based on repeatable patterns in the ECG signal and PCG signal associated with the heart rate.

[0121] Example features associated with an ECG signal are derived from R peaks, P peaks, Q points, S points, and T peaks, and may include time intervals between R peaks, differences between consecutive R peak time intervals, and the time between P peaks and R peaks.

[0122] The following characteristics can be used to characterize the R, P, Q, S and T related features:

[0123] Frequency domain characteristics obtained using the Fast Fourier Transform (FFT) algorithm, such as mean, variance, higher-order moments, skewness, kurtosis, and spectral entropy, which indicates spectral kurtosis; and

[0124] Time domain and time-frequency domain features, including wavelets or wavelet packets (WP) and synchronized wavelet packets (SWP).

[0125] Wavelets can be derived from the R peak of the ECG signal, and various statistical energy measures can be calculated based on the wavelet coefficients associated with the shape of the R wave. WP or SWP are similar, except that they are associated with the complete ECG signal. Typically, the fourth level of the wavelet packet tree corresponding to 16 coefficients is used. Statistical measures similar to FFT features can be used on these 16 coefficients.

[0126] Features associated with PCG signals are usually related to the PCG cycle, which can be divided into four states: S1, systole, S2, and diastole. After segmenting the PCG signal, time domain features, frequency domain features, energy features, entropy features, and kurtosis are defined.

[0127] In one specific example, it has been observed that changes in frequency power of specific frequency bands at specific segments of the cardiac cycle can be indicative of coronary artery disease (CAD). Thus, by using appropriate bandpass filters between 3 and 9 frequency bands, a set of features that can be used to distinguish between non-CAD subjects and CAD subjects can be obtained.

[0128] In particular, the inventors realized that a frequency band between 20 Hz-30 Hz can be used to differentiate between non-CAD subjects and CAD subjects.

[0129] The frequency band between 250Hz-1000Hz is also useful, and the current arrangement uses the ambient audio sensing component 30 to reduce noise in the PCG signal so that this frequency band can be used to obtain relevant features indicating patient abnormalities. It can be appreciated that by suppressing external noise, higher frequency heart-related sounds can be enhanced.

[0130] However, alternative methods for determining features from the PCG signal are contemplated. For example, the complete PCG signal may be analyzed by analyzing wavelets / wavelet packets (WP) and / or synchronized wavelet packets (SWP) associated with the PCG signal.

[0131] Table 1 below shows example features that may be used in feature vectors that may be obtained from body PCG signals and ECG signals. However, it will be appreciated that other features and feature vectors may be used.

[0132]

[0133]

[0134] Table 1

[0135] Fig.15 Example waveforms of PCG signals acquired from four locations on a patient are shown, from a normal subject 170 and a subject 172 with coronary artery disease (CAD) - a) pulmonary location, b) tricuspid valve location, c) mitral valve location, and d) mid-axillary location.

[0136] Example medical anomalies that may be associated with one or more feature vectors include:

[0137] heart

[0138] Valvular disease;

[0139] Heart failure

[0140] Cardiac arrhythmias, including atrial fibrillation, atrial flutter, supraventricular tachycardia, ventricular fibrillation, and ventricular tachycardia;

[0141] lung

[0142] air and fluid in the lungs caused by infectious and noninfectious diseases;

[0143] Changes in chest structure;

[0144] pneumonia;

[0145] Pleural effusion;

[0146] pneumothorax;

[0147] Chronic obstructive pulmonary disease (COPD);

[0148] asthma

[0149] The feature vectors generated by the feature extractor 142 are transmitted to the data analyzer 112 which, in this example, is located remotely from the data processing device 104 and is accessed via a wide area network such as the Internet 110 .

[0150] The data analyzer includes a machine learning component 114 arranged to use the input feature vector to generate a prediction about a medical abnormality and a data storage device 116 for storing the feature vector and the diagnosis prediction. Storing the feature vector and the diagnosis prediction allows the disease progression to be monitored over time. The data storage device 116 may also be arranged to store processed audio and body generated electrical signals, such as the audio and body generated electrical signals received by the data processing device 104 from the vest 82.

[0151] The machine learning component may include any suitable machine learning arrangement that may be hardware and / or software based, and the machine learning component may be of any suitable type, including a classifier implemented using a k-nearest neighbor algorithm (KNN), a support vector machine (SVM), and / or an artificial neural network (ANN), which may include a convolutional neural network (CNN).

[0152] The machine learning component may be trained in any suitable manner, such as using supervised learning to establish a relationship between a defined input feature vector and a medical abnormality.

[0153] Fig.14 A method 150 for monitoring an animal's body according to an embodiment of the present invention is shown. The method 150 uses Figure 7 Monitoring system 80 is shown.

[0154] In use, a user, such as a clinician, first places the vest 82 on a patient and uses the data processing device 104, which in this example is dedicated software implemented on the data processing device 104, to synchronously capture body audio and ambient audio signals using the PCG device, and to synchronously capture ECG signals using the ECG electrodes, as shown in steps 152 and 154. As shown in step 156, the processed ambient audio signal is then used to remove noise associated with the environment from the body audio signal to improve the signal-to-noise ratio of the body audio signal. Then, as shown in step 160, the captured signal is filtered and conditioned to reduce noise and prepare the signal for analysis. As shown in step 162, features relevant to diagnosing a medical condition are extracted from the processed noise-reduced body audio and body electrical signals, and the extracted features are then used to generate a feature vector, which is used as an input to a machine learning component that is trained to recognize the relationship between the feature vector and a medical abnormality.

[0155] It will be appreciated that the present invention facilitates early diagnosis of medical abnormalities, particularly cardiac and pulmonary diseases, and can identify patients suspected of having CVD and / or pulmonary disease without relying on the expertise of a clinician.

[0156] It will also be appreciated that the use of multiple body sensors, and in particular multiple PCG sensing devices, significantly enhances the depth of data available for diagnosing medical abnormalities, particularly cardiac and pulmonary abnormalities, and therefore the diagnostic granularity possible with the present invention is greater than corresponding diagnostic arrangements known to date.

[0157] It will also be appreciated that the present invention can obtain a wider range of results than conventional auscultation by a clinician due to the frequency range covered being wider than the typical human hearing range. In particular, the present invention is arranged to analyse relatively low frequencies that are relevant to the diagnosis of specific medical abnormalities but are inaudible to humans.

[0158] It will also be appreciated that because the results generated by the monitoring system are progressively stored, the progression of the disease can be monitored over time.

[0159] The functional components of the monitoring system 180 according to another embodiment of the present invention are Fig.16 The same and similar features are indicated by the same reference numerals.

[0160] Monitoring system 160 and Figure 7 The monitoring system 80 shown is similar in that a plurality of PCG devices 10 and a plurality of ECG electrodes 86, 88, 90 are included in a wearable garment 182. In this embodiment, in addition to the PCG devices 10 and the ECG electrodes 86, 88, 90, the monitoring system 180 also includes at least one ultrasonic sensor 184, which is also incorporated into the wearable garment 182 in this example, and at least one photoplethysmography (PPG) sensor 186, which is also incorporated into the wearable garment 182 in this example.

[0161] The ultrasonic sensor 184 provides information that can be used to obtain respiratory and cardiac cycle data. For example, the ultrasonic sensor 184 can be placed adjacent to the patient's heart or chest when in use, and simultaneously evaluate information from the PPG and PCG sensors 10, 186 and the ECG electrodes 86, 88, 90 to obtain an indication of the left ventricular ejection fraction (LVEF) by constructing a feature vector from multiple information sources and applying trained machine learning to the feature vector. LFEV data can be used to obtain an indication of heart failure or cardiomyopathy. The ejection fraction is a measure of the amount of blood in the heart compared to the amount of blood pumped out of the heart, so the ejection fraction is used to indicate the efficiency of the heart pumping blood to the body.

[0162] A photoplethysmography (PPG) sensor 186 may be used to monitor changes in blood volume and blood oxygenation, and thus using the PPG sensor 186, pulsatile arterial blood flow may be monitored and an associated photoplethysmographic waveform may be obtained, which may be used to derive physiological parameters including heart rate variability, blood pressure, ankle-brachial pressure, cardiovascular disease, aging, neurological disorders, lung disease, and respiratory rate. The PPG sensor 186 may be of a type arranged to measure the amount of light absorbed or reflected by tissue.

[0163] The data acquired from the ultrasonic sensor 184 may also be combined with information derived from the PPG sensor 186 to detect the presence of fluid in the lungs.

[0164] In the following claims and the foregoing description of the invention, unless the context requires otherwise due to explicit language or necessary implication, the word "comprise" or variations such as "include" or "contain" are used in an inclusive sense, i.e. specifying the presence of features but not excluding the presence or addition of further features in various embodiments of the invention.

[0165] It should be understood that, if any prior art publication is referred to herein, this reference does not constitute an admission that the publication forms part of the common general knowledge in the art, in Australia or any other country.

[0166] Modifications and variations obvious to those skilled in the art are deemed to be within the scope of the present invention.

Claims

1. A phonocardiogram (PCG) sensing device, comprising: a body audio sensing device, the body audio sensing device comprising a body audio sensing transducer arranged to sense animal body audio when the body audio sensing device is disposed adjacent to an animal body at a body audio sensing location, wherein when the body audio sensing device is disposed at the body audio sensing location, an acoustic path for body audio is defined between the animal body and the body audio sensing transducer; and an ambient audio sensing device, the ambient audio sensing device comprising an ambient audio sensing transducer arranged to sense ambient audio present in an environment proximate to the phonocardiogram sensing device when the body audio sensing device is disposed at the body audio sensing location; The body audio sensing device generates a body audio signal indicative of animal body audio, and the ambient audio sensing device generates an ambient audio signal indicative of the ambient audio, the ambient audio signal being usable to increase a signal-to-noise ratio of the body audio signal.

2. The PCG sensing device of claim 1, comprising a diaphragm defining an audio cavity when the body audio sensing device is positioned adjacent to the animal's body, the diaphragm enhancing body sounds emanating from the animal's body.

3. The PCG sensing device according to claim 1 or 2, wherein: The body audio sensing device is connected to the ambient audio sensing device using a vibration damping connection.

4. The PCG sensing device according to claim 3, wherein: The vibration damping connection includes a resilient O-ring received in opposing circumferential grooves when the body audio sensing device is connected to the ambient audio sensing device.

5. A PCG sensing device according to any one of the preceding claims, wherein: The body audio sensing transducer comprises a MEMS microphone.

6. A PCG sensing device according to any one of the preceding claims, wherein: The body audio sensing device comprises a PCG filtering component arranged to filter a body audio signal generated by the body audio sensing transducer.

7. A PCG sensing device according to any one of the preceding claims, wherein: The body audio sensing device comprises a PCG amplification component arranged to amplify a body audio signal generated by the body audio sensing transducer.

8. The PCG sensing device according to claim 7, wherein: The PCG amplification component has an associated gain of approximately 5.

9. A PCG sensing device according to any one of the preceding claims, wherein: The ambient audio sensing transducer comprises a MEMS microphone.

10. A PCG sensing device according to any one of the preceding claims, wherein: The ambient audio sensing device comprises an ambient filtering component arranged to filter an ambient audio signal generated by the ambient audio sensing transducer.

11. A PCG sensing device according to any one of the preceding claims, wherein: The ambient audio sensing device comprises an ambient amplification component arranged to amplify an ambient audio signal generated by the ambient audio sensing transducer.

12. The PCG sensing device according to claim 11, wherein: The ambient amplification component has an associated gain of approximately 20.

13. A wearable garment comprising at least one PCG sensing device as claimed in any one of the preceding claims.

14. The wearable garment of claim 13, wherein: The wearable garment includes a plurality of PCG sensing devices disposed at selected locations to optimize acquisition of relevant body audio signals.

15. The wearable garment of claim 14, wherein: The multiple PCG sensing devices include a front multiple PCG devices, and the front multiple PCG devices include a PCG device that is set on the opposite side of the sternum of the animal's body and aligned with the pulmonary artery when in use, a PCG device that is set on the opposite side of the sternum of the animal's body and aligned with the tricuspid valve when in use, a PCG device that is set on the mid-clavicular line adjacent to the mitral valve area when in use, and a PCG device that is set on the mid-axillary line adjacent to the mid-axillary area when in use.

16. The wearable garment according to claim 14 or 15, wherein: The multiple PCG sensing devices include multiple rear PCG devices, including PCG devices arranged on opposite sides of the vertical centerline directly below the shoulder blades of the animal's body, PCG devices arranged on opposite sides of the vertical centerline at the middle part of the back of the animal's body, and PCG devices arranged on opposite sides of the vertical centerline adjacent to the lower lobe of the lung of the animal's body.

17. A wearable garment according to any one of claims 14 to 16, wherein: The plurality of PCG sensing devices include at least one cervical PCG device disposed at a neck region of the patient to sense audio frequencies from a carotid artery.

18. A wearable garment according to any one of claims 13 to 17, wherein: The wearable garment includes a plurality of ECG electrodes disposed at selected locations to optimize acquisition of relevant body electrical signals.

19. The wearable garment of claim 18, wherein: The plurality of ECG electrodes include a RA ECG electrode disposed on a right side of an atrium of the animal's body and a LA ECG electrode disposed on a left side of the atrium.

20. The wearable garment of claim 19, wherein: The plurality of ECG electrodes include RLD ECG electrodes disposed vertically at a location below the heart of the animal's body and aligned with the RA ECG electrodes.

21. The wearable garment of claim 20, wherein: The wearable garment includes a right leg drive (RLD) amplifier, which uses signals obtained from the RA electrode and the LA electrode to drive a feedback electrical signal into the animal's body at the RLD electrode, and the feedback signal is used to improve common-mode rejection by canceling the common-mode signal that would otherwise exist in the RA electrode and the LA electrode.

22. A wearable garment according to any one of claims 13 to 21, comprising at least one ultrasonic sensor.

23. The wearable garment of claim 22, wherein: Information from the at least one ultrasonic sensor is used to acquire respiratory and cardiac cycle data.

24. A wearable garment according to claim 22 or 23, wherein: The at least one ultrasound sensor is disposed on the wearable device such that, when in use, the at least one ultrasound sensor is disposed adjacent to the patient's heart or chest.

25. The wearable garment of any one of claims 13 to 24, comprising at least one photoplethysmography (PPG) sensor.

26. The wearable garment of claim 25, wherein: Data from the at least one photoplethysmography (PPG) sensor is used to monitor changes in blood volume and blood oxygenation.

27. The wearable garment of claim 26, wherein: Data from the at least one photoplethysmography (PPG) sensor is used to derive physiological parameters associated with heart rate variability, blood pressure, ankle-brachial pressure, cardiovascular disease, aging, neurological disorders, lung disease, and / or respiratory rate.

28. A wearable garment according to any one of claims 25 to 27 when dependent on claim 22, wherein: Data acquired from the at least one ultrasound sensor and data derived from the at least one PPG sensor are used to detect the presence of fluid in the lungs of the patient.

29. A wearable garment according to any one of claims 18 to 28, wherein: The wearable garment comprises a data acquisition device arranged to synchronously receive signals from the PCG device and the ECG electrodes.

30. The wearable garment of claim 29 when dependent upon claim 22, wherein: The data acquisition device is arranged to synchronously receive signals from at least one ultrasound device, the PCG device, and the ECG electrodes.

31. A wearable garment according to claim 30 when dependent on claim 25, wherein: The data acquisition device is arranged to synchronously receive signals from at least one PPG device, the at least one ultrasound device, the PCG device, and the ECG electrodes.

32. A wearable garment according to any one of claims 29 to 31, wherein: The data acquisition device comprises a PCG signal filtering stage arranged to filter a PCG signal received from the PCG device.

33. The wearable garment of claim 32, wherein: The PCG signal filtering stage comprises an AC coupling filter arranged to remove a DC offset voltage in the PCG signal.

34. A wearable garment according to claim 32 or 33 when dependent on claim 19, wherein: The data acquisition device includes an instrumentation amplifier that receives ECG signals from the RA ECG electrode and the LA ECG electrode and amplifies a difference between the RA ECG signal and the LA ECG signal.

35. The wearable garment of claim 34, wherein: The data acquisition device includes a DC blocking amplifier arranged to remove a DC component in an amplified signal produced by the instrumentation amplifier.

36. A wearable garment according to any one of claims 29 to 35, wherein: The data acquisition device comprises a notch filter arranged to suppress disturbances of approximately 50 Hz.

37. A wearable garment according to any one of claims 29 to 36, wherein: The data acquisition device comprises an A / D converter arranged to convert analog PCG signals and analog ECG signals into digital PCG signals and digital ECG signals.

38. The wearable garment of claim 37, wherein: The A / D converter is a Σ-Δ A / D converter.

39. The wearable garment of claim 32, wherein: The PCG signal filtering stage comprises an anti-aliasing filter arranged to ensure that substantially no aliasing occurs in the A / D converter.

40. A wearable garment according to any one of claims 13 to 39, wherein: The wearable garment comprises a memory arranged to store data indicative of a PCG signal and an ECG signal.

41. A wearable garment according to any one of claims 13 to 40, wherein: The wearable garment includes a wireless transceiver arranged to facilitate transmission of data indicative of the PCG signal and the ECG signal from the wearable device.

42. A system for monitoring body generated data, the system comprising: A wearable garment according to any one of claims 13 to 41; a noise cancellation component arranged to remove environment-related noise from the body audio signal using the ambient audio signal, thereby producing a noise-processed body audio signal, the noise-processed body audio signal having an improved signal-to-noise ratio; and a feature extractor arranged to extract a plurality of features from the noise-processed body audio signal and generate at least one feature vector comprising the plurality of features; and A data analysis component is arranged to predict at least one medical abnormality associated with the animal's body using the at least one feature vector.

43. The system of claim 42, wherein: The data analysis component uses at least one machine learning component that is trained to learn the relationship between the feature vectors and the medical diagnosis.

44. A system according to claim 42 or 43, wherein: The extracted features are associated with the ECG signal, and are derived using the R peak, P peak, Q point, S point, and T peak of the ECG signal, and the extracted features are derived using the time intervals between R peaks, the difference between consecutive R peak time intervals, and / or the time between P peaks and R peaks.

45. The system of claim 44, wherein: The extracted features associated with the ECG signal are obtained using: frequency domain characteristics, which can include mean, variance, and higher-order moments; skewness; kurtosis; and / or Time domain and time-frequency domain characteristics, which can be obtained using wavelets or wavelet packets (WP) and / or synchronized wavelet packets (SWP) of the ECG signal.

46. ​​A system according to claim 44 or 45, wherein: The extracted features associated with the PCG signal are derived by using the S1, systolic, S2, and diastolic states of the PCG cycle.

47. A system according to any one of claims 44 to 46, wherein: The extracted features associated with the PCG signal are obtained by using time domain characteristics, frequency domain characteristics, energy characteristics, entropy characteristics and / or kurtosis characteristics.

48. A system according to any one of claims 44 to 47, wherein: The extracted features include features corresponding to changes in frequency power of a specific frequency band of a specific segment of the cardiac cycle.

49. A system according to any one of claims 42 to 48, wherein: The machine learning component includes a classifier implemented using a k-nearest neighbor algorithm (KNN), a support vector machine (SVM), and / or an artificial neural network (ANN).

50. A method of monitoring an animal for a medical abnormality, the method comprising: A plurality of phonocardiogram sensing devices are provided, each phonocardiogram sensing device comprising: a body audio sensing device, the body audio sensing device comprising a body audio sensing transducer arranged to sense animal body audio when the body audio sensing device is disposed adjacent to an animal body at a body audio sensing location, wherein when the body audio sensing device is disposed at the body audio sensing location, an acoustic path for body audio is defined between the animal body and the body audio sensing transducer; and an ambient audio sensing device, the ambient audio sensing device comprising an ambient audio sensing transducer arranged to sense ambient audio present in an environment proximate to the phonocardiogram sensing device when the body audio sensing device is disposed at the body audio sensing location; placing the body audio sensing device at a defined location on the animal's body; generating, using the body audio sensing devices, a body audio signal indicative of the animal's body audio and an ambient audio signal indicative of the ambient audio present in the environment proximate each body audio sensing device; removing environment-related noise from the body audio signal using the ambient audio signal, thereby producing a noise-processed body audio signal, the noise-processed body audio signal having an improved signal-to-noise ratio; extracting a plurality of features from the noise-processed body audio signal and generating at least one feature vector including the plurality of features; and At least one medical abnormality associated with the animal's body is predicted using the at least one feature vector.

51. The method of claim 50, comprising defining an audio cavity using a diaphragm, the diaphragm enhancing body sounds emanating from the animal's body when the body audio sensing device is positioned adjacent the animal's body.

52. A method as claimed in claim 50 or 51 comprising connecting the body audio sensing device to the ambient audio sensing device using a vibration damping connection.

53. The method of claim 50 or 51, wherein: The body audio sensing transducer and / or the ambient audio sensing transducer include a MEMS microphone.

54. A method according to any one of claims 50 to 52, comprising filtering a body audio signal produced by the body audio sensing transducer and / or an ambient audio signal produced by the ambient audio sensing transducer.

55. A method according to any one of claims 50 to 53, comprising amplifying a body audio signal produced by the body audio sensing transducer and / or an ambient audio signal produced by the ambient audio sensing transducer.

56. The method according to any one of claims 50 to 55, wherein: The at least one PCG sensing device is incorporated into a wearable garment.

57. The method of claim 56, comprising positioning a plurality of PCG sensing devices at selected locations on the wearable garment to optimize acquisition of relevant body audio signals.

58. The method of claim 57, comprising placing a plurality of front PCG devices on the wearable garment, the plurality of front PCG devices comprising a PCG device on the opposite side of the sternum of the animal's body and aligned with the pulmonary artery, a PCG device on the wearable garment on the opposite side of the sternum of the animal's body and aligned with the tricuspid valve, a PCG device on the wearable garment on the midclavicular line adjacent to the mitral valve area, and a PCG device on the midaxillary line adjacent to the midaxillary area when in use.

59. The method according to claim 57 or 58 includes setting multiple rear PCG devices on the wearable garment, and the multiple rear PCG devices include PCG devices on the wearable garment on opposite sides of the vertical centerline directly below the shoulder blades of the animal's body, PCG devices on opposite sides of the vertical centerline set at the middle part of the back of the animal's body, and PCG devices on opposite sides of the vertical centerline adjacent to the lower lobe of the lung of the animal's body.

60. The method of any one of claims 56 to 59, comprising positioning at least one cervical PCG device disposed on the wearable garment at the patient's neck region to sense audio from a carotid artery.

61. A method according to any one of claims 56 to 60, comprising providing a plurality of ECG electrodes at selected locations of the wearable garment to optimize acquisition of relevant body electrical signals.

62. The method of claim 61, wherein: The plurality of ECG electrodes include an RA ECG electrode disposed on a right side of an atrium of the animal's body and a LA ECG electrode disposed on a left side of the atrium.

63. The method of claim 62, wherein: The plurality of ECG electrodes include RLD ECG electrodes, the RLDECG electrodes being disposed vertically at a location below the heart of the animal's body and aligned with the RAECG electrodes.

64. The method of claim 63, comprising driving a feedback electrical signal into the animal's body at the RLD electrode using signals acquired from the RA electrode and the LA electrode, the feedback signal being used to improve common mode rejection by canceling out common mode signals that would otherwise be present in the RA electrode and the LA electrode.

65. A method according to any one of claims 50 to 64, comprising acquiring respiratory and cardiac cycle data using at least one ultrasonic sensor.

66. The method of claim 65, comprising positioning the at least one ultrasound sensor on the wearable device such that, in use, the at least one ultrasound sensor is positioned adjacent to the patient's heart or chest.

67. A method according to any one of claims 50 to 66, comprising acquiring photoplethysmography (PPG) data using at least one PPG sensor.

68. The method of claim 67, wherein: The PPG data is used to monitor changes in blood volume and blood oxygenation.

69. The method of claim 68, comprising using the PPG data to derive physiological parameters associated with heart rate variability, blood pressure, ankle-brachial pressure, cardiovascular disease, aging, neurological disorders, lung disease and / or respiratory rate.

70. A method according to any one of claims 67 to 69 when dependent on claim 65, comprising using data acquired from the at least one ultrasonic sensor and the PPG data to detect the presence of fluid in the patient's lungs.

71. A method according to any one of claims 61 to 70, comprising synchronously receiving signals from the PCG device and the ECG electrodes at a data acquisition device.

72. The method of claim 71 when dependent upon claim 65, comprising synchronously receiving at the data acquisition device signals from at least one ultrasound device, the PCG device, and the ECG electrodes.

73. The method of claim 72 when referring to claim 67, comprising synchronously receiving at the data acquisition device signals from at least one PPG device, the at least one ultrasound device, the PCG device, and the ECG electrodes.

74. A method according to any one of claims 50 to 73, comprising filtering a PCG signal received from the PCG device.

75. The method of claim 74, comprising using an AC coupling filter to remove a DC offset voltage in the PCG signal.

76. The method of claim 74 or 75 when dependent on claim 62, comprising receiving ECG signals from the RA ECG electrode and the LA ECG electrode, and amplifying the difference between the RA ECG signal and the LA ECG signal to produce an amplified difference signal.

77. The method of claim 76, comprising removing a DC component from the amplified difference signal.

78. A method according to any one of claims 50 to 77, comprising using a notch filter to suppress interference at approximately 50 Hz.

79. The method of any one of claims 71 to 78, comprising converting the analog PCG signal and the analog ECG signal into digital PCG signal and digital ECG signal using an A / D converter.

80. The method of claim 79, comprising using an anti-aliasing filter to ensure that substantially no aliasing occurs in the A / D converter.

81. The method of any one of claims 50 to 80, comprising removing the environment-related noise from the body audio signal using the ambient audio signal, thereby producing the noise-processed body audio signal, the noise-processed body audio signal having an improved signal-to-noise ratio.

82. The method of claim 81 , comprising extracting the plurality of features from the noise-processed body audio signal, generating the at least one feature vector comprising the plurality of features, and predicting the at least one medical abnormality associated with the animal's body using the at least one feature vector.

83. The method of claim 82, comprising using at least one machine learning component trained to learn the relationship between the feature vector and the medical diagnosis.

84. The method of claim 82 or 83, wherein: The extracted features are associated with the ECG signal, and are derived using the R peak, P peak, Q point, S point, and T peak of the ECG signal, and the extracted features are derived using the time intervals between R peaks, the difference between consecutive R peak time intervals, and / or the time between P peaks and R peaks.

85. The method of claim 84, comprising obtaining the extracted features associated with the ECG signal using: frequency domain characteristics, which can include mean, variance, and higher-order moments; skewness; kurtosis; and / or Time domain and time-frequency domain characteristics, which can be obtained using wavelets or wavelet packets (WP) and / or synchronized wavelet packets (SWP) of the ECG signal.

86. The method of claim 84 or 85, comprising using the S1, systolic, S2, and diastolic states of the PCG cycle to derive the extracted features associated with the PCG signal.

87. The method according to any one of claims 84 to 86 comprises using time domain characteristics, frequency domain characteristics, energy characteristics, entropy characteristics and / or kurtosis characteristics to obtain the extracted features associated with the PCG signal.

88. The method of any one of claims 84 to 87, wherein: The extracted features include features corresponding to changes in frequency power of a specific frequency band of a specific segment of the cardiac cycle.

89. The method of any one of claims 82 to 88, comprising using a classifier implemented by using a k-nearest neighbor algorithm (KNN), a support vector machine (SVM) and / or an artificial neural network (ANN).

Images