Wearable device and method for assessing respiratory tract infections

Abstract

The application provides a wearable device and a method for evaluating respiratory tract infection. The wearable device can include a first sensor, a second sensor, and a processor. The first sensor can obtain an audio signal of a user, the second sensor can obtain a physiological parameter signal of the user, and the processor can obtain a respiratory tract infection evaluation report of the user according to the physiological parameter signal and the audio signal. Based on the respiratory tract infection evaluation report, respiratory tract infection can be detected in an early stage, so as to facilitate confirmation, intervention, and treatment.

Description

Technical Field

[0001] This application relates to the technical field of wearable devices, specifically to a wearable device and a method for assessing respiratory infections. Background Technology

[0002] Currently available wearable devices can detect physiological parameters such as heart rate, blood oxygen, and body temperature. However, for diseases involving multiple physiological parameters, such as respiratory infections, wearable devices cannot assess the user's risk of developing the disease based on the obtained physiological parameters. Summary of the Invention

[0003] In view of this, this application provides a wearable device and a respiratory infection assessment method to solve the problem that existing wearable devices cannot assess the risk of a user having a disease.

[0004] Firstly, this application provides a wearable device. The wearable device may include a first sensor, a second sensor, and a processor. The first sensor is used to acquire the user's audio signal. The second sensor is used to acquire the user's physiological parameter signals. The processor is used to obtain the user's respiratory rate based on the physiological parameter signals, and to obtain a respiratory infection assessment report based on the audio signal and respiratory rate. It should be understood that, due to its portability and lightweight nature, the wearable device allows users to conveniently assess respiratory infections as needed. Based on the respiratory infection assessment report, the risk of the user having a respiratory infection can be assessed; furthermore, based on the wearable device, respiratory infections can be detected early, allowing for timely confirmation, intervention, and treatment to prevent the condition from worsening.

[0005] In some implementations, the second sensor may include a PPG sensor. The PPG sensor is used to acquire the user's PPG signal, and the processor is used to obtain the user's first respiratory rate based on the PPG signal. In some cases, the wearable device can obtain a respiratory infection assessment report for the user based on the first respiratory rate and audio signals.

[0006] In some implementations, the second sensor may also include an ACC sensor. The ACC sensor is used to acquire the user's ACC signal, and the processor is used to obtain the user's second respiratory rate based on the ACC signal. In some cases, the wearable device can obtain a respiratory infection assessment report for the user based on the second respiratory rate and audio signal.

[0007] In some implementations, the processor is also used to filter and fuse the first and second respiratory rates to obtain the user's third respiratory rate. It should be understood that obtaining the third respiratory rate from the first and second respiratory rates can improve the accuracy of the measured respiratory rate. Based on this, the accuracy of respiratory infection assessment reports can be improved.

[0008] In some implementations, the processor is also used to obtain the user's posture classification result based on the ACC signal; the posture classification result includes a first posture and a second posture, where the first posture includes the user's wearing parts being in a supported state, and the second posture includes the user's wearing parts being in a naturally placed state. It should be understood that by detecting the user's posture, a respiratory infection assessment report can be obtained using the corresponding respiratory rate, thereby improving the accuracy of the assessment report.

[0009] In some implementations, when the user is in the first posture, the processor generates a respiratory infection assessment report based on the first respiratory rate and the audio signal. It should be understood that when the user is in the first posture, the PPG sensor can acquire a PPG signal with a high signal-to-noise ratio. Based on this, the first respiratory rate can be obtained from the PPG signal, and thus the respiratory infection assessment report can be derived.

[0010] In some implementations, when the user is in a second posture, the processor generates a respiratory infection assessment report based on the third respiratory rate and audio signal. It should be understood that when the user is in a second posture, the PPG sensor can acquire a PPG signal with a high signal-to-noise ratio, and the ACC sensor can also acquire an ACC signal with a high signal-to-noise ratio. Based on this, the third respiratory rate can be derived from the first and second respiratory rates, and thus the respiratory infection assessment report can be obtained.

[0011] In some implementations, the wearable device may also include a low-pass filter for filtering the ACC signal; the cutoff frequency of the low-pass filter is 1Hz. The processor calculates the mean and standard deviation of the ACC signal, and calculates the power spectrum based on the low-pass filtered ACC signal to obtain the location and amplitude of the power spectrum peaks. The processor also inputs the mean and standard deviation of the ACC signal, and the location and amplitude of the power spectrum peaks into a classification model to obtain attitude classification results.

[0012] In some implementations, the processor can also prompt the user to switch to a second posture when the user is in a first posture and the first respiratory rate is outside a preset range. In response to the user's switch to the second posture, a third respiratory rate is obtained based on the PPG and ACC signals; and a respiratory infection assessment report is obtained based on the third respiratory rate and the audio signal.

[0013] In some implementations, the wearable device can include electronic devices such as wearable watches, wearable bracelets, or wearable monitors. Among these, wearable monitors can include multi-parameter monitors.

[0014] In some implementations, the wearable device may further include a first bandpass filter and a second bandpass filter. The first bandpass filter is used to bandpass filter the PPG signal to obtain the peak and trough positions of the PPG signal. The second bandpass filter is used to bandpass filter the PPG signal to obtain the peak and trough amplitudes of the PPG signal. The processor is also used to obtain the user's first respiratory rate based on the peak and trough positions and amplitudes of the PPG signal. It should be understood that the waveform of the PPG signal can be obtained through the second bandpass filter, and the peak and trough amplitudes of the PPG signal can be obtained by combining the peak and trough positions of the PPG signal.

[0015] In some implementations, the first bandpass filter allows the signal to pass through a frequency band of 0.5Hz to 10Hz, and the second bandpass filter allows the signal to pass through a frequency band of 0.1Hz to 10Hz.

[0016] In some implementations, the wearable device may also include a third bandpass filter for bandpass filtering the baseline-removed ACC signal. The processor is also used to obtain the user's second respiratory rate based on the bandpass-filtered ACC signal.

[0017] In some implementations, the third bandpass filter allows the signal to pass through a frequency band of 0.1Hz to 0.5Hz.

[0018] Secondly, this application provides a respiratory infection assessment method based on wearable devices. This assessment method may include: acquiring the user's audio signal; acquiring the user's physiological parameter signals; obtaining the user's respiratory rate based on the physiological parameter signals; and obtaining the user's respiratory infection assessment report based on the audio signal and respiratory rate. It should be understood that, based on the respiratory infection assessment report, the risk of the user having a respiratory infection can be assessed; furthermore, based on this assessment method, respiratory infections can be detected at an early stage, facilitating confirmation, intervention, and treatment by medical personnel to prevent the condition from worsening.

[0019] In some implementations, acquiring the user's physiological parameter signals specifically includes: acquiring the user's PPG signal; wherein the PPG signal includes physiological parameter signals. Based on the physiological parameter signals, the user's respiratory rate is obtained, specifically including: obtaining the user's first respiratory rate based on the PPG signal. In some cases, a respiratory infection assessment report for the user can be obtained based on the first respiratory rate and audio signals.

[0020] In some implementations, acquiring the user's physiological parameter signals specifically includes: acquiring the user's PPG signal and ACC signal; wherein both the PPG signal and the ACC signal include physiological parameter signals. Based on the physiological parameter signals, the user's respiratory rate is obtained, specifically including: obtaining the user's first respiratory rate based on the PPG signal; obtaining the user's second respiratory rate based on the ACC signal; and filtering and fusing the first and second respiratory rates to obtain the user's third respiratory rate. It should be understood that obtaining the third respiratory rate from the first and second respiratory rates can improve the accuracy of the measured respiratory rate. Based on this, the accuracy of respiratory infection assessment reports can be improved.

[0021] In some implementations, before acquiring the user's physiological parameter signals, the method further includes: acquiring the user's ACC signal; and obtaining the user's posture classification result based on the ACC signal. The posture classification result includes a first posture and a second posture. The first posture includes the user's wearing part being in a supported state, and the second posture includes the user's wearing part being in a naturally placed state. It should be understood that by detecting the user's posture, a respiratory infection assessment report can be obtained using the corresponding respiratory rate, thereby improving the accuracy of the assessment report.

[0022] In some implementations, obtaining the user's attitude classification result based on the ACC signal specifically includes: calculating the mean and standard deviation of the ACC signal; performing low-pass filtering on the ACC signal and calculating the power spectrum of the ACC signal to obtain the position and amplitude of the peak point of the power spectrum; wherein the cutoff frequency of the low-pass filter is 1Hz; and inputting the mean and standard deviation of the ACC signal, the position and amplitude of the peak point of the power spectrum into the classification model to obtain the attitude classification result.

[0023] In some implementations, obtaining a respiratory infection assessment report based on the audio signal and respiratory rate specifically includes: obtaining a respiratory infection assessment report based on a first respiratory rate and audio signal when the user is in a first posture; and obtaining a respiratory infection assessment report based on a third respiratory rate and audio signal when the user is in a second posture. It should be understood that when the user is in the first posture, the PPG sensor can acquire a PPG signal with a high signal-to-noise ratio. Based on this, the first respiratory rate can be obtained from the PPG signal, leading to the respiratory infection assessment report. When the user is in the second posture, in addition to the PPG signal, the ACC sensor can also acquire an ACC signal with a high signal-to-noise ratio. Based on this, the third respiratory rate can be obtained from the first and second respiratory rates, leading to the respiratory infection assessment report.

[0024] In some implementations, when the user is in the first posture and the first respiratory rate is outside the preset range, the evaluation method may also include: prompting the user to switch to the second posture; responding to the user's operation of switching to the second posture, and obtaining the third respiratory rate based on the PPG signal and ACC signal.

[0025] Thirdly, this application also provides a wearable device. The wearable device may include a processor, an ACC sensor, and a PPG sensor. The ACC sensor is used to acquire the user's ACC signal, and the PPG sensor is used to acquire the user's PPG signal. The processor is used to obtain the user's posture classification result based on the ACC signal; wherein the posture classification result includes a first posture and a second posture; the first posture includes the user's wearing part being in a supported state, and the second posture includes the user's wearing part being in a naturally placed state. When the user is in the first posture, the processor is also used to obtain the user's first respiratory rate based on the PPG signal. When the user is in the second posture, the processor is also used to obtain the user's first respiratory rate based on the PPG signal and the user's second respiratory rate based on the ACC signal, and to filter and fuse the first and second respiratory rates to obtain the user's third respiratory rate. It should be understood that based on the detection of the user's posture, the wearable device can obtain a posture classification result. And based on this posture classification result, the wearable device can use a corresponding method to obtain the user's respiratory rate, thereby improving the accuracy of the measured respiratory rate.

[0026] Fourthly, this application also provides a method for measuring respiratory rate based on a wearable device. This measurement method may include: acquiring the user's ACC signal and PPG signal; obtaining the user's posture classification result based on the ACC signal; wherein the posture classification result includes a first posture and a second posture; the first posture includes the user's wearing part being in a supported state, and the second posture includes the user's wearing part being in a naturally placed state; when the user is in the first posture, obtaining the user's first respiratory rate based on the PPG signal; when the user is in the second posture, obtaining the user's first respiratory rate based on the PPG signal and the user's second respiratory rate based on the ACC signal, and filtering and fusing the first and second respiratory rates to obtain the user's third respiratory rate. It should be understood that posture classification results can be obtained based on the detection of the user's posture. Based on this posture classification result, the user's respiratory rate can be obtained using a corresponding method to improve the accuracy of the measured respiratory rate.

[0027] This application obtains a respiratory infection assessment report by acquiring the user's audio and physiological parameter signals, thereby enabling the assessment of respiratory infections. Based on this, respiratory infections can be detected at an early stage, facilitating confirmation, intervention, and treatment. Attached Figure Description

[0028] Figure 1This is a frame diagram of a wearable device according to an embodiment of this application.

[0029] Figure 2 This is a frame diagram of a wearable device according to an embodiment of this application.

[0030] Figure 3 This is a framework diagram of obtaining a first respiratory rate based on a PPG signal according to an embodiment of this application.

[0031] Figure 4 This is a framework diagram of obtaining a second respiratory rate based on an ACC signal according to an embodiment of this application.

[0032] Figure 5 This is a diagram illustrating the user's initial posture.

[0033] Figure 6 This is a diagram illustrating the user in a second posture.

[0034] Figure 7 This is a framework diagram of obtaining attitude classification results based on ACC signals according to an embodiment of this application.

[0035] Figure 8 This is a schematic diagram of the interactive interface of a wearable device according to an embodiment of this application.

[0036] Figure 9 This is a flowchart of a respiratory infection assessment method according to an embodiment of this application.

[0037] Figure 10 This is a flowchart illustrating a method for obtaining a first respiratory rate based on a PPG signal according to an embodiment of this application.

[0038] Figure 11 This is a flowchart illustrating a method for obtaining a first respiratory rate based on an ACC signal according to an embodiment of this application.

[0039] Figure 12 This is a flowchart illustrating a method for extracting pose-related features according to an embodiment of this application. Detailed Implementation

[0040] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise.

[0041] References to "one embodiment" or "some embodiments" as described in this specification mean that at least one embodiment of this application includes a specific feature, structure, or characteristic described in connection with that embodiment. Therefore, phrases such as "in one embodiment," "in some embodiments," "in other embodiments," "in still other embodiments," "in some implementations," etc., appearing in different parts of this specification do not necessarily refer to the same embodiment, but rather mean "one or more, but not all, embodiments," unless otherwise specifically emphasized. The terms "comprising," "including," "having," and variations thereof mean "including but not limited to," unless otherwise specifically emphasized.

[0042] Typical wearable devices can be worn by users to obtain physiological parameters such as heart rate, electrocardiogram, blood oxygen saturation, and body temperature. These wearable devices can be, for example, the user's wrist, upper arm, or forearm. Therefore, users can view these physiological parameters through the wearable device. Alternatively, users can view them through a smart terminal wirelessly connected to the wearable device; this smart terminal can include a mobile phone. Alternatively, medical personnel can view them through a smart terminal wirelessly connected to the wearable device; this smart terminal can include a bedside monitor or a central station.

[0043] It should be understood that typical wearable devices usually only provide the function of acquiring and displaying physiological parameters, and they cannot use these physiological parameters to assess the user's risk of having a disease.

[0044] Respiratory infections and other illnesses typically affect a user's physical condition. For example, users may experience symptoms such as coughing, difficulty breathing, or fever. In the early stages of a respiratory infection, changes in the user's physical condition are relatively minor, and fluctuations in physiological parameters are also small. Users generally find it difficult to detect early symptoms of respiratory infections, and typical wearable devices cannot assess respiratory infections based on the acquired physiological parameters.

[0045] To address the aforementioned problems, the following embodiments of this application provide a wearable device and a respiratory infection assessment method based on the wearable device. This wearable device and assessment method acquire the user's audio signals and physiological parameter signals to obtain a respiratory infection assessment report based on the user, thereby indicating the user's risk of having a respiratory infection. In some cases, based on the wearable device and assessment method provided in the following embodiments, respiratory infections can be detected at an early stage, shifting the disease prevention and control focus forward, facilitating intervention and treatment.

[0046] In some embodiments, the wearable device may include a wearable watch. In other embodiments, the wearable device may also include electronic devices such as wearable bracelets or wearable monitors, without limitation.

[0047] It should be understood that, in addition to assessing respiratory infections, the wearable devices provided in each embodiment can also perform functions such as fall detection, wireless communication, or message alerts, without limitation.

[0048] Please refer to Figure 1 A wearable device 10 provided in one embodiment of this application may include a processor 12 and a first sensor 14. The processor 12 may be electrically or wirelessly connected to the first sensor 14, without limitation. The first sensor 14 may be used to acquire audio signals from a user 20. These audio signals may measure sounds emitted by the user 20 from their respiratory system. It should be understood that the first sensor 14 may also be used to acquire sounds emitted by the user 20 within a preset time period, such as 10 seconds, 20 seconds, 30 seconds, or 1 minute, without limitation.

[0049] In some embodiments, the wearable device 10 may prompt the user 20 to cough, and the first sensor 14 may measure the cough to obtain an audio signal. Based on this audio signal, the processor 12 may analyze it in conjunction with other signals to assess the user 20's risk of having a respiratory infection, and the results may be presented as a respiratory infection assessment report.

[0050] As described above, the wearable device 10 can prompt the user 20 to make a sound in various ways. For example, the wearable device 10 can prompt the user 20 to cough by voice or display prompt information about the user 20 coughing. Based on this, after the user 20 coughs, the wearable device 10 can respond to the cough by obtaining an audio signal.

[0051] In other embodiments, the smart terminal may also prompt the user 20 to cough in various ways so that the first sensor 14 can collect the sound. For example, the smart terminal may provide a voice prompt or a visual prompt. After the user 20 coughs, the wearable device 10 can respond to the cough by acquiring the audio signal.

[0052] It should be understood that the cough sounds emitted from the respiratory system by a user 20 without a respiratory infection and a user 20 with a respiratory infection will differ significantly. This difference can be reflected in their respective audio signals. Based on this, the processor 12 can perform operations such as feature extraction on the audio signals, and classify the audio signals through methods such as machine learning, in order to subsequently assess the risk of user 20 having a respiratory infection.

[0053] In some embodiments, the processor 12 may refer to an MCU (Microcontroller Unit) or a CPU (Central Processing Unit) in the wearable device 10, but is not limited thereto. The processor 12 may also refer to a circuit board assembly in the wearable device 10; the circuit board assembly may integrate electronic devices such as MCUs, CPUs, radio frequency chips, or filters.

[0054] Please refer to this again. Figure 1 In some embodiments, the wearable device 10 may further include a second sensor 16. Similar to the first sensor 14, the second sensor 16 may be electrically or wirelessly connected to the processor 12. The second sensor 16 can be used to acquire physiological parameter signals of the user 20. These physiological parameter signals may include at least a signal for measuring the user 20's respiratory rate, such as, but not limited to, the PPG and ACC signals mentioned below. Depending on the number and type of sensors configured in the wearable device 10, the physiological parameter signals may also include signals for measuring physiological parameters such as the user 20's heart rate, electrocardiogram, or blood oxygen saturation.

[0055] It should be understood that for a user 20 suffering from a respiratory infection, their physiological parameters will generally fluctuate abnormally; such abnormal fluctuations may refer to at least some of the user 20's physiological parameters being outside the normal range. Based on this, in the wearable device 10 provided in the various embodiments of this application, the processor 12 can analyze the user 20's audio signals and physiological parameter signals to determine whether the user 20's physical condition has changed, and thus comprehensively assess the risk of the user 20 suffering from a respiratory infection.

[0056] In some embodiments, after assessing respiratory infection in user 20, the corresponding respiratory infection assessment report for user 20 may record information such as "no abnormalities found," "low risk of respiratory infection," "medium risk of respiratory infection," or "high risk of respiratory infection." It should be understood that the wearable device 10 may present the corresponding respiratory infection assessment report on its interactive interface for user 20 or medical personnel to view. Alternatively, the smart terminal may also present the respiratory infection assessment report on its interactive interface; there are no limitations on this.

[0057] For example, for user 20 with an early respiratory infection, it is difficult for user 20 to notice the physical changes they experience in response to the infection. However, the second sensor 16 of the wearable device 10 can acquire physiological parameter signals corresponding to these changes. The processor 12 can analyze these physiological parameter signals to obtain user 20's respiratory rate. Based on this, the analysis reveals an abnormal respiratory rate in user 20. Combined with user 20's audio signal, a respiratory infection assessment report can be generated. This assessment report can record information indicating a high risk of respiratory infection.

[0058] It should be understood that, based on the wearable device 10 provided in this application embodiment, user 20 can conveniently and quickly assess respiratory infections. In some cases, based on this wearable device 10, respiratory infections can be detected at an early stage, facilitating medical personnel to confirm, intervene in, and treat the respiratory infection, preventing the condition from worsening.

[0059] Please refer to Figure 2 In some embodiments, the second sensor 16 may include a PPG (Photoplethysmography) sensor 16a. This PPG sensor 16a can acquire the PPG signal of the user 20. It should be understood that, due to the commonalities of human anatomy, the user 20's breathing movements affect blood flow in the blood vessels, and this effect can be reflected in the acquired PPG signal. Based on this, the processor 12 can extract feature signals related to the user 20's breathing movements from the PPG signal as physiological parameter signals, thereby calculating the user 20's first respiratory rate based on these physiological parameter signals.

[0060] In some embodiments, the second sensor 16 may further include an ACC (Accelerometer) sensor 16b. The ACC sensor 16b can acquire the ACC signal of the user 20. It should be understood that the user 20's breathing movements can cause the rise and fall of the chest and abdomen, as well as the swinging of the user 20's upper limbs. In some cases, this upper limb swinging motion can be captured by the ACC sensor 16b and modulated into the ACC signal. Based on this, the processor 12 can extract feature signals related to the user 20's breathing movements from the ACC signal as physiological parameter signals, and calculate the user 20's second respiratory rate based on these physiological parameter signals.

[0061] In some embodiments, the PPG signal or ACC signal can also be understood as a physiological parameter signal. The wearable device 10 can process the PPG signal and ACC signal to obtain the respiratory rate of the user 20, without necessarily needing to extract the feature signal related to the respiratory rate from the PPG signal or ACC signal as a physiological parameter signal.

[0062] In some embodiments, the number of first sensors can be multiple to improve the accuracy of the acquired signals. Similarly, the number of second sensors can also be multiple to improve the accuracy of the acquired signals.

[0063] Please refer to Figure 3 In some embodiments, corresponding to the PPG signal, the wearable device 10 may further include a first bandpass filter 18a and a second bandpass filter 18b. The first bandpass filter 18a can be used to bandpass filter the PPG signal to obtain the peak and valley positions of the PPG signal. The second bandpass filter 18b can also be used to bandpass filter the PPG signal to obtain the waveform of the PPG signal.

[0064] By matching the waveform and peak / valley positions of the PPG signal, the peak / valley amplitudes of the PPG signal can be obtained. Based on the peak / valley positions and amplitudes of the PPG signal, the baseline drift (BW), amplitude modulation (AM), and frequency modulation (FM) characteristics related to the respiratory rate can be obtained. Based on this, the first respiratory rate based on the PPG signal can be obtained by calculating the baseline drift, amplitude, and frequency characteristics.

[0065] In some embodiments, the first bandpass filter 18a allows signals to pass through a frequency band of 0.5 Hz to 10 Hz. The second bandpass filter 18b allows signals to pass through a frequency band of 0.1 Hz to 10 Hz.

[0066] Please refer to Figure 4 In some embodiments, the baseline of the ACC signal can be removed by software to prevent baseline drift interference. The wearable device 10 may also include a third bandpass filter 18c, which can be used to bandpass filter the baseline-removed ACC signal to retain the frequency band corresponding to the respiratory rate. The processor 12 can also be used to extract the peaks of the filtered ACC signal to obtain a second respiratory rate based on the ACC signal.

[0067] In some embodiments, the frequency band that the third bandpass filter 18c allows the signal to pass through can be 0.1Hz to 0.5Hz.

[0068] In some embodiments, the processor 12 can also be used to filter and fuse the first respiratory rate and the second respiratory rate to obtain the third respiratory rate of the user 20.

[0069] It should be understood that the second respiratory rate obtained through the ACC signal is more accurate than the first respiratory rate obtained through the PPG signal. However, the ACC signal is also more easily affected by the wearing method of the wearable device 10, the shaking movements and posture of the user 20, etc., thus limiting its applicability. In some cases, by filtering and fusing the first and second respiratory rates, a more accurate third respiratory rate can be obtained, thereby improving the accuracy of respiratory infection assessment reports.

[0070] In some embodiments, the first respiratory rate, the second respiratory rate, and the third respiratory rate can be understood as the respiratory rate of user 20 obtained through different methods. In the wearable device 10 provided in various embodiments, the respiratory infection assessment report of user 20 is mainly obtained using the first respiratory rate or the third respiratory rate, but is not limited thereto.

[0071] In some other embodiments, a respiratory infection assessment report for user 20 can be obtained using the second respiratory rate.

[0072] In some embodiments, the processor 12 can also be used to obtain the attitude classification result of the user 20 based on the ACC signal acquired by the ACC sensor 16b.

[0073] It should be understood that the ACC signal may include a triaxial acceleration signal associated with user 20. By analyzing the ACC signal, the posture classification result of user 20 can be obtained. Based on the posture classification result, wearable device 10 can select different methods to obtain the respiratory rate of user 20 to improve the accuracy of respiratory infection assessment reports.

[0074] The posture classification result can include a first posture and a second posture. The first posture can refer to the state in which the user 20's wearing part is under force support. The second posture can refer to the state in which the user 20's wearing part is in a natural position.

[0075] When user 20 is in the first posture, the first respiratory rate of user 20 can be obtained through the PPG signal. When user 20 is in the second posture, the third respiratory rate of user 20 can be obtained through the PPG signal and the ACC signal. It should be understood that, based on the posture classification result, the wearable device 10 can adaptively select the corresponding measurement method, thereby improving the accuracy of the measured respiratory rate.

[0076] In some embodiments, the wearable device 10 may prompt the user 20 to measure the respiratory rate in a first posture or a second posture on the interactive interface.

[0077] In some other embodiments, the wearable device 10 can also automatically obtain the posture classification result of the user 20 based on the ACC signal; based on the posture classification result, the corresponding sensor (14, 16a, 16b) can be selected to cooperate with the processor 12 to obtain the breathing rate of the user 20.

[0078] It should be understood that when the user 20 is in a specific first or second posture, interference caused by the external environment and the user 20’s unintentional shaking or trembling can be eliminated. Based on this, the wearable device 10 can acquire PPG or ACC signals with a high signal-to-noise ratio, thereby improving the accuracy of the obtained first or third respiratory rate and the accuracy of the respiratory assessment report.

[0079] Let's take a wearable device 10 as a watch, and the part worn by the user 20 as the wrist as an example.

[0080] Figure 5 The example illustrates the user's initial posture. Please refer to it simultaneously. Figures 1 to 5 The seated user 20 has their forearm stably resting on the table, with an interaction force between their forearm and the table. In this situation, the PPG signal of user 20 can be acquired via PPG sensor 16a. By processing the PPG signal, the user 20's first respiratory rate can be obtained.

[0081] Figure 6 The example illustrates a user in a second pose. Please refer to it simultaneously. Figures 1 to 6 The seated user 20 rests their hands naturally on their thighs, with their wrists relaxed. In this position, along with the user 20's breathing, the PPG sensor 16a acquires the user 20's PPG signal, and the ACC sensor 16b acquires the user 20's ACC signal. By processing the PPG and ACC signals, a first respiratory rate and a second respiratory rate can be obtained. The processor 12 can also filter and fuse the first and second respiratory rates to obtain the user 20's third respiratory rate.

[0082] It should be understood that the above Figure 5 The first posture and Figure 6 The second posture is given as an example only and is not intended to be limiting. In other embodiments, the first posture may further include the user 20 placing their forearm on the armrest of the chair. The second posture may further include the seated user 20 placing their palm on their abdomen; or the standing user 20 having their arms naturally lowered, etc.

[0083] In some embodiments, the processor 12 can extract attitude-related features from the ACC signal and input the extracted features into a classification model. Based on these features, the classification model can classify the user 20's attitude to distinguish between a first attitude and a second attitude.

[0084] In some embodiments, the classification model may include SVM (Support Vector Machine), decision tree, XGboost, neural network, or deep learning, etc.

[0085] Please refer to Figure 7 In some embodiments, the processor 12 can calculate the mean and standard deviation of the ACC signal. Furthermore, the wearable device 10 may also include a low-pass filter 18d that can perform low-pass filtering on the ACC signal; wherein the cutoff frequency of the low-pass filter 18d can be 1Hz.

[0086] Processor 12 can also calculate the power spectrum of the low-pass filtered ACC signal to obtain the peak position and amplitude of the power spectrum. Based on the mean and standard deviation of the ACC signal, and the position and amplitude of the peak position of the power spectrum, the classification model can classify the ACC signal to determine the attitude of user 20 at the current moment. For example, after processing the ACC signal, the attitude of user 20 at the current moment can be determined to be the first attitude; or, the attitude of user 20 at the current moment can be determined to be the second attitude.

[0087] In some embodiments, the wearable device 10 may display a guide image of a first posture or a second posture on its interactive interface to guide the user 20 to adjust their posture accordingly. In other embodiments, the smart terminal may also display a guide image of a first posture or a second posture on its interactive interface, without limitation.

[0088] In some embodiments, when the user 20's respiratory rate (e.g., a first respiratory rate or a third respiratory rate) is outside a preset range, the wearable device 10 can also display a prompt message on its interactive interface. This prompt message can encourage the user 20 to change posture and re-measure. If the user 20 chooses to change posture and re-measure, signals can be acquired using the corresponding sensors in the changed posture. If the user 20 does not choose to change posture or actively chooses not to change posture, the risk of the user 20 having a respiratory infection can be assessed based on the previously obtained respiratory rate.

[0089] In some embodiments, the prompts from the wearable device 10 may also prompt the user 20 to adjust the position of the wearable device 10 on the wearing area or to put the wearable device 10 back on.

[0090] In some embodiments, the preset range of respiratory rate can be 9 BPM to 24 BPM; where BPM is the number of breaths per minute.

[0091] For example, when user 20 is in the first posture, the wearable device 10 obtains a respiratory rate of 8 BPM, which is outside the preset range. Based on this, the wearable device 10 can prompt user 20 to switch to the second posture and re-measure on its interface. If user 20 chooses to switch, the wearable device 10 can respond to user 20's action and remeasure user 20's respiratory rate. If user 20 chooses not to switch, the wearable device 10 can obtain a respiratory infection assessment report for user 20 based on the previously obtained respiratory rate (i.e., 8 BPM) and the audio signal.

[0092] In some embodiments, during the process of assessing respiratory infection, the wearable device 10 may also provide prompts through images, text, or voice to prompt the user 20 to perform relevant operations.

[0093] Please refer to Figure 8 In some embodiments, the wearable device 10 can display the respiratory infection assessment report and also display information related to respiratory rate. For example, the wearable device can exemplarily display the changes in the user 20's respiratory rate over a period of time. This period of time can refer to the most recent few minutes, tens of minutes, or several hours. Figure 8 As an example, the respiratory infection assessment report showed no abnormalities, and the wearable device 10 also displayed the user 20's respiratory rate changes from 1 hour to 7 hours.

[0094] In some embodiments, the wearable device 10 can also be used to measure the respiratory rate of the user 20 in daily work and life. For example, based on the user 20's posture, a first respiratory rate or a third respiratory rate can be obtained. The first respiratory rate or the third respiratory rate can be displayed in real time on the main interface or default interface of the wearable device for the user 20 to view. The main interface can be a manufacturer-defined interface, or it can refer to the interface that the user sees after the wearable device is powered on without any operation. The default interface can be an interface in the wearable device that the user can customize; or, the default interface can be the main interface.

[0095] It should be understood that for some wearable devices, after loading different theme plugins, the main interface of these wearable devices may or may not display the respiratory rate, but the main interface should still be considered as the main interface or default interface described in the above embodiments.

[0096] Please refer to Figure 9This application also provides a respiratory infection assessment method based on a wearable device. Similar to the wearable device provided in the above embodiments, this assessment method can also assess respiratory infections. The assessment method may include, but is not limited to, the following steps:

[0097] 101: Obtain the user's audio signal.

[0098] In some embodiments, the audio signal can measure sounds emitted by the user's respiratory system. For example, an audio signal can be obtained by measuring the user's cough. Based on the audio signal, the assessment method can combine it with other signals for analysis to assess the user's risk of having a respiratory infection, and the results can be presented as a respiratory infection assessment report.

[0099] In some embodiments, the user can be prompted to make a sound on the user interface of the wearable device, such as a cough, and the device can then acquire an audio signal in response to the cough. Alternatively, the user can be prompted to cough on the user interface of a smart terminal, and the wearable device can acquire an audio signal in response to the cough.

[0100] It should be understood that the cough sounds produced by healthy users (i.e., those without respiratory infections) and those with respiratory infections will differ significantly. This difference can be reflected in their respective audio signals. Based on this, feature extraction and machine learning methods can be used to classify the audio signals to facilitate subsequent assessment of a user's risk of respiratory infection.

[0101] 102: Obtain the user's physiological parameter signals.

[0102] This physiological parameter signal can be used to measure a user's respiratory rate, but is not limited to that. Depending on the number and type of sensors configured in the wearable device, this physiological parameter signal can also be used to measure physiological parameters such as the user's heart rate, electrocardiogram, or blood oxygen saturation.

[0103] It should be understood that, provided there is no conflict, there is no necessary order between the steps. For example, regarding steps 101 and 102, step 101 can be executed first, followed by step 102. Alternatively, step 102 can be executed first, followed by step 101.

[0104] 103: Obtain the user's respiratory rate based on physiological parameter signals.

[0105] The physiological parameter signal can be a portion of the PPG signal or the ACC signal. By analyzing the PPG signal or the ACC signal, the physiological parameter signal can be obtained, and thus the user's respiratory rate can be determined.

[0106] It should be understood that PPG or ACC signals can also be directly interpreted as physiological parameter signals. Assessment methods can process PPG and ACC signals to obtain the user's respiratory rate, without necessarily needing to extract respiratory rate-related feature signals from the PPG or ACC signals as physiological parameter signals.

[0107] 104: Based on the audio signal and respiratory rate, obtain the user's respiratory infection assessment report.

[0108] In some embodiments, since a user's respiratory rate is directly related to the health of their respiratory system, analyzing the user's audio signal and respiratory rate can determine whether the user's physical condition has changed, thereby enabling a comprehensive assessment of the user's risk of respiratory infection.

[0109] In some embodiments, after assessing a user for respiratory infection, the corresponding user's respiratory infection assessment report may record information such as "no abnormalities found," "low risk of respiratory infection," "medium risk of respiratory infection," or "high risk of respiratory infection." It should be understood that the respiratory infection assessment report may be presented on the interactive interface of a wearable device or smart terminal for the user's convenience.

[0110] For example, users with early-stage respiratory infections may find it difficult to notice changes in their physical condition as they respond to the infection. In the assessment method provided in this application, by acquiring physiological parameter signals related to changes in the user's physical condition and analyzing these signals, the user's respiratory rate can be obtained. Based on this, analysis reveals an abnormal respiratory rate, and combined with the user's audio signal, a respiratory infection assessment report can be generated. This report records information indicating a high risk of the user having a respiratory infection.

[0111] It should be understood that, based on the assessment method provided in the embodiments of this application, users can conveniently and quickly assess respiratory infections using wearable devices. In some cases, this assessment method can enable early detection of respiratory infections, facilitating medical personnel to confirm, intervene in, and treat the infections, thus preventing the condition from worsening.

[0112] In some embodiments, the respiratory infection assessment report also records information related to respiratory rate. For example, the respiratory infection assessment report may record the distribution of a user's respiratory rate over a period of time; where the period of time may refer to the most recent few minutes, tens of minutes, or several hours.

[0113] Prior to step 101, the evaluation method provided in this application embodiment may further include: detecting whether the user is wearing a wearable device.

[0114] In some embodiments, if the user is wearing a wearable device, step 101 can be performed to assess the user's respiratory infection.

[0115] In some other embodiments, step 101 will not proceed if the user is not wearing the wearable device. In some cases, the evaluation method may also prompt the user to wear the wearable device on the interface of the wearable device or smart terminal to perform step 101.

[0116] In some embodiments, the evaluation method may specifically include acquiring the user's PPG signal in terms of obtaining physiological parameter signals.

[0117] It should be understood that respiratory-related characteristic signals can be extracted from PPG signals as physiological parameter signals, and the user's first respiratory rate can be calculated from these signals.

[0118] Please refer to Figure 10 In some embodiments, corresponding to the PPG signal, the evaluation method may further include:

[0119] 111: Perform a first bandpass filter on the PPG signal to obtain the peak and valley positions of the PPG signal.

[0120] In some embodiments, the PPG can be filtered by a first bandpass filter. The first bandpass filter allows the signal to pass through a frequency band of 0.5 Hz to 10 Hz.

[0121] 112: Perform a second bandpass filter on the PPG signal, and combine the peak and valley positions of the PPG signal to obtain the peak and valley amplitude of the PPG signal.

[0122] In some embodiments, the PPG signal can be filtered by a second bandpass filter to obtain the waveform of the PPG signal. By matching the waveform of the PPG signal with the peak and valley positions, the peak and valley amplitudes of the PPG signal can be obtained.

[0123] In some embodiments, the second bandpass filter allows signals to pass through a frequency band of 0.1 Hz to 10 Hz.

[0124] 113: Based on the peak and trough positions and amplitudes of the PPG signal, the first respiratory rate based on the PPG signal is obtained.

[0125] In some embodiments, the base drift, amplitude, and frequency characteristics related to the respiratory rate can be obtained by analyzing the peak and trough positions and amplitudes of the PPG signal. The first respiratory rate based on the PPG signal can be obtained by calculating the base drift, amplitude, and frequency characteristics.

[0126] In some embodiments, the evaluation method may further include, in terms of acquiring physiological parameter signals, acquiring the user's ACC signal.

[0127] It should be understood that a user's breathing movements can cause the rise and fall of the chest and abdomen, as well as the swinging of the user's upper limbs. In some cases, this upper limb swinging motion can be captured by the ACC sensor and modulated into the ACC signal. Based on this, breathing-related characteristic signals can be extracted from the ACC signal as physiological parameter signals, thereby calculating the user's second respiratory rate.

[0128] Please refer to Figure 11 In some embodiments, for ACC signals, the evaluation method may specifically include:

[0129] 121: Remove the baseline of the ACC signal.

[0130] Based on step 121, the baseline of the ACC signal can be removed by software to prevent baseline drift interference of the ACC signal, thereby improving the accuracy of the measured respiratory rate.

[0131] 122: Bandpass filtering is applied to the baseline-removed ACC signal.

[0132] In some embodiments, the ACC signal can be filtered by a third bandpass filter to preserve the frequency band corresponding to the respiratory rate of the ACC signal. The frequency band that the signal is allowed to pass through by the third bandpass filter can be 0.1Hz to 0.5Hz.

[0133] 123: Peak extraction is performed on the filtered ACC signal to obtain the second respiratory rate based on the ACC signal.

[0134] In some embodiments, the assessment method may further include, in obtaining the respiratory rate:

[0135] The first and second respiratory rates are filtered and fused to obtain the user's third respiratory rate.

[0136] It should be understood that the second respiratory rate obtained through the ACC signal is more accurate than the first respiratory rate obtained through the PPG signal. However, the ACC signal is also more easily affected by the way wearable devices are worn, the user's shaking movements and posture, etc., thus limiting its applicability. In some cases, by filtering and fusing the first and second respiratory rates, a more accurate third respiratory rate can be obtained, thereby improving the accuracy of respiratory infection assessment reports.

[0137] In some embodiments, the first respiratory rate, second respiratory rate, and third respiratory rate can be understood as the user's respiratory rate obtained through different methods. In the assessment methods provided in various embodiments, the user's respiratory infection assessment report is primarily obtained using the first respiratory rate or the third respiratory rate, but this is not a limitation. In other embodiments, the user's respiratory infection assessment report can also be obtained using the second respiratory rate.

[0138] In some embodiments, to improve the accuracy of respiratory rate measurement, the assessment method may further include: obtaining the user's posture classification result based on the ACC signal before acquiring the physiological parameter signal.

[0139] It should be understood that the ACC signal may include a triaxial acceleration signal associated with the user. Analysis of the ACC signal can yield the user's posture classification results. Based on the posture classification results, different methods can be selected to obtain the user's respiratory rate, thereby improving the accuracy of respiratory infection assessment reports.

[0140] In some embodiments, the posture classification result may include a first posture and a second posture. The first posture may refer to a state where the user's wearing part is under stress and supported. The second posture may refer to a state where the user's wearing part is in a natural, relaxed position.

[0141] When the user is in the first posture, the first respiratory rate can be obtained through the PPG signal. When the user is in the second posture, the third respiratory rate can be obtained through the PPG and ACC signals. It should be understood that, based on the posture classification results, the corresponding measurement method can be adaptively selected, thereby improving the accuracy of the measured respiratory rate and the accuracy of respiratory infection assessment reports.

[0142] In some embodiments, similar to prompting the user to speak as described above, the user can be prompted to measure their respiratory rate in a first or second posture on the interactive interface of the wearable device.

[0143] In some other embodiments, the user's posture classification result can be automatically obtained based on the ACC signal; based on the posture classification result, a corresponding evaluation method can be selected to obtain the user's respiratory rate.

[0144] In some embodiments, the evaluation method for obtaining a user's pose classification result based on the ACC signal may specifically include: extracting pose-related features from the ACC signal and inputting the extracted features into a classification model. Based on these features, the classification model can classify the user's pose.

[0145] Please refer to Figure 12 In some embodiments, the evaluation method may specifically include, in terms of extracting pose-related features:

[0146] 131: Calculate the mean and standard deviation of the ACC signal.

[0147] 132: Perform low-pass filtering on the ACC signal.

[0148] In some embodiments, the ACC signal can be filtered using a low-pass filter. The cutoff frequency of the low-pass filter can be 1 Hz.

[0149] 133: Calculate the power spectrum of the ACC signal after low-pass filtering to obtain the position and amplitude of the peak point of the power spectrum.

[0150] In some embodiments, based on the mean and standard deviation of the ACC signal, and the location and amplitude of the peak points of the power spectrum, the classification model can classify the ACC signal to determine the user's posture at the current moment.

[0151] In some embodiments, a guide image of a first posture or a second posture can be displayed on the interactive interface of the wearable device to guide the user to adjust their posture accordingly. In other embodiments, the guide image of the first posture or the second posture can also be displayed on the interactive interface of a smart terminal, without limitation.

[0152] In some embodiments, when a user's respiratory rate (e.g., the first respiratory rate or the third respiratory rate) is outside a preset range, a prompt message can be displayed on the wearable device's interface. This prompt message can encourage the user to change posture and remeasure. If the user chooses to change posture and remeasure, the signal can be acquired using the corresponding sensor in the new posture. If the user does not choose to change posture or actively chooses not to change posture, the risk of the user having a respiratory infection can be assessed based on the previously obtained respiratory rate.

[0153] In some embodiments, the prompt message may also prompt the user to adjust the position of the wearable device on the wearing area or to put the wearable device back on.

[0154] In some embodiments, the preset range of respiratory rate can be 9 BPM to 24 BPM. For example, when the user is in the first posture, the respiratory rate obtained through the PPG signal is 25 BPM, which is outside the preset range. Based on this, the user can be prompted on the wearable device's interface to switch to the second posture and re-measure. If the user chooses to switch, the user's respiratory rate can be remeasured in response to the user's action. If the user chooses not to switch, a respiratory infection assessment report can be obtained based on the previously obtained respiratory rate (i.e., 26 BPM) combined with the audio signal.

[0155] In some embodiments, during the implementation of the respiratory infection assessment method, prompts may be provided through images, text, or voice to guide users to perform relevant operations.

[0156] In some embodiments, the assessment method may further include providing guidance information through images, text, or voice. This guidance information can be used to guide the user in performing relevant operations to facilitate the acquisition and processing of audio signals and physiological parameter signals.

Claims

1. A wearable device, comprising: The wearable device comprises: a first sensor configured to acquire an audio signal of a user; a second sensor configured to acquire a physiological parameter signal of the user; a processor configured to obtain a respiration rate of the user according to the physiological parameter signal, and obtain a respiratory tract infection assessment report of the user according to the audio signal and the respiration rate; the second sensor comprises a PPG sensor configured to acquire a PPG signal of the user, and the processor is configured to obtain a first respiration rate of the user according to the PPG signal; the second sensor further comprises an ACC sensor configured to acquire an ACC signal of the user, and the processor is further configured to obtain a second respiration rate of the user according to the ACC signal; the processor is further configured to obtain a posture classification result of the user according to the ACC signal, wherein the posture classification result comprises a first posture and a second posture, the first posture comprises that a wearing part of the user is in a force-supported state, and the second posture comprises that the wearing part of the user is in a natural placement state; when the user is in the first posture, the processor is configured to obtain the respiratory tract infection assessment report according to the first respiration rate and the audio signal; when the user is in the second posture, the processor is configured to obtain the respiratory tract infection assessment report according to a third respiration rate and the audio signal; the processor is further configured to filter and fuse the first respiration rate and the second respiration rate to obtain a third respiration rate of the user.

2. The wearable device of claim 1, wherein, The wearable device further comprises a low-pass filter configured to perform low-pass filtering on the ACC signal, wherein a cutoff frequency of the low-pass filter is 1 Hz; the processor is configured to calculate a mean value and a standard deviation of the ACC signal, and calculate a power spectrum according to the low-pass filtered ACC signal to obtain a position and an amplitude of a peak point of the power spectrum; the processor is further configured to input the mean value and the standard deviation of the ACC signal, and the position and the amplitude of the peak point of the power spectrum into a classification model to obtain the posture classification result.

3. The wearable device of claim 1, wherein, The processor is further configured to prompt the user to switch to the second posture when the user is in the first posture and the first respiration rate is out of a preset range; the processor is further configured to obtain the third respiration rate according to the PPG signal and the ACC signal in response to an operation of the user switching to the second posture, and obtain the respiratory tract infection assessment report according to the third respiration rate and the audio signal.

4. The wearable device of any one of claims 1 to 3, wherein, The type of the wearable device comprises a wearable watch, a wearable bracelet, or a wearable monitor.

Images