Respiratory rate processing method, device and computer-readable storage medium

Abstract

The embodiment of the present invention proposes a method, device and medium for processing respiratory rate. The method part includes: obtaining a PPG signal of a PPG sensor; obtaining a first signal feature of at least one dimension based on the PPG signal; processing the first signal feature to obtain a second signal feature, converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map; based on the frequency domain energy map, extracting at least two respiratory rate values by selecting the frequency domain energy value with the highest score; fusing and calculating at least two respiratory rate values to obtain a target respiratory rate value. The present invention first obtains two or more first signal features of at least one dimension, and then obtains at least two frequency domain energy maps corresponding to the first signal feature, obtains the respiratory rate value corresponding to the frequency domain energy value with the highest score for each frequency domain energy map, ensures the obtained respiratory rate value, and then fusing and calculating at least two respiratory rate values, and finally obtains the target respiratory rate value, thereby improving the accuracy of the respiratory rate measurement value.

Description

Technical Field

[0001] The present invention relates to the technical field of wearable devices, and in particular to a respiratory rate processing method, device, and computer-readable storage medium. Background Art

[0002] As people's living standards improve, health indicators are becoming increasingly important, and respiratory function is receiving increasing attention. A key parameter of respiratory function is respiratory rate, making accurate respiratory rate monitoring crucial for the prevention and identification of respiratory diseases. Respiratory rate is typically monitored using direct measurement methods such as airflow, including temperature, carbon dioxide, and humidity, or breath sound measurement and chest block. Indirect measurement methods include respiratory signals obtained through electromyography, electrocardiography, or infrared imaging. These methods are primarily used for clinical respiratory monitoring and are difficult to use in everyday life. Photoplethysmography (PPG) signal monitoring can indirectly obtain respiratory rate information. PPG is a photoelectric technology that can monitor changes in blood volume in human tissue during the cardiac cycle. It is non-invasive and simple to operate.

[0003] With the development of the smart wearable industry, users have higher and higher expectations for the intelligence of wearable devices. However, the current measurement of respiratory rate by smart wearable devices is easily affected by the user's environment, which affects the accuracy of the measured respiratory rate. Summary of the Invention

[0004] In view of this, the purpose of this application is to provide a respiratory rate processing method to detect the respiratory rate of a user. The method first obtains two or more first signal features in at least one dimension, then obtains at least two frequency domain energy maps corresponding to the first signal features, obtains the respiratory rate value corresponding to the highest-scoring frequency domain energy value for each frequency domain energy map, ensures the obtained respiratory rate value, and then fuses and calculates the at least two respiratory rate values to finally obtain a target respiratory rate value, thereby improving the accuracy of the respiratory rate measurement value.

[0005] In a first aspect, the present application provides a respiratory rate processing method, the method comprising: obtaining a PPG signal from a PPG sensor; obtaining a first signal feature of at least one dimension based on the PPG signal; processing the first signal feature to obtain a second signal feature, converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map; and extracting at least two respiratory rate values based on the frequency domain energy map by selecting the frequency domain energy value with the highest score;

[0006] The at least two respiratory rate values are fused and calculated to obtain a target respiratory rate value.

[0007] Optionally, in combination with the first aspect, in a possible implementation manner, the first signal feature includes an upper envelope and a lower envelope.

[0008] Optionally, in combination with the first aspect, in a possible implementation, processing the first signal feature to obtain a second signal feature, and converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map includes:

[0009] The upper envelope and the lower envelope are filtered respectively to obtain the upper envelope and the lower envelope in the respiratory frequency band; the upper envelope and the lower envelope are Fourier transformed respectively to obtain the upper envelope frequency domain energy map and the lower envelope frequency domain energy map.

[0010] Optionally, in combination with the first aspect, in a possible implementation, extracting at least two respiratory rate values by selecting the frequency domain energy value with the highest score based on the frequency domain energy map includes:

[0011] The score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy graph and the score of the lower envelope energy value of the lower envelope frequency domain energy graph are calculated by the following formula:

[0012] Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value;

[0013] Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph;

[0014] According to the frequency domain energy value score, determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the lower envelope frequency domain energy value with the highest score.

[0015] Optionally, in combination with the first aspect, in a possible implementation, performing fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value includes:

[0016] For each time point, the upper envelope frequency domain energy value with the highest score and the lower envelope frequency domain energy value with the highest score are fused and calculated to obtain the target respiratory rate value corresponding to each time point.

[0017] Optionally, in combination with the first aspect, in a possible implementation manner, obtaining a first signal feature of at least one dimension based on the PPG signal includes:

[0018] First signal features in at least two dimensions are acquired based on the PPG signal, where the first signal features include an upper envelope, a lower envelope, and an RR interval.

[0019] Optionally, in combination with the first aspect, in a possible implementation, processing the first signal feature to obtain a second signal feature, and converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map includes:

[0020] Filtering the first envelope feature to obtain a second envelope feature; preprocessing the RR interval value to obtain a fused RR interval value;

[0021] Performing Fourier transform on the second envelope feature to obtain an envelope frequency domain energy map; performing Fourier transform on the fused RR interval value to obtain an RR interval frequency domain energy map.

[0022] Optionally, in combination with the first aspect, in a possible implementation, the method further includes:

[0023] The PPG signal is transformed from the time domain to the frequency domain to obtain the PPG signal frequency domain energy map.

[0024] Optionally, in combination with the first aspect, in a possible implementation, extracting at least one first respiratory rate value by selecting the frequency domain energy value with the highest score includes:

[0025] Based on the envelope frequency domain energy map, the RR interval frequency domain energy map and the PPG signal frequency domain energy map, according to a pre-constructed frequency domain energy value scoring mechanism, the first respiratory rate value, the second respiratory rate value and the third respiratory rate value at each time point on the frequency domain energy map are extracted respectively.

[0026] Optionally, in combination with the first aspect, in a possible implementation, performing a fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value includes:

[0027] The first respiratory rate value, the second respiratory rate value, and the third respiratory rate value are fused to obtain a target respiratory rate value at each time point.

[0028] Optionally, in combination with the first aspect, in a possible implementation, the method further includes:

[0029] The fourth respiratory rate is obtained based on EEMD decomposition and reconstruction.

[0030] Optionally, in combination with the first aspect, in a possible implementation, performing fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value includes:

[0031] The first respiratory rate value, the second respiratory rate value, the third respiratory rate value and the fourth respiratory rate value are fused to obtain a target respiratory rate value at each time point.

[0032] Optionally, in combination with the first aspect, in a possible implementation method, obtaining a fourth respiratory rate based on EEMD decomposition and reconstruction includes: adding normally distributed white noise with different amplitudes to the PPG signal each time to obtain at least two mixed signals; performing EEMD decomposition on each mixed signal to obtain an IMF component; performing an averaging operation on all the IMF components to obtain a target IMF of the EEMD decomposition; and extracting the respiratory rate value corresponding to the target IMF.

[0033] Optionally, in combination with the first aspect, in a possible implementation, before obtaining the PPG signal of the PPG sensor, it also includes: if it is a first measurement mode, determining whether the wearable device is stationary according to the acceleration signal, and if it is stationary, turning on the PPG signal; if it is a second measurement mode, identifying whether the wearable device has fallen asleep according to a sleeping algorithm, and if it has fallen asleep, turning on the PPG signal of a second time every first time interval; the second time is less than the first time.

[0034] Optionally, in combination with the first aspect, in a possible implementation, the PPG signal is a green light signal, further comprising: judging the signal quality of the PPG signal according to an interval characteristic of the PPG signal; the interval characteristic of the PPG signal is obtained by calculating the interval difference between adjacent peaks; when adjacent intervals undergo continuous mutations, judging that the signal quality does not meet a preset condition; if the signal quality meets the preset condition, reading the signal, and preprocessing the PPG signal to remove the baseline signal to obtain a preprocessed PPG signal.

[0035] Optionally, in combination with the first aspect, in a possible implementation, obtaining a first signal feature of at least one dimension based on the PPG signal includes: obtaining the peaks and troughs of the preprocessed PPG signal; and interpolating the peaks and troughs to obtain upper and lower envelope features of the green light signal.

[0036] Optionally, in combination with the first aspect, in a possible implementation manner, obtaining the peaks and troughs of the preprocessed PPG signal includes: determining whether a PPG point of the preprocessed PPG signal is a maximum value or a minimum value;

[0037] If it is a maximum value, determine whether it is the maximum value of the adjacent window length N; if it is the maximum value, it is a valid peak; wherein, the selection of N is determined according to the frequency of the PPG signal; if it is a minimum value, determine whether it is the minimum value of the adjacent window length N; if it is the minimum value, it is a valid trough; wherein, the selection of N is determined according to the frequency of the PPG signal.

[0038] In a second aspect of the present application, a respiratory rate processing device is provided, the device comprising:

[0039] A first acquisition module is configured to acquire a PPG signal from a PPG sensor;

[0040] a second acquisition module, configured to acquire a first signal feature of at least one dimension based on the PPG signal;

[0041] a conversion module configured to process the first signal feature to obtain a second signal feature, and convert the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map;

[0042] an extraction module configured to extract at least two respiratory rate values by selecting the frequency domain energy value with the highest score based on the frequency domain energy map;

[0043] The fusion calculation module is configured to perform fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value.

[0044] Optionally, in combination with the second aspect, in a possible implementation manner, the first signal feature includes an upper envelope and a lower envelope.

[0045] Optionally, in combination with the second aspect, in a possible implementation, the conversion module includes:

[0046] The first filtering unit is configured to filter the upper envelope and the lower envelope respectively to obtain the upper envelope and the lower envelope in the respiratory frequency band; the first Fourier transform unit is configured to perform Fourier transform on the upper envelope and the lower envelope respectively to obtain the upper envelope frequency domain energy map and the lower envelope frequency domain energy map.

[0047] Optionally, in combination with the second aspect, in a possible implementation, the extraction module includes:

[0048] A calculation unit is configured to calculate the score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy map and the score of the lower envelope energy value of the lower envelope frequency domain energy map by the following formula,

[0049] Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value;

[0050] Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph;

[0051] The determination unit is configured to determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score based on the frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the lower envelope frequency domain energy value with the highest score.

[0052] Optionally, in combination with the second aspect, in a possible implementation, the fusion calculation module includes:

[0053] The first fusion calculation unit is configured to, for each time point, perform fusion calculation on the upper envelope frequency domain energy value with the highest score and the lower envelope frequency domain energy value with the highest score to obtain a target respiratory rate value corresponding to each time point.

[0054] Optionally, in combination with the second aspect, in a possible implementation, the second acquisition module includes:

[0055] The first acquisition unit is configured to acquire first signal features of at least two dimensions based on the PPG signal, where the first signal features include an upper envelope, a lower envelope, and an RR interval.

[0056] Optionally, in combination with the second aspect, in a possible implementation, the conversion module includes:

[0057] The second filtering unit is configured to filter the first envelope feature to obtain a second envelope feature; and preprocess the RR interval value to obtain a fused RR interval value;

[0058] The second Fourier transform unit is configured to perform Fourier transform on the second envelope feature to obtain an envelope frequency domain energy map; and perform Fourier transform on the fused RR interval value to obtain an RR interval frequency domain energy map.

[0059] Optionally, in combination with the second aspect, in a possible implementation, the respiratory rate processing device further includes:

[0060] The PPG signal is transformed from the time domain to the frequency domain to obtain the PPG signal frequency domain energy map.

[0061] Optionally, in combination with the second aspect, in a possible implementation, the extraction module includes:

[0062] The first extraction unit is configured as a first extraction unit, configured to extract the first respiratory rate value, the second respiratory rate value and the third respiratory rate value at each time point on the frequency domain energy map based on the envelope frequency domain energy map, the RR interval frequency domain energy map and the PPG signal frequency domain energy map according to a pre-constructed frequency domain energy value scoring mechanism.

[0063] Optionally, in combination with the second aspect, in a possible implementation, the fusion calculation module includes:

[0064] The second fusion calculation unit is configured to fuse the first respiratory rate value, the second respiratory rate value and the third respiratory rate value to obtain a target respiratory rate value at each time point.

[0065] Optionally, in combination with the second aspect, in a possible implementation, the respiratory rate processing device further includes:

[0066] The reconstruction and acquisition module is configured to acquire a fourth respiratory rate based on EEMD decomposition and reconstruction.

[0067] Optionally, in combination with the second aspect, in a possible implementation, the fusion calculation module includes: a third fusion calculation unit, configured to fuse the first respiratory rate value, the second respiratory rate value, the third respiratory rate value and the fourth respiratory rate value to obtain the target respiratory rate value at each time point.

[0068] Optionally, in combination with the second aspect, in a possible implementation, the reconstruction acquisition module includes:

[0069] The mixing unit is configured to add normally distributed white noise with different amplitudes to the PPG signal each time to obtain at least two mixed signals; the decomposition unit is configured to perform EEMD decomposition on each mixed signal to obtain an IMF component;

[0070] an averaging operation unit, configured to perform an averaging operation on all the IMF components to obtain a target IMF for EEMD decomposition;

[0071] The second extraction unit is configured to extract the respiratory rate value corresponding to the target IMF.

[0072] Optionally, in combination with the second aspect, in a possible implementation, the respiratory rate processing device further includes:

[0073] a first determination module configured to, if in the first measurement mode, determine whether the wearable device is stationary based on the acceleration signal, and activate the PPG signal if the wearable device is stationary;

[0074] The second determination module is configured to, if in the second measurement mode, identify whether the user has fallen asleep according to a sleeping algorithm, and if the user has fallen asleep, activate a PPG signal for a second time interval at intervals of a first time interval; the second time interval is less than the first time interval.

[0075] Optionally, in combination with the second aspect, in a possible implementation, the respiratory rate processing device further includes:

[0076] a signal quality determination module configured to determine the signal quality of the PPG signal based on an interval characteristic of the PPG signal; the interval characteristic of the PPG signal is obtained by calculating the interval difference between adjacent peaks;

[0077] The preprocessing module is configured to determine that the signal quality does not meet the preset conditions when mutations occur continuously in adjacent intervals; if the signal quality meets the preset conditions, the signal is read and the PPG signal is preprocessed to remove the baseline signal to obtain a preprocessed PPG signal.

[0078] Optionally, in combination with the second aspect, in a possible implementation, the second acquisition module includes:

[0079] a second acquiring unit, configured to acquire peaks and troughs of the preprocessed PPG signal;

[0080] The difference unit is configured to interpolate the peaks and troughs to obtain upper and lower envelope features of the green light signal.

[0081] Optionally, in combination with the second aspect, in a possible implementation, the second acquiring unit includes:

[0082] a judgment subunit, configured to judge whether the PPG point of the preprocessed PPG signal is a maximum value or a minimum value;

[0083] If it is a maximum value, determine whether it is the maximum value of the adjacent window length N; if it is a maximum value, it is a valid peak; where N is determined by the frequency of the PPG signal;

[0084] If it is a minimum value, it is determined whether it is the minimum value of the adjacent window length N; if it is a minimum value, it is a valid trough; wherein, the selection of N is determined according to the frequency of the PPG signal.

[0085] In a third aspect of the present application, a wearable device is provided, comprising a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, the method in any possible implementation manner of the first aspect to the first aspect is implemented.

[0086] In a fourth aspect of the present application, a computer-readable storage medium is provided, on which a computer program is stored, characterized in that when the computer program is executed by a processor, the method in any possible implementation of the first aspect to the first aspect is implemented.

[0087] One of the solutions provided in this application includes: obtaining a PPG signal from a PPG sensor; obtaining a first signal feature of at least one dimension based on the PPG signal; processing the first signal feature to obtain a second signal feature, converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map; extracting at least two respiration rate values based on the frequency domain energy map by selecting the frequency domain energy value with the highest score; and fusing the at least two respiration rate values to obtain a target respiration rate value. As an embodiment of the present invention, firstly, two or more first signal features of at least one dimension are obtained, and then, by obtaining at least two frequency domain energy maps corresponding to the first signal feature, the respiration rate value corresponding to the highest-scoring frequency domain energy value is obtained for each frequency domain energy map, ensuring the obtained respiration rate value, and then fusing the at least two respiration rate values to finally obtain the target respiration rate value, thereby improving the accuracy of the respiration rate measurement value. This avoids deviations in the accuracy of the respiration rate measurement value caused by errors in a certain first signal feature due to external factors. BRIEF DESCRIPTION OF THE DRAWINGS

[0088] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0089] Figure 1 A schematic diagram illustrating the waveform modulation effect of the respiratory process on the PPG signal provided in an embodiment of the present application;

[0090] Figure 2 A flow chart of a respiratory rate data processing method provided in an embodiment of the present application;

[0091] Figure 3 A schematic diagram of an upper envelope frequency domain energy diagram provided in an embodiment of the present application;

[0092] Figure 4 A schematic diagram of the structure of a respiratory rate data processing device provided in an embodiment of the present application;

[0093] Figure 5 A schematic diagram of a module of a wearable device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0094] The following will be combined with the accompanying drawings to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations.

[0095] Therefore, the following detailed description of the embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention as claimed, but is merely intended to represent selected embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0096] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0097] The respiratory rate is the number of breaths taken per unit time. The amount of light absorbed by the human body's muscles and venous blood is constant, but as the heart beats, the volume of arterial blood absorbs light in a cyclical pattern. When the heart beats strongly and the blood content in the blood vessels is high, the blood absorbs more light, resulting in a lower intensity of projected or reflected light. When the heart beats weakly and the blood content in the blood vessels is low, the blood absorbs less light, resulting in a higher intensity of projected or reflected light. The velocity of blood flow changes with the respiratory rate, and the intensity of the received PPG signal also changes. Therefore, the respiratory rate can be determined from the PPG signal.

[0098] In a calm state, the normal respiratory rate for adults is approximately 16 to 20 breaths per minute. Women's respiratory rate is slightly higher, on average, by 2 to 3 breaths per minute than men's. The fastest respiratory rate for a newborn is approximately 40 to 50 breaths per minute, while for children under one year old, it is approximately 30 to 40 breaths per minute. Children over seven years old generally have an adult respiratory rate.

[0099] Specifically, the principle of measuring respiratory rate is: use a light-emitting diode to illuminate the measured part, and then use a photodiode to receive the transmitted / reflected light and convert the optical signal into an electrical signal.

[0100] See Figure 1 , the modulation effects of the respiratory process on the PPG waveform include baseline modulation (BW) and pulse amplitude modulation (AM). Baseline modulation is that during the entire respiratory cycle, the changing intrathoracic pressure will cause venous blood return. The intrathoracic pressure will decrease during inspiration, causing a decrease in central venous pressure and an increase in venous return. The exhalation process is the opposite of the inhalation process, that is, the intrathoracic pressure will increase during exhalation, causing an increase in central venous pressure and a decrease in venous return. When more blood continues to be shunted to the low-pressure venous system, the modulation effect will act on the baseline (the baseline is No mod), causing the baseline waveform to move up and down. Pulse amplitude modulation is due to the change in intrathoracic pressure during inspiration. With each ventricular beat, the blood stroke volume will decrease, causing the respiratory pulse amplitude to decrease, such as Figure 1 As shown in the AM waveform, the amplitude of some respiratory waves is reduced compared to the baseline.

[0101] As an embodiment of the present invention, refer to Figure 2 , provides a respiratory rate processing method for use in wearable devices, the method specifically includes:

[0102] S101: Acquire a PPG signal from a PPG sensor.

[0103] The PPG sensor includes one or more green LEDs and a photodetector. Green light was chosen as the light source based on the following characteristics: skin melanin absorbs a significant amount of shorter wavelengths; moisture in the skin also absorbs a significant amount of UV and IR light; green (500nm) and yellow (600nm) light that enters skin tissue is largely absorbed by red blood cells; red light and light near the IR wavelengths pass through skin tissue more easily than other wavelengths; blood absorbs more light than other tissues; and green (green-yellow) light is absorbed by both oxyhemoglobin and deoxyhemoglobin compared to red light. Therefore, both green and red light can be used as measurement light sources. Green light provides a better signal and a better signal-to-noise ratio than other light sources, so green light is used in this example. However, considering varying skin conditions (skin color, sweat), the PPG sensor, in a preferred embodiment, includes one or more green LEDs, one or more infrared LEDs, and a photodetector, enabling automatic use and switching among green, red, and IR light sources depending on the situation. For example, in a high-temperature environment, when a user is sweating profusely in the gym, the moisture on the skin surface increases. Since more green light has been absorbed, it is more difficult to detect the green light reflected under the skin. The green light can be automatically turned off and switched to infrared light to adapt to various scenarios and ensure the accuracy of the measured respiratory rate.

[0104] S102: Acquire a first signal feature of at least one dimension based on the PPG signal.

[0105] The dimension can be understood as a type. The number of first signal features is at least two, for example, it can include an upper envelope and a lower envelope. It can be understood that the upper envelope and the lower envelope both belong to the first signal features of the envelope dimension.

[0106] S103: Process the first signal feature to obtain a second signal feature, and convert the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map.

[0107] The processing method may be filtering, including but not limited to IIR, FIR, or sliding filtering. After filtering, the second signal feature is placed within a respiratory rate band of 3 to 50 breaths / min. Then, the second signal feature is Fourier transformed to obtain a frequency domain energy map, namely, an upper envelope frequency domain energy map and a lower envelope frequency domain energy map.

[0108] S104: Based on the frequency domain energy map, extract at least two respiratory rate values by selecting the frequency domain energy value with the highest score.

[0109] In this embodiment, for the upper envelope frequency domain energy graph in S103, the upper envelope frequency domain energy value with the highest score is selected, and then the respiration rate value corresponding to the upper envelope frequency domain energy value is extracted. For the lower envelope frequency domain energy graph in S103, the lower envelope frequency domain energy value with the highest score is selected, and then the respiration rate value corresponding to the lower envelope frequency domain energy value is extracted. It will be understood that each time point has a corresponding respiration rate value. Therefore, in this embodiment, for each time point, there are two respiration rate values: one obtained by the upper envelope and the other obtained by the lower envelope.

[0110] S105: Perform fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value.

[0111] In this step, for each time point, the two respiratory rate values obtained in S104 are fused to obtain the target respiratory rate value corresponding to each time point. This prevents deviations in the target respiratory rate value caused by errors in the upper or lower envelope, thereby ensuring the accuracy of the target respiratory rate value.

[0112] As an embodiment of the present invention, two or more first signal features in at least one dimension are first obtained. Then, at least two frequency domain energy maps corresponding to the first signal features are obtained, and a corresponding respiratory rate value is obtained for each frequency domain energy map. Then, the at least two respiratory rate values are fused and calculated to finally obtain a target respiratory rate value, thereby improving the accuracy of the respiratory rate measurement value. This avoids deviations in the accuracy of the respiratory rate measurement value caused by errors in a particular first signal feature due to external factors.

[0113] As a preferred embodiment of the present invention, S103 specifically includes:

[0114] S201: Filter the upper envelope and the lower envelope respectively to obtain the upper envelope and the lower envelope in the respiratory frequency band.

[0115] S202: Perform Fourier transform on the upper envelope and the lower envelope respectively to obtain an upper envelope frequency domain energy map and a lower envelope frequency domain energy map.

[0116] In this embodiment, two graphs are obtained: an upper envelope frequency domain energy graph and a lower envelope frequency domain energy graph.

[0117] Based on the previous embodiment, as a preferred embodiment of the present invention, S104 specifically includes:

[0118] S301: Calculate the score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy graph and the score of the lower envelope energy value of the lower envelope frequency domain energy graph by the following formula:

[0119] Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value;

[0120] Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph. Figure 3 , shows a schematic diagram of the upper envelope frequency domain energy diagram, where the horizontal axis is the time point and the vertical axis is the frequency value. The clearer curve is the frequency with the highest frequency domain energy value score, and each frequency value corresponds to the respiratory rate value.

[0121] S302: According to the frequency domain energy value score, determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the lower envelope frequency domain energy value with the highest score.

[0122] Based on the previous embodiment, as a preferred embodiment of the present invention, S104 specifically includes:

[0123] For each time point, the upper envelope frequency domain energy value with the highest score and the lower envelope frequency domain energy value with the highest score are fused and calculated to obtain the target respiratory rate value corresponding to each time point.

[0124] As an embodiment of the present invention, the first signal feature may include an upper envelope, a lower envelope, and an RR interval, and S102 includes:

[0125] First signal features in at least two dimensions are acquired based on the PPG signal, where the first signal features include an upper envelope, a lower envelope, and an RR interval.

[0126] In the embodiment of the present invention, the upper envelope and the lower envelope are used as one dimension, and the RR interval is used as another dimension. The RR interval value is the interval value between two adjacent peaks (or adjacent troughs).

[0127] Based on the previous embodiment, as a preferred embodiment of the present invention, S103 includes:

[0128] S401: Filter the first envelope feature to obtain the second envelope feature; preprocess the RR interval value to obtain a fused RR interval value.

[0129] S402: Perform Fourier transform on the second envelope feature to obtain an envelope frequency domain energy map; perform Fourier transform on the fused RR interval value to obtain an RR interval frequency domain energy map.

[0130] As a preferred embodiment of the present invention, it also includes:

[0131] S403: Transform the PPG signal from the time domain to the frequency domain to obtain a PPG signal frequency domain energy map.

[0132] Based on the above S401, S402 and S403, S104 specifically includes:

[0133] S501: Based on the envelope frequency domain energy map, the RR interval frequency domain energy map and the PPG signal frequency domain energy map, according to a pre-constructed frequency domain energy value scoring mechanism, the first respiratory rate value, the second respiratory rate value and the third respiratory rate value at each time point on the frequency domain energy map are respectively extracted.

[0134] Based on the above embodiment, S105 specifically includes:

[0135] The first respiratory rate value, the second respiratory rate value, and the third respiratory rate value are fused to obtain a target respiratory rate value at each time point.

[0136] As an embodiment of the present invention, the first respiratory rate obtained from the upper envelope and the lower envelope, the second respiratory rate obtained from the RR interval, and the third respiratory rate obtained from the PPG signal are fused, that is, the weights of the first respiratory rate, the second respiratory rate and the third respiratory rate are adaptively adjusted so that the weight of the large error is small and less than the preset threshold, and the weight of the small error is large and greater than the preset threshold. In this way, fusion based on multiple respiratory rates can be achieved, so that the target respiratory rate value achieves maximum accuracy.

[0137] As an embodiment of the present invention, the following steps are also included:

[0138] S502: Obtain a fourth respiratory rate based on EEMD decomposition and reconstruction.

[0139] As an embodiment of the present invention, S502 may specifically include the following steps:

[0140] S601: Adding normally distributed white noise with different amplitudes to the PPG signal each time to obtain at least two mixed signals.

[0141] S602: Perform EEMD decomposition on each mixed signal to obtain IMF components.

[0142] S603: Perform an average operation on all the IMF components to obtain a target IMF of EEMD decomposition.

[0143] S604: Extracting the respiratory rate value corresponding to the target IMF.

[0144] Based on the above S501 and S502 embodiments, S105 includes:

[0145] The first respiratory rate value, the second respiratory rate value, the third respiratory rate value and the fourth respiratory rate value are fused to obtain a target respiratory rate value at each time point.

[0146] As an embodiment of the present invention, the first respiratory rate obtained from the upper envelope and the lower envelope, the second respiratory rate obtained from the RR interval, the third respiratory rate obtained from the PPG signal, and the fourth respiratory rate obtained based on EEMD decomposition and reconstruction are fused, that is, the weights of the first respiratory rate, the second respiratory rate, the third respiratory rate and the fourth respiratory rate are adaptively adjusted, so that the weight of the large error is small and less than the preset threshold, and the weight of the small error is large and greater than the preset threshold. In this way, fusion based on multiple respiratory rates can be achieved, and the target respiratory rate value can further achieve maximum accuracy.

[0147] It should be noted that the above-mentioned respiratory rate processing method can be applied to wearable devices, such as wristbands and watches. During application, this method can include, but is not limited to, two usage scenarios. First, single-click respiratory rate measurement. This means that when a user clicks on the respiratory rate measurement option on the wearable device's graphical interface, the user's respiratory rate is measured. Second, sleeping respiratory rate measurement. This means that the user's respiratory rate is measured while the user is asleep.

[0148] As an embodiment of the present invention, before S101, the method further includes:

[0149] S106: If the wearable device is in the first measurement mode, determine whether it is stationary according to the acceleration signal, and if it is stationary, turn on the PPG signal.

[0150] The first measurement mode may be a click breathing rate measurement mode.

[0151] S107: If the second measurement mode is selected, whether the user has fallen asleep is determined based on a sleeping algorithm. If the user has fallen asleep, a PPG signal is activated for a second time interval every first time interval; the second time interval is shorter than the first time interval.

[0152] The second measurement mode may be a sleep breathing measurement mode. The sleep onset algorithm may be implemented through ACC amplitude recognition, which is not limited here. For ease of illustration, the second time period may be three minutes, and the first time period may be one minute. If the user is determined to be asleep, a three-minute green light signal is activated every ten minutes for signal extraction.

[0153] As an embodiment of the present invention, if the PPG signal is a green light signal, then before S101, the method further includes:

[0154] S108: Determine the signal quality of the PPG signal according to an interval feature of the PPG signal; the interval feature of the PPG signal is obtained by the interval difference between adjacent peaks.

[0155] S109: When mutations occur continuously in adjacent intervals, it is determined that the signal quality does not meet the preset conditions; if the signal quality meets the preset conditions, the signal is read, and the PPG signal is preprocessed to remove the baseline signal to obtain a preprocessed PPG signal.

[0156] Based on the above corresponding embodiments of S108 and S109, as an embodiment of the present invention, S102 specifically includes:

[0157] S701: Acquire peaks and troughs of the preprocessed PPG signal.

[0158] S702: Interpolate the peaks and troughs to obtain upper and lower envelope features of the green light signal.

[0159] The interpolation method may include but is not limited to cubic spline interpolation, discrete smoothing interpolation, etc.

[0160] As an embodiment of the present invention, S701 specifically includes:

[0161] S801: Determine whether the PPG point of the preprocessed PPG signal is a maximum value or a minimum value.

[0162] S802: If it is a maximum value, determine whether it is the maximum value of the adjacent window length N; if it is a maximum value, it is a valid peak; wherein the selection of N is determined according to the frequency of the PPG signal.

[0163] S803: If it is a minimum value, determine whether it is the minimum value of the adjacent window length N; if it is a minimum value, it is a valid trough; wherein the selection of N is determined according to the frequency of the PPG signal.

[0164] For steps S801 to S803, determine whether the preprocessed PPG point is larger than its adjacent points, then determine whether it is the maximum value within the adjacent window length N. If so, it is a valid peak. Determine whether the preprocessed PPG point is smaller than its adjacent points, then determine whether it is the minimum value within the adjacent window length N. If so, it is a valid trough. The choice of N is determined based on the PPG signal frequency. This method is highly resistant to noise and easy to calculate.

[0165] refer to Figure 4 As one embodiment of the present invention, a respiratory rate processing device 40 is provided, the device comprising:

[0166] A first acquisition module 410 is configured to acquire a PPG signal from a PPG sensor;

[0167] A second acquisition module 420 is configured to acquire a first signal feature of at least one dimension based on the PPG signal;

[0168] a conversion module 430 configured to process the first signal feature to obtain a second signal feature, and convert the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map;

[0169] An extraction module 440 is configured to extract at least two respiratory rate values by selecting the frequency domain energy value with the highest score based on the frequency domain energy map;

[0170] The fusion calculation module 450 is configured to perform fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value.

[0171] Optionally, in a possible implementation manner, the first signal feature includes an upper envelope and a lower envelope.

[0172] Optionally, in a possible implementation, the conversion module includes:

[0173] The first filtering unit is configured to filter the upper envelope and the lower envelope respectively to obtain the upper envelope and the lower envelope in the respiratory frequency band; the first Fourier transform unit is configured to perform Fourier transform on the upper envelope and the lower envelope respectively to obtain the upper envelope frequency domain energy map and the lower envelope frequency domain energy map.

[0174] Optionally, in a possible implementation, the extraction module includes:

[0175] A calculation unit is configured to calculate the score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy map and the score of the lower envelope energy value of the lower envelope frequency domain energy map by the following formula,

[0176] Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value;

[0177] Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph;

[0178] The determination unit is configured to determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score based on the frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the breathing rate value of the lower envelope frequency domain energy value with the highest score.

[0179] Optionally, in a possible implementation, the fusion calculation module includes:

[0180] The first fusion calculation unit is configured to, for each time point, perform fusion calculation on the upper envelope frequency domain energy value with the highest score and the lower envelope frequency domain energy value with the highest score to obtain a target respiratory rate value corresponding to each time point.

[0181] Optionally, in a possible implementation, the second obtaining module includes:

[0182] The first acquisition unit is configured to acquire first signal features of at least two dimensions based on the PPG signal, where the first signal features include an upper envelope, a lower envelope, and an RR interval.

[0183] Optionally, in a possible implementation, the conversion module includes:

[0184] The second filtering unit is configured to filter the first envelope feature to obtain a second envelope feature; and preprocess the RR interval value to obtain a fused RR interval value;

[0185] The second Fourier transform unit is configured to perform Fourier transform on the second envelope feature to obtain an envelope frequency domain energy map; and perform Fourier transform on the fused RR interval value to obtain an RR interval frequency domain energy map.

[0186] Optionally, in a possible implementation, the respiratory rate processing device further includes:

[0187] The PPG signal is transformed from the time domain to the frequency domain to obtain the PPG signal frequency domain energy map.

[0188] Optionally, in a possible implementation, the extraction module includes:

[0189] The first extraction unit is configured as a first extraction unit, configured to extract the first respiratory rate value, the second respiratory rate value and the third respiratory rate value at each time point on the frequency domain energy map based on the envelope frequency domain energy map, the RR interval frequency domain energy map and the PPG signal frequency domain energy map according to a pre-constructed frequency domain energy value scoring mechanism.

[0190] Optionally, in a possible implementation, the fusion calculation module includes:

[0191] The second fusion calculation unit is configured to fuse the first respiratory rate value, the second respiratory rate value and the third respiratory rate value to obtain a target respiratory rate value at each time point.

[0192] Optionally, in a possible implementation, the respiratory rate processing device further includes:

[0193] The reconstruction and acquisition module is configured to acquire a fourth respiratory rate based on EEMD decomposition and reconstruction.

[0194] Optionally, in a possible implementation, the fusion calculation module includes: a third fusion calculation unit, configured to fuse the first respiratory rate value, the second respiratory rate value, the third respiratory rate value and the fourth respiratory rate value to obtain a target respiratory rate value at each time point.

[0195] Optionally, in a possible implementation, the reconstruction acquisition module includes:

[0196] The mixing unit is configured to add normally distributed white noise with different amplitudes to the PPG signal each time to obtain at least two mixed signals; the decomposition unit is configured to perform EEMD decomposition on each mixed signal to obtain an IMF component;

[0197] an averaging operation unit, configured to perform an averaging operation on all the IMF components to obtain a target IMF for EEMD decomposition;

[0198] The second extraction unit is configured to extract the respiratory rate value corresponding to the target IMF.

[0199] Optionally, in a possible implementation, the respiratory rate processing device further includes:

[0200] a first determination module configured to, if in the first measurement mode, determine whether the wearable device is stationary based on the acceleration signal, and activate the PPG signal if the wearable device is stationary;

[0201] The second determination module is configured to, if in the second measurement mode, identify whether the user has fallen asleep according to a sleeping algorithm, and if the user has fallen asleep, activate a PPG signal for a second time interval at intervals of a first time interval; the second time interval is less than the first time interval.

[0202] Optionally, in a possible implementation, the respiratory rate processing device further includes:

[0203] a signal quality determination module configured to determine the signal quality of the PPG signal based on an interval characteristic of the PPG signal; the interval characteristic of the PPG signal is obtained by calculating the interval difference between adjacent peaks;

[0204] The preprocessing module is configured to determine that the signal quality does not meet the preset conditions when mutations occur continuously in adjacent intervals; if the signal quality meets the preset conditions, the signal is read and the PPG signal is preprocessed to remove the baseline signal to obtain a preprocessed PPG signal.

[0205] Optionally, in a possible implementation, the second obtaining module includes:

[0206] a second acquiring unit, configured to acquire peaks and troughs of the preprocessed PPG signal;

[0207] The difference unit is configured to interpolate the peaks and troughs to obtain upper and lower envelope features of the green light signal.

[0208] Optionally, in a possible implementation, the second acquiring unit includes:

[0209] a judgment subunit, configured to judge whether the PPG point of the preprocessed PPG signal is a maximum value or a minimum value;

[0210] If it is a maximum value, determine whether it is the maximum value of the adjacent window length N; if it is a maximum value, it is a valid peak; where N is determined by the frequency of the PPG signal;

[0211] If it is a minimum value, it is determined whether it is the minimum value of the adjacent window length N; if it is a minimum value, it is a valid trough; wherein, the selection of N is determined according to the frequency of the PPG signal.

[0212] like Figure 5 As shown, the wearable device 100 may include one or more processors 101, a memory 102, a communication module 103, a sensor module 104, a display 105, an audio module 106, a speaker 107, a microphone 108, a camera module 109, a motor 110, a button 111, an indicator 112, a battery 113, and a power management module 114. These components may communicate via one or more communication buses or signal lines.

[0213] The processor 101 is the final execution unit for information processing and program execution. It can run an operating system or application program to execute various functional applications and data processing of the wearable device 100. The processor 101 may include one or more processing units. For example, the processor 101 may include a central processing unit (CPU), a graphics processing unit (GPU), an image signal processor (ISP), a sensor hub processor or a communication processor (CP), an application processor (AP), etc. In some embodiments, the processor 101 may include one or more interfaces. The interface is used to couple peripheral devices to the processor 101 to transmit instructions or data between the processor 101 and the peripheral devices. In an embodiment of the present application, the processor 101 is also used to identify the target motion type corresponding to the motion data collected by the acceleration sensor and the gyroscope sensor, for example, walking / running / cycling / swimming, etc. Specifically, the processor 101 compares the motion waveform characteristics corresponding to the received motion data with the motion waveform characteristics corresponding to the target motion type to identify the target motion type corresponding to the motion data. The processor 101 is also used to determine whether the motion data within a preset time period all meet the preset motion intensity requirements associated with the target motion type. When it is determined that the motion data within the preset time period all meet the preset motion intensity requirements associated with the target motion type, the processor 101 controls the activation of the sensor group associated with the target motion type.

[0214] The memory 102 can be used to store computer executable program code, and the executable program code includes instructions. The memory 102 may include a program storage area and a data storage area. Among them, the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc. The data storage area may store data created during the use of the wearable device 100, such as the user's exercise parameters for each exercise, such as the number of steps, stride, pace, heart rate, respiratory rate, blood sugar concentration, energy consumption (calories), etc. The memory may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, a universal flash storage (UFS), etc. In an embodiment of the present application, the memory 102 is capable of storing sensor waveform regularity characteristic data corresponding to target sports such as walking, running, cycling, and swimming.

[0215] The communication module 103 can support the wearable device 100 to communicate with the network and mobile terminals through wireless communication technology. The communication module 103 converts electrical signals into electromagnetic signals for transmission, or converts received electromagnetic signals into electrical signals. The communication module 103 may include one or more of a cellular mobile communication module, a short-range wireless communication module, a wireless Internet module, and a location information module. The mobile communication module can send or receive wireless signals based on the technical standards of mobile communications and can use any mobile communication standard or protocol, including but not limited to Global System for Mobile Communications (GSM), Code Division Multiple Access (CDMA), Code Division Multiple Access 2000 (CDMA2000), Wideband CDMA (WCDMA), Time Division Synchronous Code Division Multiple Access (TD-SCDMA), Long Term Evolution (LTE), LTE-A (Long Term Evolution Advanced), etc. The wireless Internet module can send or receive wireless signals via a communication network based on wireless Internet technologies, including wireless LAN (WLAN), Wireless Fidelity (Wi-Fi), Wi-Fi Direct, Digital Living Network Alliance (DLNA), Wireless Broadband (WiBro), etc. The short-range wireless communication module can send or receive wireless signals based on short-range communication technologies, including Bluetooth, radio frequency identification (RFID), infrared data communication (IrDA), ultra-wideband (UWB), ZigBee, near-field communication (NFC), wireless fidelity (Wi-Fi), Wi-Fi direct, wireless USB (wireless universal serial bus), etc. The location information module can obtain the location of the wearable device based on the global navigation satellite system (GNSS), which can include one or more of the global positioning system (GPS), global satellite navigation system (Glonass), Beidou satellite navigation system and Galileo satellite navigation system.

[0216] The sensor module 104 is used to measure physical quantities or detect the operating state of the wearable device 100. The sensor module 104 may include an acceleration sensor 104A, a gyroscope sensor 104B, an air pressure sensor 104C, a magnetic sensor 104D, a biometric sensor 104E, a proximity sensor 104F, an ambient light sensor 104G, a touch sensor 104H, etc. The sensor module 104 may also include a control circuit for controlling one or more sensors included in the sensor module 104.

[0217] The accelerometer 104A can detect the magnitude of acceleration of the wearable device 100 in all directions. When the wearable device 100 is stationary, it can detect the magnitude and direction of gravity. It can also be used to identify the posture of the wearable device 100, enabling applications such as switching between landscape and portrait modes and pedometers. In one embodiment, the accelerometer 104A can be combined with the gyroscope 104B to monitor the user's stride length, cadence, and pace during exercise.

[0218] The gyro sensor 104B can be used to determine the motion posture of the wearable device 100. In some embodiments, the angular velocity of the wearable device 100 around three axes (ie, x, y, and z axes) can be determined by the gyro sensor 104B.

[0219] The air pressure sensor 104C is used to measure air pressure. In some embodiments, the wearable device 100 calculates altitude using the air pressure value measured by the air pressure sensor 104C to assist in positioning and navigation.

[0220] The GPS sensor 104D may be used to record user activity tracks to determine the user's location.

[0221] The biometric sensor 104E is used to measure the user's physiological parameters, including but not limited to a photoplethysmography (PPG) sensor, an ECG sensor, an EMG sensor, a blood glucose sensor, and a temperature sensor. For example, the wearable device 100 can measure the user's heart rate, respiratory rate, and blood pressure data using signals from the photoplethysmography sensor and / or the ECG sensor, and identify the user's blood glucose level based on data generated by the blood glucose sensor. In an embodiment of the present application, the PPG sensor is used to detect the user's heart rate. Specifically, after being turned on, the PPG sensor can continuously detect signal data related to the user's heart rate and transmit it to the processor 101, which then calculates the heart rate value using a heart rate algorithm. In an embodiment of the present application, the temperature sensor is used to detect a first temperature of the user's wrist skin. Specifically, after being turned on, the temperature sensor can continuously obtain temperature data of the user's wrist skin and transmit it to the processor 101, which then calculates the corresponding temperature value with a physical meaning using a temperature algorithm based on the electrical signal data from the temperature sensor.

[0222] The proximity sensor 104F is used to detect the presence of an object near the wearable device 100 without any physical contact. In some embodiments, the proximity sensor 104F may include a light-emitting diode (LED) and a light detector. The LED may emit infrared light, and the wearable device 100 uses the light detector to detect reflected light from a nearby object. When the reflected light is detected, it can be determined that an object is near the wearable device 100. The wearable device 100 can use the proximity sensor 104F to detect its wearable state.

[0223] The ambient light sensor 104G is used to sense the brightness of the ambient light. In some embodiments, the wearable device 100 can adaptively adjust the brightness of the display screen according to the sensed ambient light brightness to reduce power consumption.

[0224] The touch sensor 104H is used to detect touch operations applied thereto or in the vicinity thereof and is also referred to as a “touch control device.” The touch sensor 104H may be disposed on the display screen 105 , and the touch sensor 104H and the display screen 105 form a touch screen.

[0225] Display screen 105 is used to display a graphical user interface (UI), which may include graphics, text, icons, videos, or any combination thereof. Display screen 105 may be a liquid crystal display (LCD), an organic light-emitting diode (OLED), or the like. If display screen 105 is a touch screen, it can collect touch signals on or above the surface of display screen 105 and input the touch signals as control signals to processor 101.

[0226] The audio module 106, speaker 107, and microphone 108 provide audio functions between the user and the wearable device 100, such as listening to music or making calls. For example, when the wearable device 100 receives a notification message from a mobile terminal, the processor 101 controls the audio module 106 to output a preset audio signal, and the speaker 107 to emit a sound to alert the user. The audio module 106 converts the received audio data into an electrical signal and sends it to the speaker 107, which then converts the electrical signal into sound. Alternatively, the microphone 108 converts the sound into an electrical signal and sends it to the audio module 106, which then converts the audio signal into audio data.

[0227] The camera module 111 is used to capture still images or videos. The camera module 111 may include an image sensor, an image signal processor (ISP), and a digital signal processor (DSP). The image sensor converts optical signals into electrical signals, the ISP converts the electrical signals into digital image signals, and the DSP converts the digital image signals into image signals in a standard format (RGB, YUV). The image sensor may be a charge coupled device (CCD) or a complementary metal-oxide-semiconductor (CMOS).

[0228] The motor 110 can convert electrical signals into mechanical vibrations to produce a vibration effect. The motor 110 can be used for vibration prompts for incoming calls and messages, and can also be used for touch vibration feedback. The button 109 includes a power button, a volume button, etc. The button 109 can be a mechanical button (physical button) or a touch button. The indicator 112 is used to indicate the status of the wearable device 100, for example, to indicate the charging status, power changes, and can also be used to indicate messages, missed calls, notifications, etc. In some embodiments, the wearable device 100 provides vibration feedback after receiving a notification message from a mobile terminal application.

[0229] Battery 113 provides power to the various components of wearable device 100. Power management module 114 manages battery charge and discharge, and monitors parameters such as battery capacity, battery cycle count, and battery health (leakage, impedance, voltage, current, and temperature). In some embodiments, power management module 114 can charge the battery via wired or wireless means.

[0230] It should be understood that in some embodiments, the wearable device 100 may be composed of one or more of the aforementioned components. The wearable device 100 may include more or fewer components than shown, or may combine or separate certain components, or arrange the components differently. The components shown may be implemented in hardware, software, or a combination of software and hardware.

[0231] It should be understood that in some embodiments, the wearable device may be composed of one or more of the aforementioned components. The wearable device may include more or fewer components than shown, or combine or separate certain components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.

[0232] The present application also provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the above-mentioned respiratory rate processing method is implemented.

[0233] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions and operations of the devices, methods and computer program products according to multiple embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a portion of code, and the module, program segment or a portion of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.

[0234] In addition, the functional modules in the various embodiments of the present invention may be integrated together to form an independent part, or each module may exist independently, or two or more modules may be integrated to form an independent part.

[0235] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0236] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A method for processing respiratory rate, characterized in that: The method comprises: Get the PPG signal from the PPG sensor; Acquire a first signal feature of at least one dimension based on the PPG signal; the first signal feature includes an upper envelope and a lower envelope; Processing the first signal feature to obtain a second signal feature, that is, filtering the upper envelope and the lower envelope respectively to obtain an upper envelope and a lower envelope in a respiratory frequency band; Converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map; that is, performing Fourier transform on the upper envelope and the lower envelope respectively to obtain an upper envelope frequency domain energy map and a lower envelope frequency domain energy map; Based on the frequency domain energy map, the frequency domain energy value with the highest score is selected to extract at least two respiratory rate values through the following steps: The score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy graph and the score of the lower envelope energy value of the lower envelope frequency domain energy graph are calculated by the following formula: Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value; Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph; According to the frequency domain energy value score, determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the respiration rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the respiration rate value of the lower envelope frequency domain energy value with the highest score; The at least two respiratory rate values are fused and calculated to obtain a target respiratory rate value.

2. The processing method according to claim 1, characterized in that The fusing and calculating the at least two respiratory rate values to obtain a target respiratory rate value includes: For each time point, the frequency value corresponding to the upper envelope frequency domain energy value with the highest score and the frequency value corresponding to the lower envelope frequency domain energy value with the highest score are fused and calculated to obtain the target respiratory rate value corresponding to each time point.

3. The processing method according to claim 1, characterized in that The obtaining of a first signal feature of at least one dimension based on the PPG signal includes: First signal features in at least two dimensions are acquired based on the PPG signal, where the first signal features include an upper envelope, a lower envelope, and an RR interval.

4. The processing method according to claim 3, characterized in that The processing of the first signal feature to obtain a second signal feature, and converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map, includes: Filtering the first envelope feature to obtain a second envelope feature; preprocessing the RR interval value to obtain a fused RR interval value; Performing Fourier transform on the second envelope feature to obtain an envelope frequency domain energy map; performing Fourier transform on the fused RR interval value to obtain an RR interval frequency domain energy map.

5. The processing method according to claim 4, characterized in that Also includes: The PPG signal is transformed from the time domain to the frequency domain to obtain the PPG signal frequency domain energy map.

6. The processing method according to claim 5, characterized in that Extracting at least one first respiratory rate value by selecting the frequency domain energy value with the highest score includes: Based on the envelope frequency domain energy map, the RR interval frequency domain energy map and the PPG signal frequency domain energy map, according to a pre-constructed frequency domain energy value scoring mechanism, the first respiratory rate value, the second respiratory rate value and the third respiratory rate value at each time point on the frequency domain energy map are extracted respectively.

7. The processing method according to claim 6, characterized in that Performing a fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value includes: The first respiratory rate value, the second respiratory rate value, and the third respiratory rate value are fused to obtain a target respiratory rate value at each time point.

8. The processing method according to claim 6, characterized in that Also includes: The fourth respiratory rate is obtained based on EEMD decomposition and reconstruction.

9. The processing method according to claim 8, characterized in that Performing a fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value includes: The first respiratory rate value, the second respiratory rate value, the third respiratory rate value and the fourth respiratory rate value are fused to obtain a target respiratory rate value at each time point.

10. The processing method according to claim 8, characterized in that The fourth respiratory rate is obtained based on EEMD decomposition and reconstruction, including: Adding normally distributed white noise with different amplitudes to the PPG signal each time to obtain at least two mixed signals; Perform EEMD decomposition on each mixed signal to obtain the IMF component; Performing an average operation on all the IMF components to obtain the target IMF of EEMD decomposition; Extract the respiratory rate value corresponding to the target IMF.

11. The processing method according to any one of claims 1 to 10, characterized in that: Before obtaining the PPG signal from the PPG sensor, it also includes: If it is the first measurement mode, it is determined whether the wearable device is stationary based on the acceleration signal, and if it is stationary, the PPG signal is turned on; If it is the second measurement mode, whether the user has fallen asleep is identified based on the sleeping algorithm. If the user has fallen asleep, the PPG signal of the second time is turned on every first time interval; the second time is less than the first time.

12. The processing method according to claim 11, characterized in that If the PPG signal is a green light signal, the method further includes: Determining the signal quality of the PPG signal based on an interval characteristic of the PPG signal; wherein the interval characteristic of the PPG signal is obtained by calculating the interval difference between adjacent peaks; When mutations occur continuously in adjacent intervals, it is judged that the signal quality does not meet the preset conditions; if the signal quality meets the preset conditions, the signal is read, and the PPG signal is preprocessed to remove the baseline signal to obtain a preprocessed PPG signal.

13. The processing method according to claim 12, characterized in that Acquiring a first signal feature of at least one dimension based on the PPG signal includes: Obtaining peaks and troughs of the preprocessed PPG signal; Interpolation is performed on the peaks and troughs to obtain upper and lower envelope features of the green light signal.

14. The processing method according to claim 13, characterized in that The obtaining of peaks and troughs of the preprocessed PPG signal includes: Determine whether the PPG point of the preprocessed PPG signal is a maximum value or a minimum value; If it is a maximum value, determine whether it is the maximum value of the adjacent window length N; if it is a maximum value, it is a valid peak; where N is determined by the frequency of the PPG signal; If it is a minimum value, it is determined whether it is the minimum value of the adjacent window length N; if it is a minimum value, it is a valid trough; wherein, the selection of N is determined according to the frequency of the PPG signal.

15. A respiratory rate processing device, characterized in that: The device comprises: A first acquisition module is configured to acquire a PPG signal from a PPG sensor; A second acquisition module is configured to acquire a first signal feature of at least one dimension based on the PPG signal; the first signal feature includes an upper envelope and a lower envelope; a conversion module configured to process the first signal feature to obtain a second signal feature, namely, filtering the upper envelope and the lower envelope respectively to obtain an upper envelope and a lower envelope in the respiratory frequency band; converting the second signal feature from the time domain to the frequency domain to obtain a frequency domain energy map, namely, performing Fourier transform on the upper envelope and the lower envelope respectively to obtain an upper envelope frequency domain energy map and a lower envelope frequency domain energy map; The extraction module is configured to extract at least two respiratory rate values based on the frequency domain energy map by selecting the frequency domain energy value with the highest score; and calculate the score of the upper envelope frequency domain energy value of the upper envelope frequency domain energy map and the score of the lower envelope energy value of the lower envelope frequency domain energy map using the following formula: Frequency domain energy score = frequency continuity score * first weight value + energy score * second weight value; Among them, the frequency continuity score is that if the absolute difference between the frequency value at the current time point and the frequency value at the previous time point is less than the threshold, the continuity score is high, otherwise the score is low; the energy value score is to sort the energy values, the greater the energy, the higher the score; the first weight value and the second weight value are selected according to the signal-to-noise ratio of the frequency corresponding to each time point on the frequency energy graph; According to the frequency domain energy value score, determine the upper envelope frequency domain energy value with the highest score corresponding to each time point and extract the respiration rate value of the upper envelope frequency domain energy value with the highest frequency domain energy value score; determine the lower envelope frequency domain energy value with the highest score corresponding to each time point and extract the respiration rate value of the lower envelope frequency domain energy value with the highest score; The fusion calculation module is configured to perform fusion calculation on the at least two respiratory rate values to obtain a target respiratory rate value.

16. A wearable device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program that can be executed by the processor, and when the computer program is executed by the processor, the method according to any one of claims 1 to 14 is implemented.

17. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 14 is implemented.

Images