Multi-frequency based bioelectrical impedance for body fat precise measurement method and system

Abstract

The application relates to a multi-frequency-based bioelectric impedance body fat precise measurement method and system, which comprises the following steps: receiving a mode selection instruction issued by a user to determine a current working mode; if the working mode is a rapid measurement mode, a single reference frequency and a reference current intensity are applied to stimulate a single set of bioelectric impedance data and calculate a rapid body fat rate result; if the working mode is a precise measurement mode, a complete electrode system is used to obtain bioelectric impedance data under multiple preset frequencies and multiple preset current intensities, the data are subjected to signal processing, impedance amplitude and phase information are extracted, an impedance characteristic curve is constructed, feature parameters related to the body fat rate are extracted from the impedance characteristic curve, the body fat rate is calculated based on the feature parameters, and a final body fat rate result of the precise measurement mode is obtained. The method can realize precise measurement of the body fat rate and improve the accuracy and reliability of the measurement result.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of health monitoring and information technology, in particular to a multi-frequency-based bioelectrical impedance body fat accurate measurement method and system. BACKGROUND

[0002] In the field of human health monitoring technology, body fat rate as an important indicator reflecting human health status, its accurate measurement is of great significance. At present, bioelectrical impedance analysis method is a commonly used body fat measurement method, which is based on the different electrical impedance characteristics of different tissues of human body (such as fat, muscle, water, etc.), by applying a specific current to the human body, measuring the electrical impedance value of the human body, and calculating the body fat rate according to the preset algorithm.

[0003] The existing bioelectrical impedance body fat measurement method usually uses single frequency current excitation, this method although simple operation, fast measurement, but because the single frequency current can only reflect the electrical impedance information of a certain aspect of the human body, it cannot comprehensively and accurately reflect the fat distribution and content of the human body, resulting in lower accuracy of the measurement result. SUMMARY

[0004] The main purpose of the present application is to provide a multi-frequency-based bioelectrical impedance body fat accurate measurement method and system, which can realize accurate measurement of body fat rate and improve the accuracy and reliability of the measurement result.

[0005] To achieve the above purpose, the embodiment of the present application provides a multi-frequency-based bioelectrical impedance body fat accurate measurement method, which comprises the following steps:

[0006] Receiving the mode selection instruction issued by the user through the body fat scale human-computer interaction interface or the paired mobile terminal, determining the current working mode, the electrode system of the body fat scale includes at least four electrodes;

[0007] If the current mode is the fast measurement mode, a preset single reference frequency and reference current intensity excitation are applied through at least two electrodes of the body fat scale to obtain a single set of bioelectrical impedance data, and a fast body fat rate result is calculated based on the single set of bioelectrical impedance data;

[0008] If the current working mode is the accurate measurement mode, the complete electrode system composed of at least four electrodes of the body fat scale is used to obtain the bioelectrical impedance data of the measured object under multiple preset frequencies and multiple preset current intensities, wherein the multiple preset frequencies cover a wide frequency band range from low frequency to high frequency, the low frequency is used to measure the extracellular fluid impedance, and the high frequency is used to penetrate the cell membrane to measure the total body fluid impedance;

[0009] performing signal processing on the bioelectrical impedance data, extracting impedance amplitude and phase information corresponding to each frequency and each current intensity, and storing the impedance amplitude and phase information as a first data set and a second data set, respectively;

[0010] According to the first data set and the second data set, the impedance modulus and impedance angle values of the object to be measured at different frequencies are calculated, and an impedance characteristic curve is constructed based on the impedance modulus and impedance angle values;

[0011] From the impedance characteristic curve, a feature parameter related to the body fat rate is extracted, including the impedance modulus change rate and the impedance angle value distribution range;

[0012] Based on the feature parameter, the body fat rate of the object to be measured is calculated to obtain the final body fat rate result of the accurate measurement mode.

[0013] In summary, by receiving the mode selection instruction issued by the user to determine the current working mode, the technical solution of the present application can meet the different measurement needs of the user. In the fast measurement mode, a single reference frequency and reference current intensity are applied to excite a single set of bioelectrical impedance data and calculate the fast body fat rate result, which can realize fast measurement. In the accurate measurement mode, the complete electrode system of the body fat scale is used to obtain bioelectrical impedance data of the object to be measured at multiple preset frequencies and multiple preset current intensities. These data cover a wide frequency band from low frequency to high frequency, and can fully reflect the electrical impedance characteristics of different tissues of the human body. The bioelectrical impedance data is processed to extract impedance amplitude and phase information, and an impedance characteristic curve is constructed, and then feature parameters related to the body fat rate are extracted therefrom, and finally the body fat rate is calculated based on these feature parameters, which can fully exploit the body fat information contained in the multi-frequency bioelectrical impedance data, effectively improving the accuracy and reliability of the body fat rate measurement. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a scene diagram of the body fat accurate measurement method based on multi-frequency bioelectrical impedance in the embodiment of the present application;

[0015] Figure 2 is a flowchart of the body fat accurate measurement method based on multi-frequency bioelectrical impedance provided in the embodiment of the present application;

[0016] Figure 3 is a flowchart of the bioelectrical impedance data processing provided in the embodiment of the present application;

[0017] Figure 4 is a flowchart of the feature extraction provided in the embodiment of the present application;

[0018] Figure 5 is a flowchart of the non-linear correction processing provided in the embodiment of the present application;

[0019] Figure 6 A flowchart of the feature weight coefficient determination provided by the embodiment of the present application is shown in the figure.

[0020] Figure 7 A flowchart of the final body fat rate calculation provided by the embodiment of the present application is shown in the figure.

[0021] Figure 8 A flowchart of the impedance angle value distribution range calculation provided by the embodiment of the present application is shown in the figure.

[0022] Figure 9 A structural diagram of the body fat precise measurement system based on multi-frequency bioelectrical impedance provided by the embodiment of the present application is shown in the figure. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application.

[0024] The embodiment of the present application provides a body fat precise measurement method based on multi-frequency bioelectrical impedance, which will be described in detail below.

[0025] In the embodiment of the present application, the body fat precise measurement method based on multi-frequency bioelectrical impedance is a method for precisely measuring the body fat rate of a human body by using the bioelectrical impedance principle through a body fat scale. The method involves applying multiple preset frequencies and multiple preset current intensities to the human body, obtaining bioelectrical impedance data, and processing and analyzing these data to extract feature parameters related to the body fat rate, and finally calculating the body fat rate. Specifically, the electrode system of the body fat scale includes at least four electrodes, which can obtain bioelectrical impedance data in different excitation modes according to the user's selected working mode. In the fast measurement mode, a single reference frequency and reference current intensity excitation are applied to obtain a single set of bioelectrical impedance data; in the precise measurement mode, the complete electrode system is used to obtain bioelectrical impedance data under multiple preset frequencies and multiple preset current intensities. The obtained data are processed to extract impedance amplitude and phase information, and an impedance characteristic curve is constructed to extract feature parameters such as impedance modulus change rate and impedance angle value distribution range, and the body fat rate is calculated based on these feature parameters. The whole process can fully utilize multi-frequency bioelectrical impedance data, effectively improving the accuracy and reliability of body fat rate measurement.

[0026] As shown in Figure 1 , a body fat precise measurement method based on multi-frequency bioelectrical impedance is provided, which can include a body fat scale, a paired mobile terminal, and a body fat measurement platform in the body fat measurement scene; wherein the body fat scale, the body fat measurement platform, and the paired mobile terminal are connected through wireless communication modes such as Bluetooth.

[0027] Taking home health monitoring as an example, such as Figure 1 As shown, the intelligent body fat scale consists of a scale body, an electrode system, a human-computer interface, and a control circuit. The electrode system includes at least four electrodes distributed on the surface of the scale body, used to contact the human body, apply excitation current, and measure the response voltage.

[0028] The body fat scale can be a four-electrode body fat scale, which includes an electrode system, a display unit, a wireless communication unit, and a power supply unit. The electrode system includes four electrode pads disposed on the surface of the scale body. In one embodiment, as... Figure 1 As shown, the body fat scale can also be an eight-electrode body fat scale. An eight-electrode body fat scale includes an electrode system, a display unit, a wireless communication unit, and a power supply unit. The electrode module in the electrode system includes eight electrode pads, with four pads positioned on the scale body surface and the other four on the handle surface. It should also be understood that the number of electrode pads on the body fat scale is not limited. Figure 1 This body fat scale is an eight-electrode body fat scale 200. Structurally, the eight-electrode body fat scale 200 includes a handle 210 and a scale body 220. The electrode system includes eight electrode plates 230, with four electrode plates 230 disposed on the surface of the scale body 220 and the other four electrode plates 230 disposed on the surface of the handle 210.

[0029] When a user stands on the body fat scale, they can select either a quick measurement mode or a precise measurement mode via the scale's user interface or a paired mobile terminal. In quick measurement mode, the scale applies a preset single reference frequency and reference current intensity to the feet via at least two electrodes. For example, the reference frequency can be set to 50kHz, and the reference current intensity can be set to 0.8mA. Simultaneously, the response voltage generated on the electrode system is measured, and a single set of bioelectrical impedance data is calculated based on the excitation current and response voltage values. The impedance modulus is extracted from this data and matched against a pre-stored mapping relationship to obtain the corresponding quick body fat percentage. This quick measurement mode is suitable for users who only need a general understanding of their body fat percentage; it is fast and easy to operate.

[0030] If the user selects the precise measurement mode, the body fat scale acquires the bioelectrical impedance data of the subject to be measured at multiple preset frequencies and multiple preset current intensities through a complete electrode system composed of at least four electrodes. The multiple preset frequencies cover a wide frequency band range from low frequency to high frequency, for example, the low frequency can be set to 1 kHz for measuring the extracellular fluid impedance; the high frequency can be set to 1 MHz for measuring the total body fluid impedance by penetrating the cell membrane. The preset current intensity can be adjusted according to different frequencies to ensure accurate measurement of the bioelectrical impedance of the human body. Then, the body fat scale can send the bioelectrical impedance data to the body fat monitoring platform for body fat rate calculation.

[0031] The body fat measurement platform is used for signal processing of the bioelectrical impedance data, extracting the impedance amplitude and phase information corresponding to each frequency and each current intensity, and storing the impedance amplitude and phase information as a first data set and a second data set respectively. According to the first data set and the second data set, the impedance modulus and impedance angle values of the subject to be measured at different frequencies are calculated, and the impedance characteristic curve is constructed based on the impedance modulus and impedance angle values; the feature parameters related to the body fat rate are extracted from the impedance characteristic curve, including the impedance modulus change rate and the impedance angle value distribution range; based on the feature parameters, the body fat rate of the subject to be measured is calculated to obtain the final body fat rate result of the precise measurement mode, and then the final body fat rate result is sent to the body fat scale.

[0032] Finally, the body fat scale displays the measured body fat rate result to the user through the human-computer interaction interface, or sends the result to the paired mobile terminal for the user to view and record. The user can understand his own body fat condition according to the measurement result, and make reasonable diet and exercise plan to maintain a healthy body state.

[0033] In an embodiment, the steps or functions realized by the above body fat monitoring platform can also be realized or executed on the body fat scale side.

[0034] Reference Figure 2 , Figure 2 is a flowchart of a body fat precise measurement method based on multi-frequency bioelectrical impedance provided by the embodiments of the present application. The execution subject of the method can be a computer device (such as a body fat scale, a body fat monitoring platform), such as a server, etc. The body fat precise measurement method based on multi-frequency bioelectrical impedance provided by the embodiments of the present application specifically includes:

[0035] S10: receiving a mode selection instruction issued by the user through the human-computer interaction interface of the body fat scale or the paired mobile terminal, determining the current working mode, and the electrode system of the body fat scale includes at least four electrodes.

[0036] In the embodiments of the present application, the body fat scale human-computer interaction interface is an interface for the user to interact with the body fat scale, which can include a display screen, a button and the like, and the user can select a measurement mode through button operation. The paired mobile terminal is a smart device such as a mobile phone or a tablet computer that is paired and connected to the body fat scale through wireless communication, and the user can install a corresponding application program on the mobile terminal and issue a mode selection instruction through the application program. The electrode system is an important component of the body fat scale, which is used to contact the human body, apply an excitation current and measure a response voltage. The design of at least four electrodes can meet the needs of different measurement modes, and two of them can be used in the fast measurement mode, and the complete electrode system can be used in the accurate measurement mode.

[0037] In an embodiment, when the user stands on the body fat scale, the body fat scale automatically starts and enters a waiting instruction state. The user can select a fast measurement mode or an accurate measurement mode through a button on the body fat scale human-computer interaction interface, or open an application program on the paired mobile terminal and select a corresponding measurement mode in the application program. After the body fat scale receives the mode selection instruction issued by the user, the instruction is analyzed and verified to determine the current working mode. This way can provide the user with multiple operation modes, making it convenient for the user to select a suitable measurement mode according to his own needs.

[0038] S20: If the current mode is the fast measurement mode, a preset single reference frequency and reference current intensity excitation are applied through at least two electrodes of the body fat scale to obtain a single set of bioelectrical impedance data, and a fast body fat rate result is calculated based on the single set of bioelectrical impedance data.

[0039] In the embodiments of the present application, the single reference frequency refers to the frequency of the excitation current applied by the body fat scale to the human body in the fast measurement mode, which is a fixed value. The reference current intensity refers to the intensity of the excitation current, which is a preset fixed value. The bioelectrical impedance data refers to the human body electrical impedance value calculated according to Ohm's law by measuring the response voltage generated by the human body when subjected to the excitation current. The single set of bioelectrical impedance data refers to a set of bioelectrical impedance data obtained under the single reference frequency and reference current intensity excitation.

[0040] In an embodiment, after determining that the current working mode is the fast measurement mode, the body fat scale applies a preset single reference frequency and reference current intensity excitation to the feet of the human body through at least two electrodes. For example, the reference frequency can be set to 50 kHz, and the reference current intensity can be set to 0.8 mA. While the excitation current is applied, the response voltage generated on the electrode system is measured synchronously. According to Ohm's law, the electrical impedance value is equal to the voltage value divided by the current value, so that a single set of bioelectrical impedance data is calculated. The impedance modulus value is extracted from the single set of bioelectrical impedance data, and the impedance modulus value is matched according to the pre-stored mapping relationship set to obtain the corresponding fast body fat rate. This way can obtain the approximate result of the body fat rate in a short time, meeting the user's demand for quickly understanding the body fat condition.

[0041] In an embodiment, the single set of bioelectrical impedance data is represented in complex form, including a real part and an imaginary part. According to the modulus calculation formula of the complex number, the impedance modulus value is equal to the square root of the sum of the square of the real part and the square of the imaginary part. For example, the single set of bioelectrical impedance data is represented as Z = a + bj (a is the real part, and b is the imaginary part), and the impedance modulus value |Z| = √(a² + b²). The impedance modulus value is calculated by the formula.

[0042] S30: If the current working mode is the accurate measurement mode, a complete electrode system composed of at least four electrodes of the body fat scale is used to obtain the bioelectrical impedance data of the measured object under multiple preset frequencies and multiple preset current intensities, wherein the multiple preset frequencies cover a wide frequency band range from low frequency to high frequency, the low frequency is used to measure the extracellular fluid impedance, and the high frequency is used to penetrate the cell membrane to measure the total body fluid impedance.

[0043] In the embodiments of the present application, the multiple preset frequencies refer to the frequencies of the excitation current applied by the body fat scale to the human body in the accurate measurement mode, and the frequencies cover a wide frequency band range from low frequency to high frequency. The multiple preset current intensities refer to the different values of the excitation current intensity corresponding to each preset frequency.

[0044] Among them, the extracellular fluid impedance refers to the electrical impedance characteristic of the extracellular fluid of the human body, and the low frequency current mainly reflects the impedance information of the extracellular fluid due to its weak penetration ability. The total body fluid impedance refers to the total electrical impedance characteristic of the intracellular fluid and the extracellular fluid of the human body, and the high frequency current has strong penetration ability and can penetrate the cell membrane to reflect the impedance information of the total body fluid.

[0045] S40: The bioelectrical impedance data is processed, the impedance amplitude and phase information corresponding to each frequency and each current intensity are extracted, and the impedance amplitude and phase information are stored as a first data set and a second data set, respectively.

[0046] In this embodiment, signal processing refers to filtering and transforming the bioelectrical impedance data to remove noise interference and extract useful information. Impedance amplitude refers to the magnitude of the bioelectrical impedance, reflecting the degree to which the human body impedes the excitation current. Phase information refers to the phase angle of the bioelectrical impedance, reflecting the phase relationship between current and voltage. The first data set is a dataset storing impedance amplitude information, and the second data set is a dataset storing phase information.

[0047] S50: Based on the first data set and the second data set, calculate the impedance magnitude and impedance angle of the object under test at different frequencies, and construct an impedance characteristic curve based on the impedance magnitude and impedance angle.

[0048] In this embodiment, the impedance magnitude refers to the magnitude of bioelectrical impedance, which is the absolute value of the impedance amplitude. The impedance angle refers to the phase angle of the bioelectrical impedance, reflecting the phase difference between current and voltage. The impedance characteristic curve describes the changes of the impedance magnitude and impedance angle with frequency. Through the impedance characteristic curve, the variation of human bioelectrical impedance at different frequencies can be observed intuitively.

[0049] In one embodiment, impedance amplitude information is obtained from a first dataset, and phase information is obtained from a second dataset. Based on the impedance amplitude and phase information, the impedance magnitude and impedance angle at different frequencies are calculated using trigonometric functions. An impedance characteristic curve is plotted with frequency on the x-axis and impedance magnitude and impedance angle on the y-axis. This method can visually display bioelectrical impedance data graphically, facilitating analysis and extraction of useful information.

[0050] S60: Extract characteristic parameters related to body fat percentage from the impedance characteristic curve, including the rate of change of impedance modulus and the range of impedance angle distribution.

[0051] In this embodiment, the characteristic parameter refers to a parameter that reflects body fat percentage. By analyzing and processing the impedance characteristic curve, characteristic parameters related to body fat percentage can be extracted. The rate of change of impedance modulus refers to the rate at which the impedance modulus changes with frequency, reflecting the degree of change in human bioelectrical impedance at different frequencies. The range of impedance angle values ​​refers to the range of impedance angle values ​​at different frequencies, reflecting the changes in the phase relationship between current and voltage.

[0052] In an embodiment, the impedance characteristic curve is subjected to piecewise fitting to obtain a plurality of fitted curve segments. The slope of each fitted curve segment is calculated to obtain a plurality of slope values. According to the plurality of slope values, an impedance modulus variation rate is calculated, which is the mean of all the slope values. The impedance characteristic curve is divided into a plurality of interval segments. The distribution range of the impedance angle value in each interval segment is counted, and the maximum value of the distribution range of the impedance angle value in all the interval segments is taken as the distribution range of the impedance angle value. In this way, a feature parameter related to the body fat rate can be extracted from the impedance characteristic curve, which provides a basis for accurately calculating the body fat rate.

[0053] S70: Based on the feature parameter, the body fat rate of the to-be-measured object is calculated to obtain the final body fat rate result of the accurate measurement mode.

[0054] In the embodiments of the present application, based on the extracted feature parameter related to the body fat rate, the body fat rate of the to-be-measured object is determined by a specific calculation method. These feature parameters contain the characteristic information of the bioelectrical impedance of the human body at different frequencies, and these information can more accurately reflect the fat content of the human body.

[0055] In an embodiment, referring to Figure 3 , step S30 can include steps S31-S34, which will be described in detail as follows:

[0056] S31: According to the electrode system configuration of the body fat scale, a combination scheme of a plurality of preset frequencies and a plurality of preset current intensities is determined to obtain a plurality of frequency-current combinations, wherein each frequency-current combination corresponds to a specific frequency and current intensity, and each frequency-current combination is used to cover the impedance response characteristics of different tissue types.

[0057] In the embodiments of the present application, the electrode system configuration of the body fat scale can determine the range and manner of the frequency and current intensity that it can apply. Different electrode system configurations can be suitable for different measurement requirements and human tissue types. The combination scheme of a plurality of preset frequencies and a plurality of preset current intensities is adopted in the embodiments of the present application to comprehensively obtain the bioelectrical impedance information of different human tissues. Different tissue types (such as fat, muscle, water, etc.) have different impedance response characteristics to different frequencies and current intensities, and by setting a plurality of frequency-current combinations, these different response characteristics can be covered.

[0058] For example, low-frequency current can be more suitable for measuring the impedance of extracellular fluid, while high-frequency current can be more suitable for penetrating the cell membrane to measure the impedance of total body fluid.

[0059] In an embodiment, a series of preset frequencies and preset current intensities are determined according to the performance parameters and measurement requirements of the body composition scale electrode system. For example, the preset frequencies can be set to 1 kHz, 10 kHz, 100 kHz, 1 MHz, etc., and the preset current intensities can be set to 0.5 mA, 0.6 mA, 0.7 mA, 0.8 mA, etc. These frequencies and current intensities are combined to obtain a plurality of frequency-current combinations, such as (1 kHz, 0.5 mA), (10 kHz, 0.6 mA), etc. Each frequency-current combination is designed for the impedance response characteristics of a specific tissue type. This way of determining frequency-current combinations can ensure that comprehensive bioelectrical impedance data is obtained.

[0060] S32: Generate a frequency scanning sequence according to the plurality of frequency-current combinations, and sequentially apply excitation currents of the corresponding frequency-current combinations to the foot of the subject through the electrode system of the body composition scale according to the frequency scanning sequence, and synchronously measure the response voltage generated on the electrode system at each time of applying the excitation current.

[0061] The frequency scanning sequence is a sequence obtained by sequentially arranging the plurality of frequency-current combinations. Sequentially applying excitation currents according to the sequence can ensure the systematicness and accuracy of the measurement process. The excitation current is the current delivered by the electrode system of the body composition scale to the foot of the human body, which is used to stimulate the electrical impedance response of the human body. The response voltage is the voltage generated on the electrode system when the human body is subjected to the excitation current, which is related to the electrical impedance of the human body.

[0062] In an embodiment, the frequency-current combinations can be sorted in order of increasing frequency to generate the frequency scanning sequence. For example, for the 25 frequency-current combinations described above, the sequence obtained by sorting by frequency is: (0.5 kHz, 0.2 mA), (0.5 kHz, 0.4 mA), (0.5 kHz, 0.6 mA) … (2 MHz, 1.2 mA). The control circuit of the body composition scale controls the electrode system to sequentially apply excitation currents of the corresponding frequency-current combinations to the foot of the subject according to the sequence. At the moment of applying the excitation current, a high-precision voltage measurement circuit is used to synchronously measure the response voltage generated on the electrode system, and the excitation current value and the response voltage value are recorded.

[0063] For example, in actual measurement, the body composition scale starts working according to the generated frequency scanning sequence. First, the excitation current of (1 kHz, 0.3 mA) is applied, and the voltage measurement circuit synchronously measures the response voltage of 0.15 V. Then the excitation current of (1 kHz, 0.6 mA) is applied, and the response voltage of 0.3 V is measured. In this way, the measurement of all frequency-current combinations is sequentially completed according to the sequence, and the excitation current value and the response voltage value of each group are recorded.

[0064] S33: Calculate the raw bioimpedance value under each frequency-current combination according to the applied excitation current value and the measured response voltage value.

[0065] In the embodiments of the present application, the raw bioimpedance value is the human body bioimpedance value calculated according to Ohm's law from the applied excitation current value and the measured response voltage value. It is the initial measurement value without any processing, reflecting the bioimpedance characteristics of the human body under a specific frequency and current intensity.

[0066] In an embodiment, the control circuit of the body fat scale calculates the raw bioimpedance value using division operation after recording the applied excitation current value and the measured response voltage value. For example, the applied excitation current value is 0.6 mA, and the measured response voltage value is 0.36 V, then the raw bioimpedance value is 0.36 V / 0.6 mA = 600 Ω. The calculated raw bioimpedance value is associated with the corresponding frequency-current combination and stored in the memory of the body fat scale.

[0067] For example, after completing the measurement of the (50 kHz, 0.8 mA) frequency-current combination, it is recorded that the excitation current value is 0.8 mA and the response voltage value is 0.48 V. According to Ohm's law, the raw bioimpedance value is 0.48 V / 0.8 mA = 600 Ω. Similarly, for the (500 kHz, 1.2 mA) combination, if the excitation current value is 1.2 mA and the response voltage value is 0.72 V, then the raw bioimpedance value is 0.72 V / 1.2 mA = 600 Ω.

[0068] S34: Check all the calculated raw bioimpedance values at different frequencies, eliminate abnormal values that are outside the reasonable physiological range, and form the final bioimpedance dataset.

[0069] In the embodiments of the present application, the reasonable physiological range is the normal value interval obtained by statistical analysis of a large amount of human bioimpedance measurement data. The raw bioimpedance values outside this range are considered abnormal values, which may be caused by various interference factors in the measurement process, such as poor electrode contact, external electromagnetic interference, measurement device error, etc. The final bioimpedance dataset is the effective data set remaining after checking and eliminating abnormal values, which is used for subsequent body fat rate calculation.

[0070] In an embodiment, upper and lower limits of a reasonable physiological range can be preset. For example, through a large amount of experimental data statistics, the bioimpedance value of the human body under common frequency and current intensity is generally between 150Ω-1800Ω. For all the original bioimpedance values calculated at all frequencies, it is checked one by one whether it is within the range. If a value is less than 150Ω or greater than 1800Ω, it is determined as an abnormal value and is excluded. The remaining normal bioimpedance values form the final bioimpedance dataset.

[0071] For example, when checking the original bioimpedance values calculated for 25 frequency current combinations, it is found that the original bioimpedance value corresponding to the (100kHz, 1.2mA) combination is 2000Ω, which is out of the preset reasonable physiological range of 150Ω-1800Ω, and the value is determined as an abnormal value and is excluded. The original bioimpedance values of the other 24 combinations are within the reasonable range, and the 24 values form the final bioimpedance dataset for subsequent body fat rate calculation.

[0072] In an embodiment, referring to Figure 4 Step S40 can include steps S41-S44, which will be described in detail below:

[0073] S41: classify the bioimpedance data according to frequency and current intensity to form a plurality of sub-datasets, each sub-dataset corresponding to a specific frequency and current intensity combination.

[0074] In the embodiment of the application, the bioimpedance data is a large amount of data obtained at a plurality of preset frequencies and a plurality of preset current intensities, and the classification according to frequency and current intensity is to facilitate subsequent targeted processing of data at different frequencies and current intensities. The specific frequency and current intensity combination refers to the unique combination of frequency and current intensity corresponding to each sub-dataset.

[0075] For example, in the precise measurement mode, bioimpedance data at 10 different frequencies and 5 different current intensities is obtained, a total of 50 groups of data. Classifying these data according to frequency and current intensity will form 50 sub-datasets, each sub-dataset corresponding to a specific frequency and current intensity combination, such as (1kHz, 0.5mA), (10kHz, 0.6mA), etc.

[0076] In an embodiment, the classification can be achieved by data sorting and screening. First, the frequency and current intensity information in the bioimpedance data are extracted, and then sorted in ascending order of frequency, and in ascending order of current intensity under the same frequency. After sorting, data screening is performed according to different combinations of frequency and current intensity, and data with the same combination is classified into a sub-data set. This classification method can make subsequent data processing more organized and improve processing efficiency.

[0077] S42: filtering processing is performed on each sub-data set to remove high-frequency noise and low-frequency drift interference, to obtain a filtered sub-data set.

[0078] In the embodiment of the application, high-frequency noise refers to interference signals with high frequency, which can come from surrounding electronic devices, electromagnetic environment, etc.; low-frequency drift interference refers to the slow change of signals in the low-frequency band, which can be caused by factors such as unstable contact between the electrode and the human body, temperature change, etc. The filtering processing is to process the signal through a specific filter to remove unwanted interference signals and retain useful bioimpedance signals.

[0079] For example, in actual measurement, high-frequency noise can be generated by mobile phone signals, electromagnetic interference of electrical appliances, etc., and low-frequency drift interference can be caused by slight changes in the contact between the electrode and the human body. These interferences can affect the accuracy of bioimpedance data and need to be filtered.

[0080] In an embodiment, step S42 can be implemented as follows:

[0081] P1: for each sub-data set, a first-order high-pass filter is used to remove low-frequency drift interference, and the cutoff frequency of the first-order high-pass filter is dynamically set according to the excitation current frequency corresponding to the sub-data set.

[0082] In the embodiment of the application, the first-order high-pass filter is a filter that allows high-frequency signals to pass through while attenuating low-frequency signals. Low-frequency drift interference is the slow change of signals in the low-frequency band, which can affect the accuracy of bioimpedance data. By dynamically setting the cutoff frequency according to the excitation current frequency corresponding to the sub-data set, the filter can better adapt to the characteristics of data at different frequencies and effectively remove low-frequency drift interference.

[0083] For example, for a sub-data set with a low excitation current frequency, the cutoff frequency can be set relatively low; for a sub-data set with a high excitation current frequency, the cutoff frequency can be set relatively high.

[0084] In an embodiment, the cut-off frequency of the first-order high-pass filter is dynamically set according to the excitation current frequency f corresponding to the sub-data set, by the formula f_cutoff = k * f (where k is a preset coefficient, adjusted according to actual conditions). The sub-data set is input into the first-order high-pass filter, the filter attenuates low-frequency signals, removes low-frequency drift interference, and outputs signals processed by high-pass filtering. This way of dynamically setting the cut-off frequency can improve the adaptability and filtering effect of the filter.

[0085] P2: For the sub-data set processed by high-pass filtering, remove high-frequency noise by a low-pass filter with a fixed cut-off frequency to obtain a filtered sub-data set.

[0086] In the embodiments of the present application, the low-pass filter with a fixed cut-off frequency is a filter that allows low-frequency signals to pass through while attenuating high-frequency signals. High-frequency noise is a signal with a high frequency, which may come from surrounding electronic devices, electromagnetic environment, etc. By using a low-pass filter, these high-frequency noises can be removed, further improving the quality of the data.

[0087] For example, a low-pass filter with a fixed cut-off frequency of 10 kHz is set to attenuate signals with a frequency higher than 10 kHz.

[0088] In an embodiment, the sub-data set processed by high-pass filtering is input into the low-pass filter with a fixed cut-off frequency. The filter attenuates high-frequency signals and retains low-frequency bioelectrical impedance signals, and outputs the filtered sub-data set. This way of using a low-pass filter to remove high-frequency noise can effectively improve the signal-to-noise ratio of the data.

[0089] P3: Calculate the signal energy ratio of the sub-data set before and after filtering, and remove the sub-data set whose signal energy ratio exceeds the preset range.

[0090] In the embodiments of the present application, the signal energy ratio refers to the ratio of the signal energy of the filtered sub-data set to the signal energy of the unfiltered sub-data set. The signal energy ratio can reflect the loss of the signal during the filtering process. If the signal energy ratio exceeds the preset range, it means that the filtering process may have caused excessive attenuation or other abnormal effects on the signal, and such sub-data set may not be reliable and needs to be removed.

[0091] For example, the range of the preset signal energy ratio is 0.8-1.2, if the signal energy ratio of a certain sub-data set is less than 0.8 or greater than 1.2, it will be removed.

[0092] In an embodiment, the signal energy of the pre-filtered and post-filtered sub-data sets are calculated respectively. The signal energy can be obtained by integrating the square of the signal. The ratio of the signal energy of the post-filtered sub-data set to the signal energy of the pre-filtered sub-data set is calculated to obtain the signal energy ratio. It is checked whether the signal energy ratio of each sub-data set is within a preset range, and if it is out of the range, the sub-data set is removed from the data set. This way of calculating the signal energy ratio and removing abnormal sub-data sets can improve the reliability of the data.

[0093] S43: Fast Fourier transform is performed on the filtered sub-data sets to extract the amplitude component and the phase component in each sub-data set.

[0094] In the embodiments of the present application, the fast Fourier transform is an efficient algorithm for converting time domain signals to frequency domain signals. Through fast Fourier transform, the bioelectrical impedance signal can be converted from time domain to frequency domain, so that the amplitude component and the phase component of the signal are conveniently extracted. The amplitude component reflects the size of the signal, and the phase component reflects the phase information of the signal.

[0095] For example, a filtered sub-data set is a time domain bioelectrical impedance signal. After fast Fourier transform, the signal in the frequency domain is obtained, which contains the amplitude and phase information of different frequency components.

[0096] In an embodiment, the fast Fourier transform algorithm is used to process the filtered sub-data sets. First, the sub-data set is input into the fast Fourier transform program, which will calculate and convert the data and output the frequency domain signal. The amplitude and phase of each frequency component are extracted from the frequency domain signal to obtain the amplitude component and the phase component. This way can accurately extract the amplitude and phase information of the bioelectrical impedance signal, providing a basis for subsequent calculations.

[0097] S44: The amplitude component and the phase component are normalized respectively to obtain the normalized impedance amplitude and phase information.

[0098] In the embodiments of the present application, the normalization process is a standardization process of the amplitude component and the phase component, which makes the value range within a certain interval. The purpose of normalization is to eliminate the difference of data under different measurement conditions, so that the data is comparable.

[0099] For example, the amplitude component and the phase component obtained under different frequencies and current intensities may have different value ranges. Through normalization, they can be unified to a standard range.

[0100] In an embodiment, for the amplitude component, a max-min normalization method is adopted. The maximum and minimum values in the amplitude component are calculated, each amplitude component is subtracted by the minimum value, and then divided by the difference between the maximum value and the minimum value to obtain the normalized amplitude component. For the phase component, a similar normalization method is also adopted. The normalized amplitude component and phase component are taken as the normalized impedance amplitude and phase information respectively. This normalization method can improve the consistency and comparability of the data, which is helpful for subsequent analysis and calculation.

[0101] In an embodiment, referring to Figure 5 , step S70 can include steps S71-S74, which will be described in detail as follows:

[0102] S71: determining two feature weight coefficients according to the impedance modulus change rate and the impedance angle value distribution range, the feature weight coefficients being used to represent the influence degree of the impedance modulus change rate and the impedance angle value distribution range on the body fat rate.

[0103] In the embodiments of the present application, the feature weight coefficients are determined according to the impedance modulus change rate and the impedance angle value distribution range, which reflect the importance of these two feature parameters in the calculation of the body fat rate. Different individuals can have different feature weight coefficients because the body structures and bioelectrical impedance characteristics of different people can be different.

[0104] For example, for some people with high muscle content, the impedance modulus change rate can have a greater influence on the body fat rate; and for some people with high fat content, the impedance angle value distribution range can have a more significant influence on the body fat rate.

[0105] In an embodiment, referring to Figure 6 , step S71 includes steps S711-S714, which will be described in detail as follows:

[0106] S711: obtaining, from a pre-stored initial estimation model, a first feature weight coefficient initial value corresponding to the impedance modulus change rate and a second feature weight coefficient initial value corresponding to the impedance angle value distribution range, respectively.

[0107] In the embodiments of the present application, the pre-stored initial estimation model is a model established based on a large amount of experimental data and statistical analysis, which contains the initial corresponding relationship between the impedance modulus change rate and the impedance angle value distribution range and the feature weight coefficients. The first feature weight coefficient initial value and the second feature weight coefficient initial value are the initial weight coefficients corresponding to the impedance modulus change rate and the impedance angle value distribution range pre-set in the model.

[0108] For example, the initial estimation model can be obtained by analyzing bioelectrical impedance data and body fat rate data of different populations, where the initial value of the first feature weight coefficient corresponding to the impedance modulus change rate is 0.6, and the initial value of the second feature weight coefficient corresponding to the impedance angle value distribution range is 0.4.

[0109] In an embodiment, the body fat scale has a pre-stored initial estimation model stored in the memory. When the feature weight coefficients need to be determined, the initial values of the first feature weight coefficient and the second feature weight coefficient corresponding to the current impedance modulus change rate and the impedance angle value distribution range are found from the model. This way of obtaining initial values from the pre-stored model can provide a basis for subsequent weight coefficient adjustment.

[0110] S712: Adjust the initial value of the first feature weight coefficient according to the pre-stored correlation parameters between the impedance modulus change rate and the body fat rate, to obtain an optimized first feature weight coefficient.

[0111] In the embodiments of the present application, the pre-stored correlation parameters between the impedance modulus change rate and the body fat rate are obtained through a large number of experiments and researches, reflecting the degree of correlation between the impedance modulus change rate and the body fat rate. Adjusting the initial value of the first feature weight coefficient according to these correlation parameters can make the first feature weight coefficient more accurately reflect the influence of the impedance modulus change rate on the body fat rate.

[0112] For example, if the correlation parameters indicate that the correlation between the impedance modulus change rate and the body fat rate is strong, the initial value of the first feature weight coefficient can be appropriately increased; if the correlation is weak, it can be appropriately reduced.

[0113] In an embodiment, an adjustment coefficient is calculated according to the pre-stored correlation parameters. Assuming that the correlation parameter is r, the adjustment coefficient is k = 1 + r. The initial value of the first feature weight coefficient is multiplied by the adjustment coefficient to obtain the optimized first feature weight coefficient. For example, the initial value of the first feature weight coefficient is 0.6, the correlation parameter r = 0.2, then the adjustment coefficient k = 1 + 0.2 = 1.2, and the optimized first feature weight coefficient is 0.6 x 1.2 = 0.72. This way of adjusting the first feature weight coefficient according to the correlation parameters can improve the accuracy of the weight coefficient, and make the body fat rate calculation more reasonably consider the influence of the impedance modulus change rate.

[0114] S713: According to the statistical characteristics of the impedance angle value distribution range, the initial value of the second feature weight coefficient is corrected to obtain a second feature weight coefficient optimized for the current measurement.

[0115] In the embodiments of the present application, the statistical characteristics of the impedance angle value distribution range include mean, variance, distribution shape and other information, which reflect the characteristics and stability of the impedance angle value distribution. According to the statistical characteristics, the initial value of the second feature weight coefficient is corrected, so that the second feature weight coefficient can better adapt to the actual situation of the impedance angle value distribution in the current measurement, thereby more accurately reflecting the influence of the impedance angle value distribution range on the body fat rate.

[0116] For example, if the variance of the impedance angle value distribution range is large, it means that the data fluctuation is large, and the initial value of the second feature weight coefficient may need to be appropriately reduced; if the variance is small, it means that the data is stable, and the initial value of the second feature weight coefficient may need to be appropriately increased.

[0117] In an embodiment, the variance of the impedance angle value distribution range is first calculated. Assuming that the variance is σ 2, a correction coefficient m is determined according to a preset rule. If σ 2 is less than a certain threshold, m is greater than 1; if σ 2 is greater than a certain threshold, m is less than 1. The initial value of the second feature weight coefficient is multiplied by the correction coefficient m to obtain the second feature weight coefficient optimized for the current measurement. For example, the initial value of the second feature weight coefficient is 0.4, the variance σ 2 is large, and the correction coefficient m = 0.8, so the optimized second feature weight coefficient is 0.4 x 0.8 = 0.32. This way of correcting the second feature weight coefficient according to the statistical characteristics can adjust the weight according to the actual measurement situation and improve the accuracy of the body fat rate calculation.

[0118] S714: The first feature weight coefficient and the second feature weight coefficient after optimization are verified using a pre-stored verification condition. If the verification is passed, the two are used as the final feature weight coefficients; if the verification is not passed, a pre-stored default feature weight coefficient is used.

[0119] In the embodiments of the present application, the pre-stored verification condition is a condition set to ensure the rationality and effectiveness of the optimized feature weight coefficients. These conditions may include the value range of the weight coefficients, the proportional relationship between the weight coefficients, etc. Verification can avoid the adverse effects of unreasonable weight coefficients on the body fat rate calculation results.

[0120] For example, the verification condition may stipulate that the sum of the first feature weight coefficient and the second feature weight coefficient must be between 0.9 and 1.1, and each weight coefficient should be between 0 and 1.

[0121] In an embodiment, the optimized first feature weight coefficient and the second feature weight coefficient are substituted into the pre-stored verification conditions for checking. If all the verification conditions are satisfied, the two weight coefficients are taken as the final feature weight coefficients; if not, the pre-stored default feature weight coefficients are used, such as the default first feature weight coefficient being 0.5 and the second feature weight coefficient being 0.5. This verification mechanism can guarantee the reliability of the feature weight coefficients and improve the stability and accuracy of the body fat rate calculation.

[0122] S72: The impedance modulus change rate and the impedance angle value distribution range are respectively multiplied by the corresponding feature weight coefficients to obtain a weighted impedance modulus change rate and a weighted impedance angle value distribution range.

[0123] In the embodiments of the present application, the impedance modulus change rate and the impedance angle value distribution range are respectively multiplied by the corresponding feature weight coefficients in order to reflect the different importance of the two feature parameters in the calculation of the body fat rate. Through the weighting processing, the feature parameter that has a greater impact on the body fat rate can occupy a more important position in the calculation.

[0124] For example, if the feature weight coefficient of the impedance modulus change rate is 0.6, the feature weight coefficient of the impedance angle value distribution range is 0.4, the impedance modulus change rate is 0.5, and the impedance angle value distribution range is 0.3, then the weighted impedance modulus change rate is 0.5x0.6 = 0.3, and the weighted impedance angle value distribution range is 0.3x0.4 = 0.12.

[0125] In an embodiment, the impedance modulus change rate and the impedance angle value distribution range can be respectively multiplied by the corresponding feature weight coefficients using multiplication operation. The calculation results are respectively taken as the weighted impedance modulus change rate and the weighted impedance angle value distribution range. This weighting processing manner can more reasonably consider the influence of the two feature parameters on the body fat rate.

[0126] S73: The weighted impedance modulus change rate and the weighted impedance angle value distribution range are linearly superimposed to obtain an intermediate body fat rate result.

[0127] In the embodiments of the present application, the linear superposition is to add the weighted impedance modulus change rate and the weighted impedance angle value distribution range to obtain a preliminary body fat rate result. This linear superposition manner is based on the assumption that the influence of the two feature parameters on the body fat rate is linearly related.

[0128] For example, the weighted impedance modulus change rate is 0.3, and the weighted impedance angle value distribution range is 0.12, then the intermediate body fat rate result is 0.3 + 0.12 = 0.42.

[0129] In an embodiment, the weighted impedance modulus rate of change and the weighted impedance angle value distribution range are added together using an addition operation to obtain an intermediate body fat rate result. This linear superposition method is simple and intuitive, and can quickly obtain a preliminary body fat rate estimate.

[0130] S74: Nonlinear correction processing is performed on the intermediate body fat rate result to obtain a final body fat rate result in an accurate measurement mode.

[0131] In the embodiments of the present application, due to the complexity of human physiological structure and individual differences, the intermediate body fat rate result may have certain deviations and needs to be subjected to nonlinear correction processing. The nonlinear correction processing is performed by considering more factors, such as basic physiological parameters of the human body, to adjust the intermediate body fat rate result, so as to improve the accuracy of body fat rate calculation.

[0132] For example, the relationship between the body fat rate and the bioelectrical impedance characteristic parameters of people of different ages, heights, and weights may be nonlinear, and needs to be subjected to nonlinear correction to eliminate such differences.

[0133] In an embodiment, the following nonlinear correction processing method can be used, which can fully consider individual differences of the human body and improve the accuracy of body fat rate measurement. Referring to Figure 7 , the method specifically includes steps S741-S745, which will be described below:

[0134] S741: Obtain basic physiological parameters of the object to be measured, the basic physiological parameters including height, weight, and age.

[0135] In the embodiments of the present application, the basic physiological parameters are parameters reflecting the basic characteristics of the human body, and the height, weight, and age have important influences on the bioelectrical impedance characteristics and body fat rate of the human body. People of different heights, weights, and ages may have different body structures and fat distributions, and therefore these factors need to be considered when calculating the body fat rate.

[0136] For example, the metabolic rates of young people and old people are different, and the fat content and distribution may also be different; people with higher height and weight may also have different bioelectrical impedance characteristics.

[0137] In an embodiment, the user can input the height, weight, and age information through the human-computer interaction interface of the body fat scale, or the information can be obtained from the paired mobile terminal, which is pre-recorded by the user. After the body fat scale receives the information, it is stored in the internal storage for subsequent use.

[0138] S742: Calculate the basal metabolic rate of the object to be measured according to the basic physiological parameters.

[0139] In the embodiments of the present application, the basal metabolic rate refers to the energy metabolic rate of a human body in a state of being awake and extremely quiet, without being affected by muscle activity, environmental temperature, food and mental stress, etc. The basal metabolic rate is closely related to factors such as height, weight and age of the human body, and the basal metabolic rate can be calculated to further understand the physiological state of the human body.

[0140] For example, a young person with a higher height and a heavier weight may have a relatively higher basal metabolic rate, while an old person with a lower height and a lighter weight may have a relatively lower basal metabolic rate.

[0141] In an embodiment, the basal metabolic rate can be calculated using the Harris-Benedict equation. For a male, the basal metabolic rate (BMR) = 88.362 + (13.397 x weight (kg)) + (4.799 x height (cm)) - (5.677 x age (years)); for a female, the basal metabolic rate (BMR) = 447.593 + (9.247 x weight (kg)) + (3.098 x height (cm)) - (4.330 x age (years)). The height, weight and age information input by the user are substituted into the corresponding formula to calculate the basal metabolic rate.

[0142] S743: Based on the basal metabolic rate, a non-linear correction factor is determined, which is used to adjust the deviation of the intermediate body fat rate result.

[0143] In the embodiments of the present application, the non-linear correction factor is determined according to the basal metabolic rate, which is used to adjust the possible deviation in the intermediate body fat rate result. Due to the complexity of the physiological structure of the human body and individual differences, the intermediate body fat rate result may not accurately reflect the actual body fat rate, and needs to be adjusted by the non-linear correction factor.

[0144] For example, a person with a higher basal metabolic rate may have a relatively lower body fat rate, and the intermediate body fat rate result needs to be appropriately adjusted downward; while a person with a lower basal metabolic rate may have a relatively higher body fat rate, and the intermediate body fat rate result needs to be appropriately adjusted upward.

[0145] In an embodiment, step S743 can be implemented in the following manner:

[0146] K1: According to the basal metabolic rate value, a pre-stored basal correction factor - metabolic rate relationship curve is queried to obtain a basal non-linear correction factor.

[0147] In the embodiments of the present application, the pre-stored basic correction factor-metabolic rate relationship curve is obtained through a large number of experiments and data analysis, and reflects the corresponding relationship between the basic metabolic rate and the basic nonlinear correction factor. The basic metabolic rate is the energy consumed by the human body in a quiet state to maintain life activities, and different basic metabolic rates correspond to different body fat metabolism conditions, so the basic nonlinear correction factor can be determined according to the basic metabolic rate.

[0148] For example, a person with a higher basic metabolic rate may have faster body fat metabolism, and the corresponding basic nonlinear correction factor may be smaller; a person with a lower basic metabolic rate may have slower body fat metabolism, and the corresponding basic nonlinear correction factor may be larger.

[0149] In an embodiment, the body fat scale stores the basic correction factor-metabolic rate relationship curve in the memory. When the basic metabolic rate value is obtained, the corresponding basic nonlinear correction factor is found in the relationship curve. For example, the basic metabolic rate is 1200 kcal / day, and by querying the relationship curve, the corresponding basic nonlinear correction factor is 0.9. This way of querying the relationship curve to obtain the basic nonlinear correction factor can utilize existing experimental data to provide a reasonable basis for nonlinear correction.

[0150] K2: respectively calculate the deviation degree of the impedance modulus change rate relative to the median value of its preset normal range, and the deviation degree of the impedance angle value distribution range relative to the median value of its preset normal range.

[0151] In the embodiments of the present application, the median value of the preset normal range is the median value of the impedance modulus change rate and the impedance angle value distribution range obtained by statistical analysis of the bioelectrical impedance data of a large number of normal human bodies. The calculation of the deviation degree can measure the difference between the currently measured impedance modulus change rate and the impedance angle value distribution range and the normal situation, thereby reflecting the abnormal situation of the human tissue impedance characteristics.

[0152] For example, the median value of the preset normal range of the impedance modulus change rate is 0.5, and the currently measured impedance modulus change rate is 0.6, and the deviation degree is (0.6-0.5) / 0.5=0.2.

[0153] In an embodiment, for the impedance modulus change rate, the current measured value is subtracted from the median value of the preset normal range, and then divided by the median value of the preset normal range to obtain the deviation degree of the impedance modulus change rate. For the impedance angle value distribution range, the similar method is also used to calculate the deviation degree. The two deviation degrees calculated are recorded respectively to prepare for subsequent calculation of the comprehensive deviation coefficient.

[0154] K3: the two deviation degrees are weighted and fused to calculate a comprehensive deviation coefficient representing the abnormal degree of the current tissue impedance characteristics.

[0155] In the embodiments of the present application, the weighted fusion is to combine the deviation degree of the impedance modulus change rate and the deviation degree of the impedance angle value distribution range according to certain weights to obtain a comprehensive index to reflect the abnormal degree of the current tissue impedance characteristics. The influence of different deviation degrees on the abnormal degree of the tissue impedance characteristics can be different, and the two factors can be more reasonably considered by the weighted fusion.

[0156] For example, assuming that the weight of the impedance modulus change rate deviation degree is 0.6, the weight of the impedance angle value distribution range deviation degree is 0.4, the impedance modulus change rate deviation degree is 0.2, and the impedance angle value distribution range deviation degree is 0.3, the comprehensive deviation coefficient is 0.2*0.6+0.3*0.4=0.24.

[0157] In an embodiment, the weights of the impedance modulus change rate deviation degree and the impedance angle value distribution range deviation degree can be pre-set. The two deviation degrees are multiplied by the corresponding weights respectively, and then added to obtain the comprehensive deviation coefficient.

[0158] K4: substituting the basic nonlinear correction factor and the comprehensive deviation coefficient into a pre-stored dynamic adjustment function for calculation, the dynamic adjustment function is configured to enhance the correction amplitude of the basic nonlinear correction factor when the comprehensive deviation coefficient increases, and output the final nonlinear correction factor used for body fat rate calculation.

[0159] In the embodiments of the present application, the pre-stored dynamic adjustment function is a function established according to experimental data and theoretical analysis, which describes the relationship between the basic nonlinear correction factor, the comprehensive deviation coefficient and the final nonlinear correction factor. When the comprehensive deviation coefficient increases, it indicates that the abnormal degree of the tissue impedance characteristics increases, and the correction amplitude of the basic nonlinear correction factor needs to be enhanced to more accurately adjust the body fat rate calculation result.

[0160] For example, the dynamic adjustment function can be f(x, y)=x*(1+y), where x is the basic nonlinear correction factor and y is the comprehensive deviation coefficient. When the comprehensive deviation coefficient increases, the final nonlinear correction factor will also increase accordingly.

[0161] In an embodiment, the basic nonlinear correction factor and the comprehensive deviation coefficient can be substituted into the pre-stored dynamic adjustment function for calculation. For example, the basic nonlinear correction factor is 0.9, the comprehensive deviation coefficient is 0.24, and the dynamic adjustment function is f(x, y)=x*(1+y), then the final nonlinear correction factor used for body fat rate calculation is 0.9*(1+0.24)=1.116. This way of calculating the nonlinear correction factor by using the dynamic adjustment function can dynamically adjust the correction amplitude according to the abnormal degree of the tissue impedance characteristics, and improve the accuracy of the body fat rate calculation.

[0162] S744: multiplying the intermediate body fat rate result by the non-linear correction factor to obtain a corrected body fat rate result.

[0163] In the embodiments of the present application, the intermediate body fat rate result is multiplied by the non-linear correction factor to adjust the intermediate body fat rate result according to the non-linear correction factor, so that it is closer to the actual body fat rate.

[0164] For example, if the intermediate body fat rate result is 0.4 and the non-linear correction factor is 0.9, the corrected body fat rate result is 0.4 x 0.9 = 0.36.

[0165] In an embodiment, the intermediate body fat rate result is multiplied by the non-linear correction factor using a multiplication operation to obtain the corrected body fat rate result. This adjustment method is simple and direct, and can effectively correct the intermediate body fat rate result.

[0166] S745: smoothing the corrected body fat rate result to eliminate fluctuations caused by individual differences to obtain a final body fat rate result.

[0167] In the embodiments of the present application, the smoothing process is a further processing of the corrected body fat rate result to eliminate fluctuations caused by individual differences. Different people may have different physical conditions and measurement conditions, which may cause fluctuations in the corrected body fat rate result. The smoothing process can make the result more stable and accurate.

[0168] For example, in multiple measurements, the corrected body fat rate result may fluctuate due to factors such as posture, electrode contact during measurement, etc. The smoothing process can reduce the impact of these fluctuations.

[0169] In an embodiment, the moving average method can be used to smooth the corrected body fat rate result. A certain number of recent measurement results are selected, and their average value is calculated as the final body fat rate result. With the appearance of new measurement results, the average value is constantly updated. This smoothing method can effectively eliminate fluctuations caused by individual differences and measurement errors, and improve the stability and accuracy of body fat rate measurement.

[0170] In an embodiment, with reference to Figure 8 , step S60 can include steps S61-S65, which will be described in detail below:

[0171] S61: performing piecewise fitting processing on the impedance characteristic curve to obtain a plurality of fitted curve segments.

[0172] In the embodiments of the present application, the segmented fitting processing is to divide the impedance characteristic curve into multiple small segments, and fit each small segment to obtain multiple fitted curve segments. The impedance characteristic curve is a curve describing the variation of the impedance modulus and the impedance angle with frequency. Since the curve can be relatively complex, the segmented fitting can more accurately describe the local characteristics of the curve.

[0173] For example, the impedance characteristic curve can have different trends in different frequency intervals, and the segmented fitting can better capture these changes.

[0174] In an embodiment, according to the characteristics of the impedance characteristic curve, the curve is divided into multiple small segments by selecting appropriate segmentation points. For each small segment, a polynomial fitting method is used for fitting. For example, for a small segment of the curve, a quadratic polynomial y = ax2 + bx + c can be used for fitting, and the coefficients a, b and c of the polynomial are determined by the least squares method. The fitted curve of each small segment is obtained, and these fitted curves form multiple fitted curve segments. This segmented fitting processing can more accurately describe the local characteristics of the impedance characteristic curve, and provide an accurate data basis for subsequent feature parameter extraction.

[0175] S62: Calculate the slope of each fitted curve segment to obtain multiple slope values.

[0176] In the embodiments of the present application, the slope of the fitted curve segment reflects the change rate of the curve segment. In the impedance characteristic curve, the slope reflects the rate of change of the impedance modulus with frequency. By calculating the slope of each fitted curve segment, the change of the impedance modulus in different frequency intervals can be understood, which provides a basis for subsequent calculation of the impedance modulus change rate.

[0177] For example, in a certain frequency interval, the slope of the fitted curve segment is large, indicating that the impedance modulus changes rapidly with frequency in this interval; and the smaller the slope, the slower the change.

[0178] In an embodiment, for each fitted curve segment, the slope is calculated according to the derivative of its fitting polynomial. Taking the quadratic polynomial fitted curve y = ax2 + bx + c as an example, its derivative is y' = 2ax + b. In the frequency interval corresponding to the fitted curve segment, several frequency points are selected, and these frequency points are substituted into the derivative formula to calculate the corresponding slope values. The slope values are averaged to obtain the average slope of the fitted curve segment. The above operation is sequentially performed on each fitted curve segment to obtain multiple slope values. This way of calculating the slope can accurately reflect the change of the fitted curve segment.

[0179] S63: According to the multiple slope values, calculate the impedance modulus change rate, and the impedance modulus change rate is the average of all slope values.

[0180] In the embodiments of the present application, the impedance modulus change rate is a parameter that comprehensively reflects the change of the impedance modulus with frequency in the entire frequency range. By calculating the average of the slope values of all the fitting curve segments, a relatively stable and accurate impedance modulus change rate can be obtained.

[0181] For example, if the slope values of different fitting curve segments differ greatly, taking the average can balance these differences to some extent and obtain a more representative change rate.

[0182] In an embodiment, the calculated slope values are added and then divided by the number of slope values to obtain the impedance modulus change rate. Assuming that n slope values k1, k2, …, kn are obtained, the impedance modulus change rate R = (k1 + k2+ … + kn) / n. This calculation method is simple and intuitive and can effectively integrate the information of each fitting curve segment to obtain the overall change rate of the impedance modulus.

[0183] S64: Intervals of the impedance characteristic curve are divided to obtain multiple interval segments.

[0184] In the embodiments of the present application, the impedance characteristic curve is divided into intervals to facilitate the statistics of the distribution of the impedance angle values in different frequency intervals. Different frequency intervals can correspond to the bioelectrical impedance characteristics of different tissues of the human body, and the distribution characteristics of the impedance angle values can be analyzed in more detail through interval division.

[0185] For example, the impedance characteristic curve can be divided according to different frequency ranges such as low frequency, medium frequency, and high frequency.

[0186] In an embodiment, the impedance characteristic curve is divided into intervals according to a preset frequency range. For example, the frequency range is divided into 1kHz - 10kHz, 10kHz - 100kHz, 100kHz - 1MHz, and the like. For each interval segment, the corresponding impedance angle value data range is determined. This interval division method can be adjusted according to actual needs to better adapt to different analysis purposes.

[0187] S65: The distribution range of the impedance angle values in each interval segment is counted, and the maximum value of the distribution range of the impedance angle values in all interval segments is taken as the distribution range of the impedance angle values.

[0188] In the embodiments of the present application, counting the distribution range of the impedance angle values in each interval segment can understand the change amplitude of the impedance angle values in different frequency intervals. Taking the maximum value of the distribution range of the impedance angle values in all interval segments as the final distribution range of the impedance angle values can highlight the interval with the most significant change of the impedance angle values and more effectively reflect the overall distribution characteristics of the impedance angle values.

[0189] For example, the impedance angle value changes greatly in some interval segments, and changes less in other interval segments. Taking the maximum value can capture the key information of the change.

[0190] In an embodiment, for each interval segment, the maximum value and the minimum value of the impedance angle value in the interval segment are found, the difference between them is calculated to obtain the distribution range of the impedance angle value in the interval segment. The distribution ranges of all interval segments are compared, and the maximum value among them is selected as the distribution range of the impedance angle value. For example, there are three interval segments, and the distribution ranges of the impedance angle values of the three interval segments are 10°, 15° and 8° respectively. The final distribution range of the impedance angle value is 15°. This statistical method can accurately reflect the distribution characteristics of the impedance angle value.

[0191] Correspondingly, in order to better implement the above method, the embodiments of the present application also provide a multi-frequency-based bioelectrical impedance body fat accurate measurement system. As shown in the figure, the multi-frequency-based bioelectrical impedance body fat accurate measurement system 80 comprises: Figure 9

[0192] The receiving module 801 is configured to receive a mode selection instruction issued by a user through a body fat scale human-computer interaction interface or a paired mobile terminal, and determine a current working mode. The electrode system of the body fat scale comprises at least four electrodes.

[0193] The fast measurement module 802 is configured to, if the current mode is a fast measurement mode, apply a preset single reference frequency and reference current intensity excitation through at least two electrodes of the body fat scale to obtain a single set of bioelectrical impedance data, and calculate a fast body fat rate result based on the single set of bioelectrical impedance data.

[0194] The accurate measurement module 803 is configured to, if the current working mode is an accurate measurement mode, obtain bioelectrical impedance data of a to-be-measured object under a plurality of preset frequencies and a plurality of preset current intensities through a complete electrode system composed of at least four electrodes of the body fat scale. The plurality of preset frequencies cover a wide frequency band range from low frequency to high frequency. The low frequency is used to measure extracellular fluid impedance, and the high frequency is used to penetrate the cell membrane to measure total body fluid impedance.

[0195] The signal processing module 804 is configured to perform signal processing on the bioelectrical impedance data, extract corresponding impedance amplitude and phase information under each frequency and each current intensity, and store the impedance amplitude and phase information as a first data set and a second data set respectively.

[0196] The impedance calculation module 805 is configured to calculate impedance modulus and impedance angle values of the to-be-measured object under different frequencies according to the first data set and the second data set, and construct an impedance characteristic curve based on the impedance modulus and the impedance angle values.

[0197] ​The feature extraction module 806 is configured to extract a feature parameter related to the body fat rate from the impedance characteristic curve, and the feature parameter comprises an impedance modulus change rate and an impedance angle value distribution range.

[0198] The body fat calculation module 807 is configured to calculate the body fat rate of the to-be-measured object based on the feature parameter, and obtain a final body fat rate result of the accurate measurement mode.

[0199] The implementation of each module can refer to the method embodiments described above, and will not be described here. The technical effects of the modules and the device can refer to the method embodiments described above.

[0200] The above only discloses the preferred embodiments of the present application, and of course cannot limit the scope of the rights of the present application, so the equivalent changes made according to the claims of the present application still belong to the scope of the present application.

Claims

1. A method for accurate measurement of body fat based on multi-frequency bioelectrical impedance, characterized in that, The method includes the following steps: The body fat scale receives mode selection commands from users through the human-computer interaction interface of the body fat scale or a paired mobile terminal, determines the current working mode, and the electrode system of the body fat scale includes at least four electrodes. If the current mode is the rapid measurement mode, a preset single reference frequency and reference current intensity excitation is applied through at least two electrodes of the body fat scale to obtain a single set of bioelectrical impedance data, and a rapid body fat percentage result is calculated based on the single set of bioelectrical impedance data. If the current working mode is the precision measurement mode, the bioelectrical impedance data of the subject under test is obtained through the complete electrode system consisting of at least four electrodes of the body fat scale at multiple preset frequencies and multiple preset current intensities. The multiple preset frequencies cover a wide frequency range from low frequency to high frequency. The low frequency is used to measure extracellular fluid impedance, and the high frequency is used to penetrate the cell membrane to measure total body fluid impedance. The bioelectrical impedance data is processed to extract the impedance amplitude and phase information corresponding to each frequency and current intensity, and the impedance amplitude and phase information are stored as a first data set and a second data set, respectively. Based on the first data set and the second data set, calculate the impedance magnitude and impedance angle of the object under test at different frequencies, and construct an impedance characteristic curve based on the impedance magnitude and impedance angle. From the impedance characteristic curve, feature parameters related to body fat percentage are extracted, including the rate of change of impedance modulus and the range of impedance angle distribution. Based on the rate of change of impedance modulus and the range of impedance angle distribution, two characteristic weighting coefficients are determined. These characteristic weighting coefficients are used to characterize the degree of influence of the rate of change of impedance modulus and the range of impedance angle distribution on body fat percentage. Multiply the rate of change of impedance magnitude and the range of impedance angle values ​​by the corresponding feature weight coefficients to obtain the weighted rate of change of impedance magnitude and the weighted range of impedance angle values. The weighted rate of change of impedance modulus and the weighted range of impedance angle distribution are linearly superimposed to obtain the intermediate body fat percentage result; The intermediate body fat percentage results are subjected to nonlinear correction based on the basal metabolic rate of the subject to obtain the final body fat percentage results of the accurate measurement mode. The rate of change of impedance modulus is the mean of the slope of each fitted curve segment obtained after segmenting the impedance characteristic curve; the impedance angle distribution range refers to the maximum value of the impedance angle distribution range within the multiple intervals obtained after dividing the impedance characteristic curve into intervals according to different frequency ranges.

2. The method according to claim 1, characterized in that, The bioelectrical impedance data is processed to extract the impedance amplitude and phase information corresponding to each frequency and current intensity, including: The bioelectrical impedance data is classified according to frequency and current intensity to form multiple subsets, each subset corresponding to a specific combination of frequency and current intensity. Each subset of data is filtered to remove high-frequency noise and low-frequency drift interference, resulting in a filtered subset of data. Perform a Fast Fourier Transform on the filtered subsets to extract the magnitude and phase components from each subset. The amplitude and phase components are normalized to obtain normalized impedance amplitude and phase information.

3. The method according to claim 2, characterized in that, The intermediate body fat percentage results are subjected to nonlinear correction processing to obtain the final body fat percentage results, including: Obtain the basic physiological parameters of the subject to be tested, including height, weight and age; Based on the aforementioned basic physiological parameters, the basal metabolic rate of the subject under test was calculated. Based on the basal metabolic rate, a nonlinear correction factor is determined, which is used to adjust for deviations in the intermediate body fat percentage results. Multiply the intermediate body fat percentage result by the nonlinear correction factor to obtain the corrected body fat percentage result; The corrected body fat percentage result is smoothed to eliminate fluctuations caused by individual differences, resulting in the final body fat percentage result.

4. The method according to claim 3, characterized in that, Acquire bioelectrical impedance data of the object under test at multiple preset frequencies and multiple preset current intensities, including: Based on the electrode system configuration of the body fat scale, a combination scheme of multiple preset frequencies and multiple preset current intensities is determined to obtain multiple frequency-current combinations. Each frequency-current combination corresponds to a specific frequency and current intensity, and each frequency-current combination is used to cover the impedance response characteristics of different tissue types. A frequency scanning sequence is generated based on the multiple frequency current combinations, and an excitation current corresponding to the frequency current combination is applied to the feet of the subject through the electrode system of the body fat scale in sequence according to the frequency scanning sequence. The response voltage generated on the electrode system is measured synchronously each time an excitation current is applied. Based on the applied excitation current value and the synchronously measured response voltage value, the original bioelectrical impedance value under each frequency current combination is calculated. The original bioelectrical impedance values ​​at all calculated frequencies are verified, and outliers that exceed the reasonable physiological range are removed to form the final bioelectrical impedance dataset.

5. The method according to claim 4, characterized in that, Each subset of the dataset is filtered to remove high-frequency noise and low-frequency drift interference, including: For each subset of data, a first-order high-pass filter is used to remove low-frequency drift interference. The cutoff frequency of the first-order high-pass filter is dynamically set according to the excitation current frequency corresponding to the subset of data. For the high-pass filtered subset, high-frequency noise is removed by a low-pass filter with a fixed cutoff frequency to obtain the filtered subset. Calculate the signal energy ratio of the subset before and after filtering, and remove the subset whose signal energy ratio exceeds a preset range.

6. The method according to claim 1, characterized in that, The determination of two characteristic weighting coefficients based on the rate of change of impedance magnitude and the range of impedance angle distribution includes: From the pre-stored initial estimation model, obtain the initial values ​​of the first feature weight coefficient corresponding to the rate of change of impedance modulus and the initial values ​​of the second feature weight coefficient corresponding to the range of impedance angle distribution. Based on the pre-stored correlation parameters between the rate of change of impedance modulus and body fat percentage, the initial value of the first feature weight coefficient is adjusted to obtain the optimized first feature weight coefficient. Based on the statistical characteristics of the impedance angle distribution range, the initial value of the second feature weight coefficient is corrected to obtain the second feature weight coefficient optimized for the current measurement. The optimized first feature weight coefficient and second feature weight coefficient are verified using pre-stored verification conditions. If the verification passes, the two are used as the final feature weight coefficients; if the verification fails, the pre-stored default feature weight coefficient pair is used.

7. The method according to claim 3, characterized in that, The determination of the nonlinear correction factor based on the basal metabolic rate includes the following steps: Based on the basal metabolic rate value, query the pre-stored basal correction factor-metabolic rate relationship curve to obtain the basal nonlinear correction factor; Calculate the deviation of the rate of change of impedance modulus from the median of its preset normal range, and the deviation of the range of impedance angle values ​​from the median of its preset normal range, respectively. The two deviations are weighted and fused to calculate a comprehensive deviation coefficient that characterizes the degree of abnormality in the current tissue impedance characteristics. The basic nonlinear correction factor and the comprehensive deviation coefficient are substituted into a pre-stored dynamic adjustment function for calculation. The dynamic adjustment function is configured to enhance the correction amplitude of the basic nonlinear correction factor when the comprehensive deviation coefficient increases, and output the final nonlinear correction factor used for body fat percentage calculation.

8. A multi-frequency bioelectrical impedance-based precise body fat measurement system using the method described in any one of claims 1-7, characterized in that, The system includes: The receiving module is used to receive mode selection instructions issued by the user through the human-computer interaction interface of the body fat scale or the paired mobile terminal, and determine the current working mode. The electrode system of the body fat scale includes at least four electrodes. The rapid measurement module is used to apply a preset single reference frequency and reference current intensity excitation through at least two electrodes of the body fat scale to obtain a single set of bioelectrical impedance data if the current mode is rapid measurement mode, and to calculate the rapid body fat percentage result based on the single set of bioelectrical impedance data. The precision measurement module is used to acquire bioelectrical impedance data of the subject under test at multiple preset frequencies and multiple preset current intensities through a complete electrode system consisting of at least four electrodes of the body fat scale if the current working mode is precision measurement mode. The multiple preset frequencies cover a wide frequency band from low frequency to high frequency. The low frequency is used to measure extracellular fluid impedance, and the high frequency is used to penetrate the cell membrane to measure total body fluid impedance. The signal processing module is used to perform signal processing on the bioelectrical impedance data, extract the impedance amplitude and phase information corresponding to each frequency and current intensity, and store the impedance amplitude and phase information as a first data set and a second data set, respectively. The impedance calculation module is used to calculate the impedance magnitude and impedance angle of the object under test at different frequencies based on the first data set and the second data set, and to construct an impedance characteristic curve based on the impedance magnitude and impedance angle. The feature extraction module is used to extract feature parameters related to body fat percentage from the impedance characteristic curve. The feature parameters include the rate of change of impedance modulus and the range of impedance angle distribution. The body fat calculation module is used to calculate the body fat percentage of the subject based on the aforementioned feature parameters, thereby obtaining the final body fat percentage result in the accurate measurement mode.

Images