A traditional chinese medicine pulse condition detection device based on vital sign radar data

Abstract

The application relates to a traditional Chinese medicine pulse condition detection device based on vital sign radar data and relates to the technical field of medical detection.The device comprises a data acquisition module, a wave peak processing module, a screening module and a calculation and output module; the data acquisition module collects human body breathing and heartbeat data through vital sign detection radar; the wave peak processing module extracts a breathing and heartbeat effective data segment, carries out peak value searching and peak value interval calculation, generates a breathing peak value frequency table and a heartbeat peak value frequency table; the screening module respectively screens and eliminates the breathing peak value frequency table and the heartbeat peak value frequency table; and the calculation and output module calculates a heartbeat frequency measurement value FreHeart according to the heartbeat peak value screened frequency table and outputs a pulse knot or a pulse facilitation according to the multiple relationship between the maximum peak value interval and the average heartbeat frequency. The application is suitable for the field of medical detection.

Description

Technical Field

[0001] This invention relates to the field of medical testing technology, specifically a traditional Chinese medicine pulse diagnosis device based on vital sign radar data. Background Technology

[0002] Nowadays, people are increasingly health-conscious, especially with the recent surge in popularity of Traditional Chinese Medicine (TCM) both domestically and internationally. TCM is a medical theoretical system gradually formed and developed by the Chinese through long-term medical practice. TCM emphasizes pulse diagnosis, which is closely related to the Qi and blood of the internal organs. Through pulse diagnosis, different pulse patterns can be observed, leading to the understanding and diagnosis of diseases. In TCM diagnosis, it is common to experience a situation where "the heart understands the pulse, but the fingers cannot pinpoint it." Standardization is needed for TCM pulse detection and interpretation. Currently, most TCM health monitoring measures heart rate and heart rate variability, offering only simple functions. Existing personal health monitoring products are mainly used for detecting blood pressure, blood sugar, and body temperature, with fewer products specifically for TCM pulse detection. Accurate medical instruments, such as electrocardiogram (ECG) monitors, require twelve or eighteen leads, are bulky, and complex to operate; they are not convenient for the average person to carry and use correctly.

[0003] Currently, the most similar patent to this invention is titled "A Non-Contact Traditional Chinese Medicine Pulse Detection Method and Device" (CN113488162A). This invention discloses a non-contact traditional Chinese medicine pulse detection method in the field of traditional Chinese medicine detection technology, comprising the following steps: S1: video acquisition; S2: face detection; S3: ROI extraction; S4: signal preprocessing; S5: blind source separation; S6: spectrum analysis; S7: pulse extraction and analysis; a non-contact traditional Chinese medicine pulse detection device, comprising a binocular camera, a video image processing unit, and a business function processing unit; this invention utilizes a common camera to capture face videos, and through digital image technologies such as face detection and three-primary-color separation, and digital signal processing technologies such as blind source separation, filtering, and discrete Fourier transform, extracts traditional Chinese medicine pulse information from the photoplethysmography pulse wave signal acquired from the captured video. By analyzing the pulse through relative traditional Chinese medicine pulse theory, the health status of the body can be more accurately determined. The aforementioned comparative patent uses a camera to capture facial data to obtain vital sign data, and analyzes the pulse based on this data; however, it is greatly affected by light, has poor accuracy, and cannot fully meet the user's all-day monitoring needs. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a TCM pulse detection device based on vital sign radar data, which solves the problems of poor accuracy and significant light interference during signal acquisition in existing patents, and realizes non-contact, all-weather, all-time TCM pulse detection.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solution: a traditional Chinese medicine pulse detection device based on vital sign radar data, comprising a data acquisition module, a peak processing module, a screening module, and a calculation and output module.

[0006] The data acquisition module is used to collect human breathing and heartbeat data through a vital signs detection radar. The peak processing module is used to set the effective data acquisition start frame as M frames, and extract the effective respiratory and heartbeat data segments from the data acquired by the data acquisition module from frame M+1. Each effective respiratory and heartbeat data segment is N frames, and each effective respiratory and heartbeat data segment includes a frame number, respiratory energy value, and heartbeat energy value. The peak processing module is also used to perform peak lookup and peak interval calculation on the effective respiratory and heartbeat data segments to generate a respiratory peak frequency table and a heartbeat peak frequency table. The filtering module is used to filter and remove peak respiratory frequency tables and peak heart rate frequency tables respectively, generating peak respiratory frequency tables and peak heart rate frequency tables after filtering. The calculation and output module is used to calculate the mean heart rate (MeanHeart) based on the frequency table after filtering heart rate peaks, and to calculate the measured heart rate (FreHeart). If 60 < FreHeart < 80, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the mean heart rate, where K is in the range of 1.05 ≤ K ≤ 2, then this pulse is output as a knotted pulse. If 80 < FreHeart < 100, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the mean heart rate, then this pulse is output as a rapid pulse.

[0007] In the peak processing module, the peak threshold for each effective data segment of respiratory and heartbeat is set to Q times the maximum peak value of that data segment, and the range of the Q value is: 0.1≤Q≤0.8.

[0008] The Q value is 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, or 0.8.

[0009] The filtering module uses the following criteria to filter and remove entries from the heart rate peak frequency table: entries are removed if any of the following conditions are met: 1. The breathing rate is less than 36 frames; the range of the value 'a' is: a≤36; that is, the breathing rate is greater than 32 breaths / minute. Second, the breathing rate is greater than 150 frames; the range of the b value is b≥150; that is, the breathing rate is less than 8 breaths / minute. The filtering module filters and removes heart rate peak frequency tables based on the following criteria: a table is removed if any of the following conditions are met: 1. Heart rate less than 10 frames per second; the range of the c value is c≤10; that is, a heart rate greater than 120 beats / minute; Second, the heart rate is greater than 30 frames per second; the range of the d value is d≥30; that is, the heart rate is less than 40 beats / minute.

[0010] The calculation and output module calculates the heart rate measurement value FreHeart, and the formula for calculating the heart rate measurement value FreHeart is as follows: FreHeart = 60000 / (MeanHeart * FramePeriod), where 60000 is the number of milliseconds in one minute, MeanHeart is the average heartbeat frequency, and FramePeriod is the frame period. If 60 < FreHeart < 80, and within each segment of valid respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, where the range of K is 1.05 ≤ K ≤ 2, then this pulse pattern is output as a knotted pulse. If 80 < FreHeart < 100, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, then this pulse pattern is output as a rapid pulse. If FreHeart ≤ 60 times, or 100 times ≤ FreHeart, then the heart rate measurement result will not be output.

[0011] The value of M is 200, and the value of N is 300.

[0012] The vital signs detection radar uses a 6843XX chip or a 1642XX chip.

[0013] This invention provides a TCM pulse diagnosis device based on vital sign radar data. Addressing the issue of light interference in existing and patented technologies, this invention uses a radar sensor for signal acquisition. This sensor is unaffected by weather or light, enabling all-weather, all-day detection with higher accuracy and a better user experience. Attached Figure Description

[0014] Figure 1 This is a diagram of the detection module of the present invention; Figure 2 This is a schematic diagram of the valid data in the data packet of the present invention; Figure 3 This is a flowchart of the detection process of the present invention; Figure 4 This is a diagram showing the specific data segment frame numbers for this invention; Figure 5 This is a diagram illustrating the peak threshold setting method of the present invention; Figure 6This is a diagram showing the peak index table and peak interval table of the present invention; Figure 7 This is a peak interval frequency table of the present invention; Figure 8 This is a diagram illustrating the respiratory and heart rate calculation method of the present invention. Figure 9 This is a table showing the correspondence between common TCM pulse patterns. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Example: like Figure 1-9 As shown, this embodiment of the invention provides a traditional Chinese medicine pulse diagnosis device based on vital sign radar data. The detection process of the device includes the following steps: Step 1: Data Acquisition: Collect human breathing and heart rate data using vital sign detection radar.

[0017] like Figure 1 As shown, the hardware module of this invention includes a data acquisition module, a peak processing module, a filtering module, and a calculation and output module.

[0018] The data acquisition module is provided by Ti's vital signs radar 6843XX and 1642XX radar chips, including: AWR6843XX, AWR1642XX, IWR6843XX, and IWR1642XX. After each frame is processed, the chip outputs a data packet and extracts the valid respiratory and heart rate data segments from the data packet.

[0019] Step 2: Data Filtering: Set the starting frame for valid data acquisition to frame M. Extract the valid respiratory and heart rate data segments from frame M+1 of the data acquired in step 1. Each valid respiratory and heart rate data segment consists of N frames. Each valid respiratory and heart rate data segment includes: frame number, respiratory energy value, and heart rate energy value. Step 3: Data Processing: Perform peak lookup and peak interval calculation on the effective data segments of respiratory and heartbeat from Step 2 to generate a respiratory peak frequency table and a heartbeat peak frequency table; filter and remove from the respiratory peak frequency table and the heartbeat peak frequency table respectively to generate a respiratory peak frequency table and a heartbeat peak frequency table after filtering. Step 4: Output Results: Calculate the mean heart rate MeanHeart based on the heart rate peak filtering frequency table in Step 3, and calculate the heart rate measurement value FreHeart. If 60 < FreHeart < 80, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the mean heart rate, where K is in the range of 1.05 ≤ K ≤ 2, then output this pulse as a knotted pulse. If 80 < FreHeart < 100, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the mean heart rate, then output this pulse as a rapid pulse.

[0020] In the third step, the peak threshold of each effective data segment of breathing and heartbeat is set to Q times the maximum peak value of that data segment, and the range of the Q value is: 0.1≤Q≤0.8.

[0021] The Q value is 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, or 0.8.

[0022] In the third step, the criteria for filtering and eliminating respiratory peak frequency tables are as follows: Any of the following conditions must be met for a table to be eliminated: 1. The breathing rate is less than 36 frames; the range of the value 'a' is: a≤36; that is, the breathing rate is greater than 32 breaths / minute. Second, the breathing rate is greater than 150 frames; the range of the b value is b≥150; that is, the breathing rate is less than 8 breaths / minute. In the third step, the criteria for filtering and eliminating heart rate peak frequency tables are as follows: Any of the following conditions must be met for a table to be eliminated: 1. Heart rate less than 10 frames; the range of the c value is b≤10; that is, a heart rate greater than 120 beats / minute; Second, the heart rate is greater than 30 frames per second; the range of the d value is d≥30; that is, the heart rate is less than 40 beats / minute.

[0023] In the fourth step, the heart rate measurement value FreHeart is calculated. The formula for calculating the heart rate measurement value FreHeart is as follows: FreHeart = 60000 / (MeanHeart * FramePeriod), where 60000 is the number of milliseconds in one minute, MeanHeart is the average heartbeat frequency, and FramePeriod is the frame period. If 60 < FreHeart < 80, and within each segment of valid respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, where the range of K is 1.05 ≤ K ≤ 2, then this pulse pattern is output as a knotted pulse. If 80 < FreHeart < 100, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, then this pulse pattern is output as a rapid pulse.

[0024] If FreHeart ≤ 60 times, or 100 times ≤ FreHeart, then the heart rate measurement result will not be output.

[0025] The value of M is 200, and the value of N is 300.

[0026] The vital signs detection radar uses a 6843XX chip or a 1642XX chip.

[0027] like Figure 2 As shown, the data packet length of the effective respiratory and heartbeat data segment is greater than 216 bytes and the data length is a multiple of 8. Each effective respiratory and heartbeat data segment includes a frame number (frameNumber), a respiratory waveform energy value (sumEnergyBreathWfm), and a heartbeat waveform energy value (sumEnergyHeartWfm). The frame period is FramePeriod, which is set to 50ms.

[0028] The peak processing module includes peak value lookup for respiratory and heart rate, peak interval calculation, and peak frequency table generation. The generated respiratory and heart rate peak frequency table is filtered and eliminated, which is the filtering module. The calculation and output module calculates the measured heart rate value and outputs the nodal pulse or rapid pulse according to the K-fold relationship between the maximum peak interval and the average heart rate.

[0029] At rest, the normal respiratory rate for men is 8.5–21.2 breaths / min, and for women it is 12.4–26.5 breaths / min. Resting heart rate varies greatly among people, generally ranging from 0.625 to 1.615 Hz (mean 1.025 Hz), or 37.5–96.9 beats / min (mean 61.5 beats / min).

[0030] To ensure data stability and accuracy, the first M frames of data after the radar is activated are discarded without processing. The value of M can be set manually; in this specific implementation, it is set to 200. Processing begins from the (M+1)th frame, extracting valid data segments of respiratory and heart rate. Each valid data segment consists of N frames; in this specific implementation, it is set to 300. The duration of each valid data segment is N*Tframe (300*100ms=30s). The frame numbers of the specific data segments are as follows: Figure 4 As shown.

[0031] Peak values ​​are searched and calculated for each segment of respiratory and heart rate data. A threshold needs to be set for each segment to find the peak value. The method for setting the peak threshold is as follows: Figure 5As shown. The calculation of each peak index value and peak interval is as follows: assuming there are three peaks, A, B, and C, the corresponding frame numbers are TAframeNumber, TBframeNumber, and TCframeNumber, respectively.

[0032] The corresponding peak interval is the frame sequence difference between two consecutive peaks, i.e., TdiffAB, TdiffBC, as shown below. Figure 6 As shown, the peak interval frequency table is a statistical table of the frequency of occurrence of each peak interval. Figure 6 In the data, TdiffAB and TdiffBC each appear once, with a frequency of 1 and a probability of 50%. Data with an occurrence probability of less than 50% can also be filtered out.

[0033] Figure 7 This is a frequency table of peak intervals. The frequency table is filtered and removed; the result after filtering is a mat matrix. Any item meeting any of the following conditions can be filtered out: A breathing rate of less than 36 frames per second; that is, a breathing rate of more than 32 breaths per minute. A breathing rate greater than 150 frames per second; that is, a breathing rate greater than 150 frames per second when breathing is less than 8 breaths per minute. Heart rate less than 10 frames per second; that is, heart rate greater than 120 beats per minute. Heart rate greater than 30 frames per second; that is, heart rate less than 40 beats per minute. After filtering the frequency table, calculate the respiration and heart rate separately. The calculation method is to first calculate the mean respiratory and heart rate (MeanHeart and MeanBreath) in the filtered frequency table: peak interval Diff (column 1 of the frequency table), corresponding to the frequency F (column 2 of the frequency table). The corresponding formula is: Mean = sum(Diff * F) / sum(F). See [link to method for calculating the mean respiratory and heart rate]. Figure 8 .

[0034] Heart rate measurement: FreHeart = 60000 / (MeanHeart * FramePeriod), where 60000 is the number of milliseconds in one minute and FramePeriod is the frame period.

[0035] If the FreHeart rate is between 60 and 80 beats per minute, and within the effective data segment (300 frames), there is a maximum peak interval that is 1.05 and 1.1 times the average heart rate, then this pulse is a knotted pulse; if the heart rate is between 80 and 100 beats per minute, and within the effective data segment (300 frames), there is a maximum peak interval that is 1.05 and 1.1 times the average heart rate, then this pulse is a rapid pulse.

[0036] The maximum peak interval is determined by removing the maximum value of the first column in the frequency table matrix (Mat), i.e., max(Mat(:,1)); the mean heart rate is MeanHeart. That is, it is determined whether max(Mat(:,1)) is between MeanHeart*1.05 and MeanHeart*1.1.

[0037] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A traditional Chinese medicine pulse diagnosis device based on vital sign radar data, characterized in that: It includes a data acquisition module, a peak processing module, a filtering module, and a calculation and output module; The data acquisition module is used to collect human breathing and heartbeat data through a vital signs detection radar. The peak processing module is used to set the effective data acquisition start frame as M frames, and to extract the effective respiratory and heartbeat data segments from the data acquired by the data acquisition module from frame M+1, with each effective respiratory and heartbeat data segment consisting of N frames. Each valid data segment of breathing and heartbeat includes: frame number, respiratory energy value, and heartbeat energy value; The peak processing module is also used to perform peak lookup and peak interval calculation on the effective data segments of respiratory and heartbeat, and generate a respiratory peak frequency table and a heartbeat peak frequency table; The filtering module is used to filter and remove peak respiratory frequency tables and peak heart rate frequency tables respectively, generating peak respiratory frequency tables and peak heart rate frequency tables after filtering. The peak threshold for each effective data segment of respiratory and heartbeat is set to Q times the maximum peak value of that data segment, and the range of the Q value is: 0.1≤Q≤0.8; The filtering module filters and removes entries from the respiratory peak frequency table under the following conditions: the entry is removed if either of the following conditions is met: respiratory rate is less than 36 frames; respiratory rate is greater than 150 frames. The filtering module filters and removes heart rate peak frequency tables under the following conditions: heart rate less than 10 frames; heart rate greater than 30 frames. The calculation and output module is used to calculate the mean heart rate (MeanHeart) based on the frequency table after filtering the heart rate peaks, and to calculate the heart rate measurement value (FreHeart). The formula for calculating the heart rate measurement value (FreHeart) is as follows: FreHeart = 60000 / (MeanHeart * FramePeriod), where 60000 is the number of milliseconds in one minute, MeanHeart is the average heartbeat frequency, and FramePeriod is the frame period. If 60 < FreHeart < 80, and within each segment of valid respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, where the range of K is 1.05 ≤ K ≤ 2, then this pulse pattern is output as a knotted pulse. If 80 < FreHeart < 100, and within each segment of effective respiratory and heart rate data, there is a maximum peak interval that is K times the average heart rate, then this pulse pattern is output as a rapid pulse.

2. The TCM pulse diagnosis device based on vital sign radar data according to claim 1, characterized in that: The Q value is 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, or 0.

8.

3. The TCM pulse diagnosis device based on vital sign radar data according to claim 1, characterized in that: The value of M is 200, and the value of N is 300.

4. A TCM pulse diagnosis device based on vital sign radar data according to claim 1, characterized in that: The vital signs detection radar uses a 6843XX chip or a 1642XX chip.

Images