A syndrome differentiation analysis method based on multi-gradient pulse condition data

Abstract

The application discloses a syndrome differentiation analysis method based on multi-gradient pulse condition data, which comprises the following steps: collecting pulse waves of the cun, guan and chi parts of the left and right hands of a patient under different pressure gradients, extracting the main wave amplitude of the obtained pulse wave original signal, determining the external pressure value corresponding to the maximum main wave amplitude as the optimal pulse-taking pressure, extracting the multi-dimensional pulse condition characteristics of the pulse wave original signal under the optimal pulse-taking pressure, calling preset three-dimensional decision tree rules to quantitatively evaluate each viscera and determine viscera syndrome differentiation evaluation parameters, and outputting viscera syndrome differentiation evaluation results; constructing a directed graph transmission network of the viscera syndrome differentiation evaluation parameters and twelve meridians, and mapping and outputting corresponding deficiency-excess cold-heat diagnosis labels according to the numerical interval of the obtained meridian energy value of the twelve meridians. The application does not need to rely on the subjective experience of doctors or the subjective questionnaire of patients, eliminates the difference in human experience, and ensures that the detection results of the same patient at different time points or on different devices are still highly consistent.

Description

Technical Field

[0001] This invention belongs to the field of TCM intelligent diagnosis and medical signal processing technology, and in particular relates to a diagnostic analysis method based on multi-gradient pulse data. Background Technology

[0002] Currently, traditional Chinese medicine pulse diagnosis theory holds that the three pulse positions of Cun, Guan, and Chi correspond to the functional states of different internal organs. The Cun, Guan, and Chi positions on the left hand correspond to the heart, liver, and kidney, respectively, while the Cun, Guan, and Chi positions on the right hand correspond to the lungs, spleen, and kidney (Mingmen). Traditional Chinese medicine obtains information about the internal organs through the three-position, nine-pulse diagnosis to differentiate syndromes and treat accordingly, demonstrating a high degree of holism and systematicity.

[0003] Existing computer-aided pulse diagnosis and analysis systems typically only output basic pulse parameters (such as pulse rate and pulse strength), lacking the ability to systematically extrapolate from pulse data to the state of the internal organs and meridians. While some systems can provide a constitution type assessment, their analysis often includes subjective questionnaires rather than being entirely based on objective pulse data, resulting in poor reproducibility of conclusions.

[0004] In terms of meridian status analysis, existing methods mainly rely on meridian electrometry (such as Liangdaoluo) or acupoint detection, and lack methods to directly deduce the patency and deficiency / excess and cold / heat status of the twelve meridians from pulse data.

[0005] The existing pulse diagnosis analysis system is difficult to meet clinical needs in terms of analysis accuracy. The main reasons are: (1) It only uses pulse data from a single pressure point, which is insufficient in terms of information; (2) It fails to make full use of the changes in pulse under different pressure conditions; (3) It lacks a systematic way to deduce the functional state of the internal organs from purely objective pulse data; and (4) It lacks a method to deduce the state of the twelve meridians from pulse data.

[0006] The prior art, with application number CN202411691131.9 and titled "A Pulse Analysis Method Based on a Pulse Diagnosis Instrument," inputs the obtained user pulse image into a pre-constructed standard pulse image model to determine the user's pulse image type and outputs the user's pulse image type result. The above-mentioned prior art relies only on a single superficial pulse image classification, without combining the theory of cun, guan, and chi pulse differentiation of viscera and organs, and without quantifying the cold, heat, deficiency, and excess of viscera and organs. It uses a single fixed pressure acquisition without stratified gradient pressure application. In application, the error in distinguishing between floating pulse, deep pulse, and deficiency and excess pulse is large, and the clinical accuracy does not meet the requirements of syndrome differentiation. Summary of the Invention

[0007] The purpose of this invention is to provide a diagnostic analysis method based on multi-gradient pulse data, which can combine traditional Chinese medicine pulse diagnosis theory with modern signal processing technology to establish a quantitative mapping model from objective pulse data to the state of the internal organs and meridians. This method does not rely on subjective questionnaires and significantly improves the objectivity and repeatability of diagnostic conclusions.

[0008] This invention adopts the following technical solution: a diagnostic analysis method based on multi-gradient pulse data, comprising the following steps:

[0009] S1: Simultaneous acquisition of pulse wave signals at the cun, guan, and chi positions of the patient's left and right hands using multi-gradient pressure. The raw pulse wave signal is acquired after the pressure stabilizes at each pressure gradient.

[0010] S2: Extract the main wave amplitude of the original pulse wave signal under each pressure gradient, determine the applied pressure value corresponding to the maximum main wave amplitude as the optimal pulse-taking pressure, and extract the multidimensional pulse features of the original pulse wave signal under the optimal pulse-taking pressure; the multidimensional pulse features include pulse position, pulse strength, pulse rate, pulse tension and pulse flow.

[0011] S3: Based on multi-dimensional pulse characteristics, the preset three-dimensional decision tree rules are called to perform a quantitative evaluation of each organ to obtain organ differentiation evaluation parameters, and the organ differentiation evaluation results are output according to the organ differentiation evaluation parameters.

[0012] S4: Construct a directed graph transmission network based on the organ differentiation assessment parameters and the exterior-interior and interior-exterior relationships of the twelve meridians, calculate the energy values ​​of the twelve meridians, and output the deficiency-excess, cold-heat diagnostic labels of each meridian based on the numerical range of the energy values ​​of the twelve meridians.

[0013] Furthermore, multi-gradient pressurization in S1 is achieved by setting the initial pressure value, the final pressure value, and the pressurization step size.

[0014] Furthermore, the process for determining the optimal pulse sampling pressure in S2 is as follows:

[0015] S201: Detect the main peak of the raw pulse wave signal under each pressure gradient and extract the representative main peak amplitude of each cardiac cycle;

[0016] S202: Construct an amplitude-pressure response scatter plot with the applied pressure value under each pressure gradient as the x-axis and the corresponding representative main wave amplitude value as the y-axis. After curve fitting, determine the applied pressure value corresponding to the maximum representative main wave peak amplitude as the optimal pulse sampling pressure.

[0017] Furthermore, the main peak detection in S201 includes the following steps:

[0018] The original pulse wave signal is preprocessed to obtain the basic pulse wave signal, and candidate peaks are obtained by detecting all local maxima points of the basic pulse wave signal using basic mathematical methods.

[0019] A dual judgment mechanism of adaptive threshold and time-domain constraint is used to screen candidate peaks to obtain the main peak amplitude sequence;

[0020] An adaptive peak threshold function that varies with time is constructed using the sliding window method. If the amplitude of a candidate peak at a certain moment is greater than or equal to the adaptive peak threshold at the corresponding moment, it is retained as a candidate peak to be screened in the time domain; otherwise, the candidate peak is considered to be noise or interference peak and noise is removed.

[0021] A lower limit for the physiological reentry period is set, which is the minimum time window. If the time interval between adjacent candidate peaks is less than the lower limit for the physiological reentry period, it is determined that there is a diphtheria wave or noise. For two adjacent candidate peaks, the one with the larger amplitude is retained.

[0022] Furthermore, the S3 visceral differentiation assessment parameters include the visceral deficiency-excess index and cold-heat attribute; among them, a positive value of the visceral deficiency-excess index represents an excess syndrome, a negative value of the visceral deficiency-excess index represents a deficiency syndrome, and a visceral deficiency-excess index of 0 represents a neutral syndrome. Combined with the visceral mapping relationship of the Cunkou pulse at each location, the visceral differentiation assessment results including the heart, liver, spleen, lung, kidney and the corresponding six fu organs are generated.

[0023] Furthermore, in S4, the energy values ​​of the twelve meridians are calculated, and the diagnostic labels for deficiency / excess and cold / heat of each meridian are output based on the numerical range of the energy values ​​of the twelve meridians. The energy values ​​of the twelve meridians include the energy values ​​of the six Zang meridians and the six Fu meridians. The steps are as follows:

[0024] S401: Obtaining the energy values ​​of the six viscera meridians based on the viscera differentiation assessment parameters;

[0025] S402: Calculate the energy values ​​of the six viscera meridians by combining the energy transmission weights of the same internal and external meridians through internal and external coupling transmission;

[0026] S403: Calculate the initial value of the unobstructedness of the corresponding meridian based on the energy values ​​of each meridian in the twelve meridian energy values;

[0027] S404: Combine the parameters for evaluating pulse flowability with the patency correction rule for the pulse flowability correction coefficient. The initial patency value is corrected according to the set patency correction rule to obtain the final patency of the corresponding meridian.

[0028] S405: Outputs the energy distribution status of the twelve meridians and labels each meridian with diagnostic tags for deficiency / excess, cold / heat.

[0029] Furthermore, in S404, pulse flowability is evaluated based on the pulse flowability coefficient obtained from the two-factor quantization index. The two-factor quantization index refers to the peak curvature variation coefficient and the high-frequency energy ratio. The peak curvature variation coefficient is obtained by taking the variation coefficient of the second derivative of the signal within a certain period interval to the left and right of the main wave peak. The high-frequency energy ratio is obtained by calculating the ratio of high-frequency energy to fundamental frequency energy after performing a fast Fourier transform (FFT) on the pulse signal.

[0030] Furthermore, the smoothness correction rule for the pulse flowability correction coefficient in S404 is as follows: the pulse flowability correction coefficient function adopts a piecewise linear mapping, and the corresponding pulse flowability correction coefficient is determined according to the value of the parameter for evaluating pulse flowability.

[0031] Furthermore, the diagnostic labels for deficiency / excess and cold / heat in S4 include deficiency / excess and cold / heat diagnostic labels. The deficiency / excess diagnostic labels include very excess, slightly excess, normal, slightly deficiency, and very deficiency; the cold / heat diagnostic labels include very heat, slightly heat, normal, slightly cold, and very cold.

[0032] Furthermore, the diagnostic analysis method also includes S5: generating a structured comprehensive diagnostic report that includes a pulse quantitative parameter report, a visceral syndrome differentiation report, a meridian status assessment, and a personalized conditioning plan.

[0033] This invention performs quantitative analysis on objective pulse wave signals acquired by a multi-gradient pressure sensor. Through a pre-defined three-dimensional decision tree rule, it transforms traditionally ambiguous pulse characteristics (such as floating, deep, wiry, and slippery) into clearly defined numerical ranges for quantitative evaluation, eliminating reliance on physician subjective experience or patient questionnaires. This eliminates human experience differences and ensures high consistency in test results for the same patient across different devices or time points.

[0034] This invention employs multi-gradient pressure synchronous acquisition technology to determine the optimal pulse-taking pressure by constructing an amplitude-pressure response curve. Compared to existing technologies that only utilize a single fixed pressure point to acquire pulse data, this invention can completely capture pulse information in the superficial, middle, and deep regions, solving the problems of misjudging superficial pulses and failing to detect deep pulses.

[0035] This invention constructs a directed graph transmission network based on the internal-external relationship of the Zang-Fu organs in Traditional Chinese Medicine, extrapolating the diagnostic results of a single Zang-Fu organ to the energy system of the twelve meridians. Through the internal-external coupling transmission mechanism, it quantifies and simulates the physiological energy transfer and pathological cold-heat transformation patterns of Zang-Fu organs as they circulate through the internal-external meridians. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention or related technologies, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0037] Figure 1 A flowchart illustrating the dialectical analysis method provided by this invention;

[0038] Figure 2 This is a schematic diagram of meridian status assessment in a specific embodiment of the dialectical analysis method provided by the present invention. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of the embodiments described herein clearer, the technical solutions in the embodiments will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments described herein, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments described herein without creative effort are within the scope of protection of this document. It should be noted that, unless otherwise specified, the embodiments and features described herein can be arbitrarily combined with each other.

[0040] The following is in conjunction with the appendix Figure 1 With appendix Figure 2 The present invention will be described in detail with reference to the embodiments:

[0041] A diagnostic analysis method based on multi-gradient pulse data, as shown in the appendix. Figure 1 As shown, it includes the following steps:

[0042] S1: Simultaneous acquisition of pulse wave signals at the cun, guan, and chi positions of the patient's left and right hands using multi-gradient pressure. The raw pulse wave signal is acquired after the pressure stabilizes at each pressure gradient.

[0043] S2: Extract the main wave amplitude of the original pulse wave signal under each pressure gradient, determine the applied pressure value corresponding to the maximum main wave amplitude as the optimal pulse taking pressure, and extract the multidimensional pulse features of the original pulse wave signal under the optimal pulse taking pressure.

[0044] S3: Based on multi-dimensional pulse characteristics, the preset three-dimensional decision tree rules are called to perform a quantitative evaluation of each organ to obtain organ differentiation evaluation parameters, and the organ differentiation evaluation results are output according to the organ differentiation evaluation parameters.

[0045] S4: Construct a directed graph transmission network based on the organ differentiation assessment parameters and the exterior-interior and interior-exterior relationships of the twelve meridians, calculate the energy values ​​of the twelve meridians, and output the deficiency-excess, cold-heat diagnostic labels of each meridian based on the numerical range of the energy values ​​of the twelve meridians.

[0046] In S1 of this invention, continuous pulse analog signals from six standard pulse positions—left cun, right cun, left guan, right guan, left chi, and right chi—are acquired using a multi-channel flexible array pressure sensor, and then converted from analog to digital to obtain digitized raw pulse wave signals. The flexible array pressure sensor is fixed to the three acquisition points—cun, guan, and chi—using a flexible substrate.

[0047] In this embodiment, the raw pulse wave signal data is organized and stored according to dimensions such as patient ID, collection date, collection time, cun, guan, chi position, and pressure gradient.

[0048] In this invention, the system provides two selectable gradient acquisition modes. Multi-gradient pressurization is achieved by setting the initial pressure value, the final pressure value, and the pressurization step size. Multiple data segments are acquired through the two gradient modes: 0 as the initial pressure value, 160 mmHg as the final pressure value, and the pressurization step size can be set to 20 mmHg or 10 mmHg.

[0049] Mode 1: The pressure step size is 20 mmHg, and 8 pressure gradients are generated in the range of 0~160 mmHg. 8 pulse data segments are obtained in each of the three parts of Cun, Guan and Chi.

[0050] Mode 2: The pressure step size is 10 mmHg, generating 16 pressure gradients within the range of 0~160 mmHg, and obtaining 16 pulse data segments at each of the three positions of Cun, Guan and Chi.

[0051] In this embodiment, no data is collected during the pressurization process. After the pressure stabilizes, pulse waveform data is continuously collected for no less than 10 seconds at each pressure gradient. The PID output PWM duty cycle drives the multi-gradient pressure control unit until the error between the current pressure value and the target pressure value is less than or equal to... If the pulse wave reading remains stable at mmHg for 1 second, it indicates that the current pressure value is stable at the target pressure value, and the acquisition time of the original pulse wave signal is greater than 10 seconds.

[0052] S2: Extract the main wave amplitude of the original pulse wave signal under each pressure gradient, and determine the applied pressure value corresponding to the maximum main wave amplitude as the optimal pulse-taking pressure; under the optimal pulse-taking pressure, extract the multidimensional pulse characteristics of the original pulse wave signal.

[0053] In S2 of this invention, the process of determining the optimal pulse sampling pressure is as follows:

[0054] S201: Detect the main peak of the raw pulse wave signal under each pressure gradient and extract the representative main peak amplitude of each cardiac cycle;

[0055] S202: Construct an amplitude-pressure response scatter plot with the applied pressure value under each pressure gradient as the x-axis and the corresponding representative main wave peak amplitude as the y-axis. After curve fitting, determine the applied pressure value corresponding to the maximum representative main wave peak amplitude, which is the optimal pulse sampling pressure.

[0056] In this embodiment, the original pulse wave signal is first subjected to a 0.5~10Hz bandpass filter to remove baseline drift and high-frequency noise to obtain the basic pulse wave signal. Then, all local maxima of the basic pulse wave signal are detected by basic mathematical methods to obtain candidate peaks. The main peak amplitude sequence is obtained by using a dual judgment mechanism of adaptive threshold and time domain constraint to ensure that only the true physiological main peak is included and diphtheria waves and noise interference are excluded.

[0057] An adaptive peak threshold Th(t) that varies with time is constructed using the sliding window method:

[0058]

[0059] Where α ranges from 0.5 to 0.6, and MA is the mean of the signal within the window. The original waveform sampling amplitude at time t, The duration of the sliding window. =1.5 seconds, β ranges from 0.1 to 0.2, and σ is the standard deviation of the signal within the window.

[0060] The process of selecting candidate peaks in the time domain is as follows:

[0061] If the candidate peak is at time amplitude Greater than or equal to the peak time Corresponding instantaneous adaptive peak threshold If so, it will be retained as a candidate peak to be screened in the time domain;

[0062] Otherwise, it is considered noise or interference peak and the noise is removed.

[0063] Temporal return period constraint: Set a lower limit for the physiological return period, i.e., the minimum time window. If the time interval between adjacent candidate peaks Less than the lower limit of physiological re-entry period If the peak value is not found, it is determined that there is a diphthous wave or noise. For two adjacent candidate peaks, the one with the larger amplitude is retained.

[0064] In this embodiment, the median of the main wave amplitude sequence retained by the dual determination mechanism of adaptive threshold and time domain constraint is taken after removing the maximum and minimum values, and is used as the representative main wave amplitude under the pressure gradient. The discrete points are fitted using the second-order polynomial least squares method, and the global maximum amplitude point is searched on the fitted curve. The applied pressure value corresponding to this point is the optimal pulse sampling pressure.

[0065] In this invention, the extracted feature parameters include pulse position, pulse strength, pulse rate, pulse tension, and pulse flowability; specifically as follows:

[0066] (1) Pulse location characteristics: The pulse location is quantified by the optimal pulse pressure. The floating pulse corresponds to a low optimal pulse pressure, the deep pulse corresponds to a high optimal pulse pressure, and the middle pulse is in the middle.

[0067] (2) Pulse strength characteristics: The difference between the peak and trough values ​​of the original pulse wave signal under the optimal pulse-taking pressure is used to characterize the strength of the pulse.

[0068] (3) Pulse rate feature: Peak detection is performed on the original pulse wave signal under the optimal pulse-taking pressure, the number of valid peaks obtained through the dual judgment mechanism of adaptive threshold and time domain constraint per unit time is counted, and the pulse rate (times / min) is calculated, with an accuracy of ±3 times / min.

[0069] (4) Pulse tension feature: Analyze the ascending branch slope of the original pulse wave signal and the descending branch slope , the tension T is quantified by normalizing and mapping the ascending branch slope of the main wave and the descending branch slope to [0, 1].

[0070] (5) Pulse fluency feature: Analyze the smoothness of the peak shape of the original pulse wave signal. A smooth and fluent peak shape indicates high fluency, i.e., a slippery pulse, while a rough and irregular peak shape indicates low fluency, i.e., a choppy pulse.

[0071] S3: Based on the multi-dimensional pulse conditions, the preset three-dimensional decision tree rules are invoked to perform quantitative evaluation of syndrome differentiation for each zang-fu organ, output zang-fu organ syndrome differentiation evaluation parameters, and output zang-fu organ syndrome differentiation evaluation results according to the zang-fu organ syndrome differentiation evaluation parameters;

[0072] In the present invention, the zang-fu organ syndrome differentiation evaluation parameters include the zang-fu organ deficiency-excess degree index VDE and cold-heat properties. The cun position of the left hand corresponds to the heart, the guan position of the left hand corresponds to the liver, the chi position of the left hand corresponds to the kidney yin, the cun position of the right hand corresponds to the lung, the guan position of the right hand corresponds to the spleen, and the chi position of the right hand corresponds to the gate of vitality.

[0073] In the present invention, the normalized pulse force P, pulse position L and pulse tension T are used as three-dimensional input features, combined with the mapping relationship between each corresponding position and the zang-fu organs corresponding to the cunkou pulse, the preset three-dimensional decision tree rules are invoked to perform quantitative syndrome differentiation evaluation on the liver at the left guan position, and the zang-fu organ deficiency-excess degree index VDE ∈ [-1,1] is output. A positive VDE value represents excess syndrome, a negative VDE value represents deficiency syndrome, and a VDE value of 0 represents balanced syndrome, so as to generate evaluation results of the deficiency-excess and cold-heat states including the heart, liver, spleen, lung, kidney and the corresponding six fu organs;

[0074] For the pulse position L: when L ≥ 0.65, the pulse position is a deep pulse; when 0.35 < L < 0.65, the pulse position is a medium pulse; when L ≤ 0.35, the pulse position is a floating pulse.

[0075] For the pulse force P: when , the pulse force is a forceful pulse; when , the pulse force is a moderate pulse; when P < 0.4, the pulse force is a weak pulse.

[0076] For the tension T: when T ≥ 0.7, the pulse condition is a wiry pulse; when 0.3 < T < 0.7, the pulse condition is a medium pulse; when T ≤ 0.3, the pulse condition is a soft pulse.

[0077] The above normalization process for pulse force P, pulse position L, and pulse tension T is a common technique in the art, and will not be elaborated herein.

[0078] In this embodiment, with the normalized pulse force P, the pulse position L, and the pulse tension T as three-dimensional input features, regarding the three-dimensional decision tree syndrome differentiation rules for quantitative syndrome differentiation and assessment of the left Guan liver, partial embodiments when the pulse force is a forceful pulse are as follows:

[0079] The pulse force is a forceful pulse, P≥0.6:

[0080] When the pulse force is a forceful pulse, the pulse position is a floating pulse, and the pulse condition is a wiry pulse, that is, P≥0.6, L≤0.35 and T≥0.7, the zang-fu deficiency-excess degree index VDE=+0.8 is output, indicating excess syndrome, liver depression transforming into heat, and excess heat of the liver;

[0081] When the pulse force is a forceful pulse, the pulse position is a deep pulse, and the pulse condition is a wiry pulse, that is, P≥0.6, L≥0.65 and T≥0.7, the zang-fu deficiency-excess degree index VDE=+0.9 is output, indicating interior excess syndrome, liver depression transforming into heat, and excess heat of the liver;

[0082] When the pulse force is a forceful pulse, the pulse position is a floating pulse, and the pulse condition is a moderate pulse or a soft pulse, that is, P≥0.6, L≤0.35 and T<0.7, the zang-fu deficiency-excess degree index VDE=+0.5 is output, indicating exterior excess, internal accumulation of dampness pathogen, and slight heat due to dampness stagnation;

[0083] When the pulse force is a forceful pulse, the pulse position is a deep pulse, and the pulse condition is a moderate pulse or a soft pulse, that is, P≥0.6, L≥0.65 and T<0.7, the zang-fu deficiency-excess degree index VDE=+0.6 is output, indicating interior excess syndrome, internal accumulation of dampness pathogen, and slight heat due to dampness stagnation;

[0084] When the pulse force is a forceful pulse, the pulse position is a middle pulse, and the pulse condition is a wiry pulse, that is, P≥0.6, 0.35<L<0.65 and T≥0.7, the zang-fu deficiency-excess degree index VDE=+0.9 is output, indicating excess syndrome, liver depression transforming into heat, and excess heat of the liver;

[0085] When the pulse force is a forceful pulse, the pulse position is a middle pulse, and the pulse condition is a moderate pulse or a soft pulse, that is, P≥0.6, 0.35<L<0.65 and T<0.7, the zang-fu deficiency-excess degree index VDE=+0.5 is output, indicating normal syndrome, internal accumulation of dampness pathogen, and slight heat due to dampness stagnation.

[0086] S4: Construct a directed graph conduction network according to the exterior-interior and collateral connection relationship between the zang-fu syndrome differentiation assessment parameters and the twelve meridians, and based on the calculated energy values of the twelve meridians map the numerical intervals to output diagnosis labels of deficiency, excess, cold and heat for each meridian;

[0087] In this invention, based on the relationship between the internal organs and their exterior and interior in traditional Chinese medicine, six fixed network combinations are formed according to the yin-yang pairing rule that yin organs govern the interior and yang organs govern the exterior. These combinations are: heart and small intestine, liver and gallbladder, spleen and stomach, lung and large intestine, kidney and bladder, and pericardium and triple burner. Each pairing achieves bidirectional direct network connection between the internal organs and their exterior and interior through the corresponding yin-yang meridians of the hands and feet. Each pair of internal organs and their exterior and interior forms an intrinsic connection based on the meridian pathways, with physiological functional synergy and coupling, and pathological pathogenic factors circulating and transforming along the meridians.

[0088] Based on the relationships between the internal organs and their exterior-interior connections in Traditional Chinese Medicine, a directed graph transmission network G=(V,B,W) is constructed.

[0089] Node set V: contains two types of nodes—Zang nodes (heart, liver, spleen, lung, kidney, pericardium, 6 nodes) and Fu nodes (small intestine, gallbladder, stomach, large intestine, bladder, triple burner, 6 nodes), for a total of 12 Zang and Fu nodes, each corresponding to one of the twelve meridians.

[0090] Directed edge set B: Directed edges are constructed based on the exterior-interior relationship of Yin meridians belonging to Zang organs and Yang meridians belonging to Fu organs and Zang organs. Specifically:

[0091] The direction of transmission of Yin meridians is from the Zang node to the Fu node to which it belongs, such as Lung → Large Intestine;

[0092] The direction of transmission of Yang meridians is from the Fu node to the Zang node to which it belongs, such as large intestine → lung;

[0093] In addition, based on the pathological transmission relationship between the internal organs (such as the transmission rules between the internal organs recorded in the "Inner Canon"), transmission boundaries are added between different internal organs, such as liver → spleen (wood controls earth), spleen → kidney (earth controls water), etc.

[0094] S401: Obtaining the energy value vector of the six viscera meridians based on the visceral differentiation assessment parameters ;

[0095] In this invention, the energy values ​​of the six internal organs are... The value is determined by the combination of the organ deficiency / excess index (VDE) and the cold / heat attribute (H) of the corresponding organ meridian:

[0096]

[0097] in: The baseline energy value is 10 in this embodiment; λ is the attenuation coefficient of energy by the VDE index of organ deficiency and excess, ranging from 0.8 to 1.0, and is 0.15 in this embodiment; The heat / cold correction factor is set as follows: heat is 1.10, cold is 0.90, and neutral is 1.00.

[0098] , The rating range is [-10, 10]; These represent the energy scores of the Heart Meridian, Liver Meridian, Kidney Meridian, Lung Meridian, Spleen Meridian, and Mingmen Meridian, respectively, which together constitute the energy value vector of the six internal organs. Since there is no independent pericardium-corresponding pulse location for the Cunkou Six-Part Pulse Diagnosis, this invention uses the Mingmen Meridian Energy Score collected from the right Chi pulse to replace the pericardium Zang Meridian Energy in the Sanjiao Fu Meridian Coupling Calculation.

[0099] In this embodiment, the peak pressure, pulsation frequency, pulse width, and descending slope of the pulse waves at six pulse positions are extracted from the digital pulse wave curve. A linear mapping correction model of the pulse diagnosis characteristic parameters to the interval [−10, 10] is established. The corresponding organ energy score is independently calculated for each pulse position: left cun corresponds to the heart, right cun to the lungs, left guan to the liver, right guan to the spleen, left chi to the kidneys, and right chi to the mingmen (GV4). These are combined to form the energy column vector of the six organ meridians. ;

[0100] Mapping rule definition: When the pulse is generally weak, with low tension and slow pulsation, a negative score is output. The closer the score is to the actual pulse position, the better. The more severe the deficiency and cold, the higher the score. When the pulse is strong, powerful, and rapid, a positive score is output. The closer the value is to 10, the more severe the excess heat. When the pulse is calm and balanced, the corresponding score is close to 0, which indicates that the cold and heat of the internal organs and meridians are balanced and the energy is stable.

[0101] S402: Based on the energy value vector of the six internal organs Combining the energy transmission weights of meridians with the same origin and interior-exterior connection, the energy values ​​of the six fu meridians are calculated through interior-exterior coupling transmission. ;

[0102] Calculate the total propagation weight of the incoming edges for any node i. For any node i, the energy it receives comes from two directions, as shown in the following formula:

[0103]

[0104] in, For the homologous pairing weights of the visceral nodes paired with visceral node i, such as large intestine ← lung, the homologous pairing weight in this example is taken as a baseline value of 0.7, and the homologous pairing weight is usually set between 0.70 and 0.80; based on the pathological transmission relationship between viscera and organs, other visceral nodes that indirectly affect visceral node i are obtained, such as large intestine ← liver. Based on the indirect path of liver disease transmitting to spleen, spleen disease to lung, and the lung and large intestine being paired, the sum of the weights of other visceral nodes that indirectly affect visceral node i is: , A value of 0.3 is acceptable, and it is typically set between 0.20 and 0.55. This is the set of other organ nodes that have an indirect influence on organ node i.

[0105] Weight of the directed edge from dirty node u to visceral node v Defined as:

[0106]

[0107] in, Let the dirty node u be the dirty meridian energy value, which is the starting point. The correlation strength coefficient between the internal organs is obtained through experiments. , This is the adjustment coefficient.

[0108]

[0109] in, The weight of energy transmission in meridians of the same origin and exterior; calculated A value of 0.7 indicates that 70% of the energy in the Fu meridians comes directly from the energy transmission of the Zang meridians, which share the same origin. A value of 0.3 indicates that the remaining 30% of the energy in the Fu meridians is regulated by the average energy level of the six internal organs as a whole, thus avoiding distortion of the Fu meridian values ​​caused by extreme fluctuations in the energy of a single organ.

[0110] Calculate the energy values ​​of the six fu meridians The formula is as follows:

[0111]

[0112] in, The energy value of the Fu meridian for the Fu nodes paired with the Zang nodes; The weight of energy transmission in meridians of the same origin and exterior; This refers to the energy value of the dirty node's internal meridian. The average base energy weight of the whole body; This represents the average energy level of the six internal organs.

[0113] The aforementioned exterior-interior coupling transmission is an energy weighted transmission mechanism based on the exterior-interior network relationship of the internal organs in traditional Chinese medicine. The energy of the paired internal organs is directly transmitted to the corresponding internal organs according to the coupling coefficient weight. At the same time, the average energy of the entire internal organ system is superimposed as a base compensation to quantitatively simulate the physiological energy transmission and pathological cold and heat synchronous transmission law of the distribution of internal organ qi to the internal organs through the exterior-interior network.

[0114] S403: Based on the energy values ​​of the twelve meridians Calculate the initial value of the patency of the corresponding meridian. ;

[0115] Energy values ​​of the twelve meridians That is, the calculated energy values ​​for each process. ;

[0116] Initial value of meridian patency The calculation is as follows:

[0117]

[0118] in, Recommended value: 0.15; This represents the energy value of the meridians. Characterizes the absolute extent to which the energy of the meridians deviates from the zero point of equilibrium; It is a hyperbolic tangent compression function;

[0119] In this invention, the energy values ​​of the twelve meridians (including the energy values ​​of the six viscera meridians and the six bowels meridians) obtained through pulse diagnosis can only characterize the cold and heat, deficiency and excess attributes of the meridians. The purpose of setting the initial value calculation step for meridian patency is to: transform the degree of cold and heat imbalance of meridian energy into a quantitative indicator of the degree of meridian blockage and stagnation, making up for the deficiency that meridian energy can only distinguish between cold and heat, deficiency and excess and cannot characterize the state of meridian passage. This initial value serves as a basic evaluation parameter, providing a standardized input for the quantitative calculation of meridian blockage classification and unblocking and regulation, and conforming to the evolutionary law of TCM pathogenesis of meridian obstruction caused by cold and heat imbalance in a nonlinear mapping manner.

[0120] S404: Parameter settings for evaluating pulse flowability, including correction coefficients for pulse flowability. The smoothness correction rule, based on the set smoothness correction rule, adjusts the initial smoothness value. The final unobstructed flow of the corresponding meridians is obtained through correction. ;

[0121] In S404 of this invention, pulse flowability is evaluated based on the pulse flowability coefficient obtained from the two-factor quantification index. The two-factor quantification index refers to the peak curvature variation coefficient. and the proportion of high-frequency energy Peak curvature variation coefficient The high-frequency energy proportion is obtained by taking the coefficient of variation of the second derivative of the signal within a certain period interval to the left and right of the main wave peak. The ratio of high-frequency energy to fundamental frequency energy is obtained by performing a Fast Fourier Transform (FFT) on the pulse signal.

[0122] Peak curvature variation coefficient The coefficient of variation of the second derivative of the signal is calculated by taking a 5% period interval to the left and right of the peak of the main wave. When the peak is smooth and rounded... Approaching 0; when there are sawtooth patterns or fluctuations at the peak. Significantly increased.

[0123] Perform a Fast Fourier Transform (FFT) on the pulse signal and determine the proportion of high-frequency energy. This was obtained by calculating the ratio of energy in the 5-15Hz high-frequency band to the energy in the 0.5-5Hz fundamental frequency band. The high-frequency energy of the hesitant pulse is significantly higher than that of the slippery pulse.

[0124] The pulse flow efficiency coefficient Fs is calculated using the following formula:

[0125]

[0126] when At this time, the pulse is slippery, indicating abundant qi and blood and good flow of the pulse; when At that time, the pulse was normal; when At this time, the pulse is choppy, indicating a higher degree of qi stagnation and blood stasis.

[0127] The pulse flowability correction coefficient function employs a piecewise linear mapping, determining the corresponding pulse flowability correction coefficient based on the value of the pulse flowability coefficient. This correction coefficient is used to initialize the meridian patency. Perform a global pulse diagnosis to uniformly correct the basic patency assessment results of individual meridians. The patency correction rules are as follows:

[0128]

[0129] The final patency level of the corresponding meridian is obtained:

[0130]

[0131] in, It is the final patency of the meridian corresponding to the initial value of patency after correction of the smoothness and roughness of the meridian and upper limit constraint. It is the standard patency index used for syndrome differentiation output.

[0132] Original meridian patency Calculations based solely on the energy-temperature deviation of a single meridian only reflect local obstruction caused by the imbalance of deficiency and excess in a single meridian. The pulse flow coefficient Fs is a macroscopic pulse characteristic of the six positions of the cun-kou (radial artery), reflecting the overall state of qi and blood circulation throughout the body. It contains global information on stagnation and fullness independent of the energy of a single meridian. This segmented correction model achieves a closed-loop correction between local meridian energy assessment and the overall smoothness and roughness of the pulse, making up for the one-sidedness of relying solely on the single dimension of meridian energy to assess smoothness.

[0133] S405: Outputs the energy distribution status of the twelve meridians and labels each meridian with diagnostic tags for deficiency / excess, cold / heat.

[0134]

[0135] The diagnostic labels for deficiency and excess include very excess, slightly excess, normal, slightly deficient, and very deficient; the diagnostic labels for cold and heat include very hot, slightly hot, normal, slightly cold, and very cold.

[0136] It can also recommend corresponding acupoints based on the diagnostic labels of deficiency / excess and cold / heat. For example, when the diagnostic label is very excess or very hot, it can be associated with acupoints that clear heat (such as Xingjian and Taichong), and when the diagnostic label is very deficiency or very cold, it can be associated with acupoints that warm and tonify.

[0137] Traditional Chinese medicine theory holds that a slippery pulse indicates abundant and smooth flow of Qi and blood, and high meridian patency; while a choppy pulse indicates Qi stagnation and blood stasis, and meridian obstruction. Even if the energy score is close to balance, if the overall pulse pattern is choppy, there is still meridian obstruction. This correction mechanism feeds back macroscopic pulse characteristics to microscopic meridian assessment, achieving closed-loop verification.

[0138] The dialectical analysis method described in this invention also includes S5: integrating multidimensional pulse characteristics, the visceral deficiency-excess index (VDE), and the final patency of the corresponding meridians. It generates a structured comprehensive diagnostic report that includes pulse quantitative parameter reports, organ syndrome differentiation reports, meridian status assessments, and personalized treatment plans.

[0139] By integrating pulse characteristic parameters, visceral syndrome differentiation results, and meridian status analysis results, a structured comprehensive pulse diagnosis analysis report is generated. The pulse quantitative parameter report includes pulse location, pulse rate, rhythm, pulse strength, tension, and fluency values ​​and trend graphs for each location (superficial, middle, and deep). The visceral syndrome differentiation report shows the deficiency / excess, cold / heat status of each viscera and corresponding common symptoms and disease risk warnings. The meridian status assessment includes energy distribution maps of each meridian and corresponding deficiency / excess, cold / heat labels. The personalized conditioning plan includes recommended acupoint locations, operation methods and frequencies, as well as dietary recommendations, lifestyle conditioning suggestions, etc.

[0140] This invention performs quantitative analysis on objective pulse wave signals acquired by a multi-gradient pressure sensor. Through a pre-defined three-dimensional decision tree rule, it transforms traditionally ambiguous pulse characteristics (such as floating, deep, wiry, and slippery) into clearly defined numerical ranges for quantitative evaluation, eliminating reliance on physician subjective experience or patient questionnaires. This eliminates human experience differences and ensures high consistency in test results for the same patient across different devices or time points.

[0141] This invention employs multi-gradient pressure synchronous acquisition technology to determine the optimal pulse-taking pressure by constructing an amplitude-pressure response curve. Compared to existing technologies that only utilize a single fixed pressure point to acquire pulse data, this invention can completely capture pulse information in the superficial, middle, and deep regions, solving the problems of misjudging superficial pulses and failing to detect deep pulses.

[0142] This invention constructs a directed graph transmission network based on the internal-external relationship of the Zang-Fu organs in Traditional Chinese Medicine, extrapolating the diagnostic results of a single Zang-Fu organ to the energy system of the twelve meridians. Through the internal-external coupling transmission mechanism, it quantifies and simulates the physiological energy transfer and pathological cold-heat transformation patterns of Zang-Fu organs as they circulate through the internal-external meridians.

[0143] This invention introduces a pulse flow coefficient based on a dual-factor quantitative index (peak curvature variation coefficient + high-frequency energy ratio), and sets a segmented linear patency correction rule accordingly. It uses the slippery and rough characteristics of the overall pulse at the six points of the cunkou to correct the patency assessment results of local meridians, thus overcoming the one-sidedness of relying solely on a single meridian energy dimension to assess patency.

[0144] This embodiment uses a 32-year-old male patient as an example to demonstrate the entire process of applying the method described in this invention to perform TCM pulse diagnosis and syndrome differentiation analysis and generate a structured comprehensive diagnostic report.

[0145] A multi-channel flexible array pressure sensor was used to simultaneously acquire pulse data at six standard pulse positions: left cun, right cun, left guan, right guan, left chi, and right chi. Multi-gradient pressure was achieved by setting the initial pressure value, the final pressure value, and the pressure step size. The pressure range was 0–160 mmHg, the step size was set to 20 mmHg, and there were a total of eight pressure gradients. After each pressure gradient stabilized, the raw pulse wave signal was continuously acquired for no less than 10 seconds.

[0146] In this embodiment, to demonstrate typical information, data from three representative pressure gradients—floating (low pressure), medium pressure, and deep pressure—are selected. Taking the left cun position as an example, the main peak of the raw pulse wave signal under each pressure gradient is detected, and the representative main wave amplitude is extracted. Using the left cun position as an example, the main wave amplitude under each pressure is fitted with a second-order polynomial to determine the optimal pulse-taking pressure corresponding to the largest amplitude. In this embodiment, the optimal pulse-taking pressure for the left cun position corresponds to the medium pressure state (approximately 120 mmHg), at which the main wave amplitude is the largest.

[0147] The table below shows the reference range of the multidimensional pulse features finally extracted from the original pulse wave signal and the multidimensional pulse features acquired from the left cun position:

[0148] Multidimensional pulse features were extracted under optimal pulse-taking pressure, including:

[0149] Pulse location: The pulse position at the left cun position is 224.59, which falls within the range of the central pulse. Pulse strength: The pulse strength under optimal pressure is 9.74 (lower than the lower reference limit of 16-25), indicating a weak pulse. Pulse rate: 91.83 beats / min, which is within the normal range. Tension: 1.19 (higher than the upper reference limit of 0.70), indicating a wiry pulse. Fluency: 0.36 (lower than the upper reference limit of 0.60), indicating a slightly hesitant pulse.

[0150] Based on the mapping relationship between the cunkou pulse and the internal organs, the pulse force P, pulse position L, and tension T under the optimal pressure of each part are normalized and then input into the preset three-dimensional decision tree rules to output the internal organ deficiency and excess index VDE and cold and heat attributes.

[0151] Taking the left radial artery (corresponding to the kidney and bladder) as an example, its optimal pulse pressure characteristics are: pulse force 62.96 (strong), pulse position 189.21 (deep pulse), and tension 0.29 (soft pulse). Based on the three-dimensional decision tree rule (strong pulse + deep pulse + soft pulse → VDE = +0.6, internal excess syndrome), and combined with a comprehensive judgment of other locations, it is concluded that the bladder is in excess syndrome, and the cold / heat attribute is heat.

[0152] The final assessment results based on the differentiation of syndromes of the internal organs were as follows: Bladder heat syndrome: damp heat descends and qi transformation is impaired; Bladder qi stagnation syndrome: qi stagnation and dysfunction of opening and closing; Bladder stones syndrome: damp heat stagnation and impurity deposition.

[0153] The energy values ​​of the six Zang-organ meridians and the six Fu-organ meridians are calculated. Ultimately, the meridian identification result for the Bladder Meridian, based on its deficiency / excess and cold / heat diagnostic labels, is "Extreme Excess" and "Extreme Heat." Clearing and purging acupoints are automatically matched, and based on the characteristics of the Bladder Meridian, the following recommendations are made:

[0154] Zhiyin acupoint: prick and bleed to clear away heat evil; Weizhong acupoint: prick and bleed to clear away bladder heat; Feiyang acupoint: massage or scrape to dredge the meridians; Shuidao acupoint: rub with the palm to facilitate water passage; Zhongji acupoint: massage to regulate bladder qi transformation.

[0155] Integrate all the above data to generate a report containing the following:

[0156] The pulse quantitative parameter report includes pulse location, pulse rate, rhythm, pulse strength, tension, and fluency values ​​and trend charts for each location (superficial, middle, and deep). The visceral syndrome differentiation report includes the diagnostic basis, symptom description, and related prescriptions for bladder heat syndrome, qi stagnation syndrome, and gallstone syndrome. The meridian status assessment includes energy distribution maps of each meridian and labels for deficiency / excess and cold / heat, as shown in the appendix. Figure 2 As shown, the personalized treatment plan includes recommended acupoint locations, operation methods and frequencies, as well as dietary recommendations, lifestyle suggestions, etc.

[0157] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A diagnostic analysis method based on multi-gradient pulse data, characterized in that: Includes the following steps: S1: Simultaneous acquisition of pulse wave signals at the cun, guan, and chi positions of the patient's left and right hands using multi-gradient pressure. The raw pulse wave signal is acquired after the pressure stabilizes at each pressure gradient. S2: Extract the main wave amplitude of the original pulse wave signal under each pressure gradient, determine the applied pressure value corresponding to the maximum main wave amplitude as the optimal pulse-taking pressure, and extract the multidimensional pulse features of the original pulse wave signal under the optimal pulse-taking pressure; the multidimensional pulse features include pulse position, pulse strength, pulse rate, pulse tension and pulse flow. S3: Based on multi-dimensional pulse characteristics, the preset three-dimensional decision tree rules are called to perform a quantitative evaluation of each organ to obtain organ differentiation evaluation parameters, and the organ differentiation evaluation results are output according to the organ differentiation evaluation parameters. S4: Construct a directed graph transmission network based on the organ differentiation assessment parameters and the exterior-interior and interior-exterior relationships of the twelve meridians, calculate the energy values ​​of the twelve meridians, and output the deficiency-excess, cold-heat diagnostic labels of each meridian according to the numerical range of the energy values ​​of the twelve meridians.

2. The dialectical analysis method according to claim 1, characterized in that: In S1, multi-gradient pressurization is achieved by setting the initial pressure value, the final pressure value, and the pressurization step size.

3. The dialectical analysis method according to claim 1, characterized in that: The process for determining the optimal pulse sampling pressure in S2 is as follows: S201: Detect the main peak of the raw pulse wave signal under each pressure gradient and extract the representative main peak amplitude of each cardiac cycle; S202: Construct an amplitude-pressure response scatter plot with the applied pressure value under each pressure gradient as the x-axis and the corresponding representative main wave amplitude value as the y-axis. After curve fitting, determine the applied pressure value corresponding to the maximum representative main wave peak amplitude as the optimal pulse sampling pressure.

4. The dialectical analysis method according to claim 3, characterized in that: The main peak detection in S201 includes the following steps: The original pulse wave signal is preprocessed to obtain the basic pulse wave signal, and candidate peaks are obtained by detecting all local maxima points of the basic pulse wave signal using basic mathematical methods. A dual judgment mechanism of adaptive threshold and time-domain constraint is used to screen candidate peaks to obtain the main peak amplitude sequence; An adaptive peak threshold function that varies with time is constructed using the sliding window method. If the amplitude of a candidate peak at a certain moment is greater than or equal to the adaptive peak threshold at the corresponding moment, it is retained as a candidate peak to be screened in the time domain; otherwise, the candidate peak is considered to be noise or interference peak and noise is removed. A lower limit for the physiological reentry period is set, which is the minimum time window. If the time interval between adjacent candidate peaks is less than the lower limit for the physiological reentry period, it is determined that there is a diphtheria wave or noise. For two adjacent candidate peaks, the one with the larger amplitude is retained.

5. The dialectical analysis method according to claim 1, characterized in that: The S3 visceral differentiation assessment parameters include the visceral deficiency-excess index and cold-heat attribute; a positive visceral deficiency-excess index represents an excess syndrome, a negative visceral deficiency-excess index represents a deficiency syndrome, and a visceral deficiency-excess index of 0 represents a neutral syndrome. Combined with the visceral mapping relationship of the Cunkou pulse at each location, the visceral differentiation assessment results including the heart, liver, spleen, lung, kidney and the corresponding six fu organs are generated.

6. The dialectical analysis method according to claim 1, characterized in that: S4 calculates the energy values ​​of the twelve meridians and outputs diagnostic labels for each meridian based on the numerical range of these energy values, mapping them to their deficiency / excess, cold / heat patterns. The energy values ​​of the twelve meridians include the energy values ​​of the six Zang meridians and the six Fu meridians. The steps are as follows: S401: Obtaining the energy values ​​of the six viscera meridians based on the viscera differentiation assessment parameters; S402: Calculate the energy values ​​of the six viscera meridians by combining the energy transmission weights of the same internal and external meridians through internal-external coupling transmission; S403: Calculate the initial value of the unobstructedness of the corresponding meridian based on the energy values ​​of each meridian in the twelve meridian energy values; S404: Combine the parameters for evaluating pulse flowability with the patency correction rule for the pulse flowability correction coefficient. The initial patency value is corrected according to the set patency correction rule to obtain the final patency of the corresponding meridian. S405: Outputs the energy distribution status of the twelve meridians and labels each meridian with diagnostic tags for deficiency / excess, cold / heat.

7. The dialectical analysis method according to claim 6, characterized in that: In S404, the parameter used to evaluate pulse flowability is the pulse flowability coefficient, which is obtained based on a two-factor quantization index. The two-factor quantization index refers to the peak curvature variation coefficient and the high-frequency energy ratio. The peak curvature variation coefficient is obtained by taking the variation coefficient of the second derivative of the signal within a certain period interval to the left and right of the main wave peak. The high-frequency energy ratio is obtained by calculating the ratio of high-frequency energy to fundamental frequency energy after performing a fast Fourier transform on the pulse signal.

8. The dialectical analysis method according to claim 7, characterized in that, The smoothness correction rule for the pulse flowability correction coefficient in S404 is as follows: the pulse flowability correction coefficient function adopts a piecewise linear mapping, and the corresponding pulse flowability correction coefficient is determined according to the value of the parameter for evaluating pulse flowability.

9. The dialectical analysis method according to claim 1, characterized in that: In S4, the diagnostic labels for deficiency / excess and cold / heat include deficiency / excess and cold / heat. The deficiency / excess diagnostic labels include very excess, slightly excess, normal, slightly deficiency, and very deficiency; the cold / heat diagnostic labels include very heat, slightly heat, normal, slightly cold, and very cold.

10. The dialectical analysis method according to claim 1, characterized in that: The diagnostic analysis method also includes S5: generating a structured comprehensive diagnostic report that includes a pulse quantitative parameter report, an organ differentiation report, a meridian status assessment, and a personalized conditioning plan.

Images