Wrist pulse spectrum fitting and analyzing method

Abstract

The invention provides a wrist pulse spectrum fitting and analysis method, which comprises the following steps of: converting a pressure signal into an electric signal, fitting to generate a human body pulse condition change curve, calculating from four dimensions of pulse position, pulse number, pulse shape and pulse potential, calculating the pulse position from two dimensions of depth and length; the pulse number is calculated from the two dimensions of frequency and rhythm, the pulse shape is calculated from the three dimensions of width, tension and fullness, the pulse potential is calculated from the three dimensions of fluency, strength and gradient, the pulse condition is analyzed by a reasonable algorithm, the wrist pulse spectrum is reasonably subjected to qualitative treatment in the aspect of traditional Chinese medicine pulse diagnosis, the accuracy of manual pulse taking is improved, and the pulse diagnosis accuracy is improved. The sensor can continuously and objectively monitor the pulse change of the patient under the influence of the mood of the pulse-taking person and external environment factors, so that more objective pulse-taking data can be obtained.

Description

Technical Field

[0001] The invention relates to the technical field of pulse image processing, in particular to a wrist pulse spectrum fitting and analysis method. Background Art

[0002] The pulse is mainly generated by the diastole and contraction of the human arteries. Most of the pulses on the market are measured by photoelectric sensors, which are used for detection of user fingers or wrist detection. Some are also detected by pressure sensors, but they are large in size, not convenient for carrying, and not suitable for home and clinic use. Traditional Chinese medicine pulse diagnosis mainly takes the pulse at the inch and mouth for palpation and pulse measurement, but it is highly subjective and difficult to standardize. The existing pulse diagnosis instruments or measuring equipment are based on empirical positioning or manual auxiliary positioning, and the positioning is inaccurate or inconvenient. If the positioning is inaccurate, the results of the collection will deviate, and the analysis results obtained may also vary greatly. In particular, the pulse of the human body is not very different in some different diseases and symptoms, and inaccurate positioning will affect the detection results. In addition, the pulse is greatly affected by the external environment, especially the temperature. If the influencing factors such as temperature are not fully considered, the measured pulse will also cause a large deviation.

[0003] To this end, some industry practitioners have developed pulse monitors, pulse monitoring belts and other equipment based on the improved sphygmomanometer to achieve the standardization, objectivity and concretization of Chinese medicine pulse diagnosis.

[0004] The Chinese invention patent with application number CN202111627390.1 discloses a wristband pulse diagnosis instrument, which is first worn at the high bone position according to the position of the human wrist, specifically at the high bone position of the wrist. At the same time, the innermost contact pressure sensor group can cover the vicinity of the cun pulse position. Before collection, an initialization operation is performed first, that is, the instrument will be automatically initialized after being worn, and the acquisition module collects the electrical signal of the contact pressure sensor group; then the electrical signal is transmitted to the processing unit, and the processing unit transmits the electrical signal to the low-pass filter unit for noise reduction filtering. After the noise reduction filtering, the zero phase shift filter unit performs phase distortion processing, and after the phase distortion processing is completed, it is input to the simulation unit, and the simulation unit is used to simulate the formation of an electrical signal waveform; the electrical signal waveform is compared with the set standard range waveform to obtain whether the position of the contact pressure sensor covers the cun pulse position. If not, the offset distance is obtained according to the comparison result, and the offset distance is sent to the control module, and the control module drives the X and Y axis moving parts to move to correspond to the coverage of the cun pulse position.

[0005] The Chinese invention patent with application number CN201810434690.X discloses a pulse collection wristband, including a wristband, a fixed plate detachably connected to the side wall of the wristband, a groove provided on the side wall of the fixed plate, three through holes provided on the inner bottom of the groove, a block fixedly connected to the inner wall of the groove, two first slide grooves and two second slide grooves symmetrically provided on the side wall of the fixed plate, a guide groove provided on the side wall of the fixed plate, a slider slidably connected in the two first slide grooves, a first spring provided between the slider and the side wall of the first slide groove, a gear rotatably connected to the inner wall of the two first slide grooves, a winding wheel fixedly connected to the side wall of the gear, and a rope provided on the side wall of the winding wheel. The pulse information collected by external reception is processed by a processor.

[0006] The above-mentioned prior art discloses indirectly realizing TCM pulse diagnosis through pressure detection equipment, by converting pressure signals into electrical signals, and fitting and generating a human pulse change curve after the conversion of electrical signals and pressure values, so as to replace the subjective judgment of manual pulse diagnosis. However, there is no corresponding processing method for the conversion distortion and image processing problems existing in the process of converting from electrical signals to human pulse change curves, resulting in the problem that the TCM pulse diagnosis standardization and objectification process realized by such means has poor accuracy and cannot truly reflect the actual pulse changes. Summary of the invention

[0007] The present invention provides a wrist pulse spectrum fitting and analysis method, which solves the above problems existing in the prior art.

[0008] The technical solution of the present invention is achieved as follows: a wrist pulse spectrum fitting and analysis method, characterized in that it includes the following steps: collecting human wrist pulse signals through a pulse monitor, converting the signals into wrist pulse spectrum, and fitting them into a pulse fitting curve, analyzing the pulse fitting curve, and qualitatively analyzing it according to a pre-set analysis algorithm, thereby achieving objective and standardized wrist pulse diagnosis.

[0009] As a preferred technical scheme, the pulse fitting curve is analyzed by following the steps:

[0010] Step S1: Calculate the wrist pulse from four dimensions: pulse position, pulse rate, pulse shape and pulse strength;

[0011] Step S2: Calculate the pulse position from the two dimensions of depth and length, and divide the pulse position into shallow pulse position, deep pulse position and very deep pulse position by comparing the single average wave height among the second fitting curve, the third fitting curve, the fourth fitting curve and the fifth fitting curve of this single average wave; and divide the pulse position into short pulse position and long pulse position by comparing the ratio of the wave area above the wave waist to the wave area below the wave waist of the reference wave of this single average wave with the system average value;

[0012] Step S3: Calculate the pulse from two dimensions: frequency and rhythm. Compare the reference wave of this single average wave with the system average value. Count the 15-second peaks to divide the pulse frequency into slow, slightly slow, fast and very fast. Compare the wave height or wavelength within the reference wave curve to divide the rhythm into irregular and paused.

[0013] Step S4: The pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count of the single average wave benchmark wave compared with the system average value; the tension is divided into tight and loose according to the y-axis value of the peak of the single average wave benchmark wave compared with the system average value; the fullness is divided into open and empty according to the y-axis value of the trough of multiple single average waves compared with each other;

[0014] Step S5: Calculate the pulse from three dimensions: fluency, strength and slope. Compare the current single average wave reference wave with the system average value to divide the fluency into smooth and rough; compare the current single average wave reference wave with the system average value, and divide the strength into weak and strong by the ratio of the single average wave area to the single average wave circumference; compare the current single average wave reference wave with the system average value, and divide the slope into gentle and rapid by comparing the ratio of the peak x-axis value to the wavelength.

[0015] As a preferred technical solution, in step S2: the pulse depth is determined by comparing the single average wave heights of the second single average wave, the third single average wave, the fourth single average wave, and the fifth single average wave.

[0016] When the following conditions are met: sH② / sH③>constant G∩Camax=Ca②, it is determined to be a superficial pulse;

[0017] When the following conditions are met: sH④ / sH③>constant G^2∩sH② / sH③<1, it is determined to be a deep pulse position;

[0018] When the following condition is satisfied: sH⑤ / sH④> constant G^2∩Camax=Ca⑤, it is determined to be an extremely deep pulse position.

[0019] Otherwise, the pulse is judged to be neither deep nor shallow;

[0020] The pulse length is determined by comparing the single average wave c with the system average value and the ratio of the wave area above the waist to the wave area below the waist:

[0021] When satisfied: When , it is judged as short pulse position;

[0022] When satisfied: When , it is determined to be a long pulse position;

[0023] Otherwise, the pulse is judged to be neither long nor short.

[0024] As a preferred technical solution, the single average wave reference wave is compared with the system average value to count the 15-second peak value to determine the frequency of the pulse.

[0025] When ν / τν<constant G^-1.25 is satisfied, it is judged as: slow

[0026] When the constant G^-1.25≤ν / τν<constant G^-0.5 is satisfied, it is judged as: slightly slow

[0027] When the constant G^1≤ν / τν<constant G^2.5 is satisfied, it is judged as: fast

[0028] When the constant G^2.5≤ν / τν is satisfied, it is judged as: extremely fast

[0029] When the constant G^-0.5≤ν / τν<constant G^1 is satisfied, it is judged as: neither fast nor slow;

[0030] This method divides the rhythm into irregular and paused rhythms by comparing the wave height or wavelength within the sub-reference wave curve. Specifically:

[0031] When satisfied: When , it is judged as: uneven;

[0032] When satisfied: When , it is judged as: pause;

[0033] Otherwise, it is judged as: the rhythm is even.

[0034] As a preferred technical solution, the pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count when comparing the single average wave reference wave with the system average value; the tension is divided into tight and loose according to the y-axis value of the peak when comparing the single average wave reference wave with the system average value; the fullness is divided into open and empty according to the y-axis value of the trough when comparing multiple single average waves.

[0035] As an optimal technical solution, this single mean wave Compare the ratio of the single-average wave area to the 15-second peak count with the system average:

[0036] When satisfied: When , it is judged as: wide;

[0037] When satisfied: When it is judged as: fine;

[0038] Otherwise it is judged as neither big nor small.

[0039] As an optimal technical solution, this single mean wave Compare the peak y-axis value with the system average to determine the tension, specifically:

[0040] When satisfied: When , it is judged as: tight;

[0041] When satisfied: When it is judged as: soft;

[0042] Otherwise it is judged as: neither loose nor tight.

[0043] As a preferred technical solution, the fullness is determined by comparing the y-axis values ​​of the troughs between the single average waves ②, ③, and ④, including

[0044] When satisfied: When it is judged as: slippery;

[0045] When satisfied: When it is judged as: difficult

[0046] Otherwise it is judged as: neither slippery nor astringent.

[0047] As an optimal technical solution, the pulse is calculated from three dimensions: fluency, strength and slope. The comparison between the single average wave reference wave and the system average value divides the fluency into smooth and difficult; the comparison between the single average wave reference wave and the system average value divides the strength into weak and strong by the ratio of the single average wave area to the single average wave circumference; the comparison between the peak x-axis value and the wavelength within the single average wave reference wave divides the slope into gentle and rapid; specifically:

[0048] When the following conditions are met: ∑sH / τ∑sH<constant G^-0.5∩ΔsH<constant G^-6, it is judged as: powerless;

[0049] When the following conditions are met: ∑sH / τ∑sH>constant G^0.5∩ΔsH<constant G^-6, it is judged as: powerful;

[0050] Otherwise, it is determined to be balanced in strength;

[0051] When satisfied: Judged as: slow;

[0052] When satisfied: When it is judged as: urgent;

[0053] Otherwise it is judged as calm and mild.

[0054] As a preferred technical solution, the pulse conditions are divided into floating pulse, sinking pulse, hidden pulse, slow pulse, slow pulse, rapid pulse, slippery pulse, wiry pulse, weak pulse, strong pulse, short pulse, long pulse, thin pulse, string pulse, moist pulse, intermittent pulse, wiry pulse, scattered pulse, firm pulse, arterial pulse, weak pulse, subtle pulse, surging pulse, tight pulse, leathery pulse, knotted pulse and hurried pulse according to the judgment results of pulse position, pulse rate, pulse shape and pulse strength.

[0055] Compared with the existing technology, this solution has the following beneficial effects: the wrist pulse is calculated from four dimensions: pulse position, pulse rate, pulse shape, and pulse strength; the pulse is analyzed with a reasonable algorithm; the wrist pulse spectrum is rationally characterized from the perspective of traditional Chinese medicine pulse diagnosis, thereby improving the accuracy of manual pulse diagnosis; it is affected by the mood of the person taking the pulse and external environmental factors, and the sensor can continuously and objectively monitor the patient's pulse changes, thereby obtaining more objective pulse diagnosis data. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 A pulse condition determination table for a wrist pulse spectrum fitting and analysis method of the present invention;

[0058] Figure 2 A pulse classification table for a wrist pulse spectrum fitting and analysis method of the present invention;

[0059] Figure 3 It is one of the pulse diagnosis results of the wrist pulse spectrum fitting and analysis method of the present invention;

[0060] Figure 4 The second pulse diagnosis result of the wrist pulse spectrum fitting and analysis method of the present invention;

[0061] Figure 5 This is the third pulse diagnosis result of the wrist pulse spectrum fitting and analysis method of the present invention. DETAILED DESCRIPTION

[0062] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] Reference Figure 1 A wrist pulse spectrum fitting and analysis method is characterized in that it comprises the following steps: collecting a human wrist pulse signal through a pulse monitor, converting the signal into a wrist pulse spectrum, and fitting it into a pulse fitting curve; analyzing the pulse fitting curve and qualitatively analyzing it according to a pre-set analysis algorithm, thereby achieving objective and standardized wrist pulse diagnosis.

[0064] Specifically, the pulse fitting curve is analyzed by following the steps:

[0065] Step S1: Calculate the wrist pulse from four dimensions: pulse position, pulse rate, pulse shape and pulse strength;

[0066] Step S2: Calculate the pulse position from the two dimensions of depth and length, and divide the pulse position into shallow pulse position, deep pulse position and very deep pulse position by comparing the single average wave height among the second fitting curve, the third fitting curve, the fourth fitting curve and the fifth fitting curve of this single average wave; and divide the pulse position into short pulse position and long pulse position by comparing the ratio of the wave area above the wave waist to the wave area below the wave waist of the reference wave of this single average wave with the system average value;

[0067] Step S3: Calculate the pulse from two dimensions: frequency and rhythm. Compare the reference wave of this single average wave with the system average value. Count the 15-second peaks to divide the pulse frequency into slow, slightly slow, fast and very fast. Compare the wave height or wavelength within the reference wave curve to divide the rhythm into irregular and paused.

[0068] Step S4: The pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count of the single average wave benchmark wave compared with the system average value; the tension is divided into tight and loose according to the y-axis value of the peak of the single average wave benchmark wave compared with the system average value; the fullness is divided into open and empty according to the y-axis value of the trough of multiple single average waves compared with each other;

[0069] Step S5: Calculate the pulse from three dimensions: fluency, strength and slope. Compare the current single average wave reference wave with the system average value to divide the fluency into smooth and rough; compare the current single average wave reference wave with the system average value, and divide the strength into weak and strong by the ratio of the single average wave area to the single average wave circumference; compare the current single average wave reference wave with the system average value, and divide the slope into gentle and rapid by comparing the ratio of the peak x-axis value to the wavelength.

[0070] Specifically, the pulse depth is determined by comparing the single average wave heights of the second single average wave, the third single average wave, the fourth single average wave, and the fifth single average wave.

[0071] When the following conditions are met: sH② / sH③>constant G∩Camax=Ca②, it is determined to be a superficial pulse;

[0072] When the following conditions are met: sH④ / sH③>constant G^2∩sH② / sH③<1, it is determined to be a deep pulse position;

[0073] When the following condition is satisfied: sH⑤ / sH④> constant G^2∩Camax=Ca⑤, it is determined to be an extremely deep pulse position.

[0074] Otherwise, the pulse is judged to be neither deep nor shallow;

[0075] The pulse length is determined by comparing the single average wave c with the system average value and the ratio of the wave area above the waist to the wave area below the waist:

[0076] When satisfied: When , it is judged as short pulse position;

[0077] When satisfied: When , it is determined to be a long pulse position;

[0078] Otherwise, the pulse is judged to be neither long nor short.

[0079] The single average wave reference wave is compared with the system average value for 15 seconds to count the peak value to determine the pulse frequency. When it satisfies: ν / τν<constant G^-1.25, it is determined to be: slow

[0080] When the constant G^-1.25≤ν / τν<constant G^-0.5 is satisfied, it is judged as: slightly slow

[0081] When the constant G^1≤ν / τν<constant G^2.5 is satisfied, it is judged as: fast

[0082] When the constant G^2.5≤ν / τν is satisfied, it is judged as: extremely fast

[0083] When the constant G^-0.5≤ν / τν<constant G^1 is satisfied, it is judged as: neither fast nor slow;

[0084] Rhythms are divided into irregular and pauses by comparing the wave height or wavelength within the sub-reference wave curve. Specifically:

[0085] When satisfied: When , it is judged as: uneven;

[0086] When satisfied: When , it is judged as: pause;

[0087] Otherwise, it is judged as: the rhythm is even.

[0088] The pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count when comparing the single average wave benchmark wave with the system average. The tension is divided into tight and loose according to the y-axis value of the peak when comparing the single average wave benchmark wave with the system average. The fullness is divided into stretch and empty according to the y-axis value of the trough when comparing multiple single average waves.

[0089] This single mean wave Compare the ratio of the single-average wave area to the 15-second peak count with the system average:

[0090] When satisfied: When , it is judged as: wide;

[0091] When satisfied: When it is judged as: fine;

[0092] Otherwise it is judged as neither big nor small.

[0093] Wherein, sC: single mean wave crest y-axis value; C: wave crest (systolic pressure), H: wave height (amplitude / pulse pressure), sV: single mean wave trough y-axis value, V: wave trough (diastolic pressure), aH: average wave height, sW: single mean wave waist y-axis value, sW=(sC+2×sV) / 3, aL: average wavelength, Single mean wave The portion of the wave area above the wave waist, Single mean wave The portion of the wave area below the wave waist.

[0094] in,

[0095] τ∑sH=system average ∑sH, <==statistical system average τ∑sH.

[0096] This single mean wave Compare the peak y-axis value with the system average to determine the tension, specifically:

[0097] When satisfied: When , it is judged as: tight;

[0098] When satisfied: When it is judged as: soft;

[0099] Otherwise it is judged as: neither loose nor tight.

[0100] Among them, sL: single mean wavelength x-axis value, ν: heart rate, Wave 15 seconds peak count, if ν=ν2+ν3+ν4+ν5,τν=system average ν,counts per second is

[0101]

[0102] sAe: single average wave area, ∑sH = (sH② + sH③ + sH④) / 3, τ∑sH = system average ∑sH, Single mean wave Peak y-axis value, Single mean wave The trough y-axis value, sAc: single mean wave circumference.

[0103]

[0104] The fullness is determined by comparing the y-axis values ​​of the troughs between the single average waves ②, ③, and ④, including:

[0105] When satisfied: When it is judged as: slippery;

[0106] When satisfied: When it is judged as: difficult;

[0107] Otherwise it is judged as: neither slippery nor astringent.

[0108] in,

[0109]

[0110] At the same time, a suggested pulse diagnosis method is provided based on the above algorithm: ∈ = (number of positives × f × e-number of negatives) / (number of positives + number of negatives + 1), the weight of the numerator with brackets f = 2 (1 positive is counted as 2), the weight of the numerator with brackets f = 1 / 2 (1 positive is counted as 0.5), ∈ ≥ 1 is established, and the three established ones with the largest scores are imported. If there is no ∈ ≥ 1, the one with the largest score is imported. Therefore, according to the judgment results of pulse position, pulse rate, pulse shape, and pulse strength, the pulse is divided into floating pulse, sinking pulse, latent pulse, slow pulse, slow pulse, rapid pulse, rapid pulse, slippery pulse, astringent pulse, weak pulse, full pulse, short pulse, long pulse, thin pulse, string pulse, moist pulse, alternate pulse, wiry pulse, scattered pulse, firm pulse, arterial pulse, weak pulse, micro pulse, surging pulse, tight pulse, leather pulse, knotted pulse, and accelerated pulse.

[0111] Figures 3 to 5 Some of the pulse diagnosis results are shown below.

[0112] The human pulse is collected through a wearable pulse monitor. In combination with the corresponding pulse processing method, the principle of feeling the pulse wave change by simulating the traditional Chinese medicine finger pressing and taking the pulse is simulated. The wristband airbag is pressurized to simulate the manual pressing pressure. The sensor monitors and collects the pulse change data after pressurization. After the AI ​​algorithm completes the conversion of electrical signals and pressure values, it fits and generates the human pulse change curve to replace the subjective judgment of manual pulse taking. The wrist pulse is calculated from four dimensions: pulse position, pulse rate, pulse shape, and pulse strength. The pulse position is calculated from two dimensions: depth and length. The pulse rate is calculated from two dimensions: frequency and rhythm. The pulse shape is calculated from three dimensions: width, tension and fullness. The pulse strength is calculated from three dimensions: fluency, strength and slope. The pulse is analyzed with a reasonable algorithm to improve the accuracy of manual pulse taking. The pulse is affected by the mood of the pulse taker and external environmental factors. The sensor can continuously and objectively monitor the patient's pulse changes, thereby obtaining more objective pulse diagnosis data.

[0113] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A wrist pulse spectrum fitting and analysis method, characterized in that: The following steps are involved: The wrist pulse signal of the human body is collected by a pulse monitor, and the signal is converted into a wrist pulse spectrum and fitted into a pulse fitting curve. The pulse fitting curve is analyzed and qualitatively characterized according to a pre-set analysis algorithm, thereby achieving objective and standardized wrist pulse diagnosis.

2. A wrist pulse spectrum fitting and analysis method as claimed in claim 1, characterized in that: The pulse fitting curve is analyzed by following the steps: Step S1: Calculate the wrist pulse from four dimensions: pulse position, pulse rate, pulse shape and pulse strength; Step S2: Calculate the pulse position from the two dimensions of depth and length, and divide the pulse position into shallow pulse position, deep pulse position and very deep pulse position by comparing the single average wave height among the second fitting curve, the third fitting curve, the fourth fitting curve and the fifth fitting curve of this single average wave; and divide the pulse position into short pulse position and long pulse position by comparing the ratio of the wave area above the wave waist to the wave area below the wave waist of the reference wave of this single average wave with the system average value; Step S3: Calculate the pulse from two dimensions: frequency and rhythm. Compare the reference wave of this single average wave with the system average value. Count the 15-second peaks to divide the pulse frequency into slow, slightly slow, fast and very fast. Compare the wave height or wavelength within the reference wave curve to divide the rhythm into irregular and paused. Step S4: The pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count of the single average wave benchmark wave compared with the system average value; the tension is divided into tight and loose according to the y-axis value of the peak of the single average wave benchmark wave compared with the system average value; the fullness is divided into open and empty according to the y-axis value of the trough of multiple single average waves compared with each other; Step S5: Calculate the pulse from three dimensions: fluency, strength and slope. Compare the current single average wave reference wave with the system average value to divide the fluency into smooth and rough; compare the current single average wave reference wave with the system average value, and divide the strength into weak and strong by the ratio of the single average wave area to the single average wave circumference; compare the current single average wave reference wave with the system average value, and divide the slope into gentle and rapid by comparing the ratio of the peak x-axis value to the wavelength.

3. A wrist pulse spectrum fitting and analysis method as claimed in claim 2, characterized in that: In step S2: the pulse depth is determined by comparing the single average wave heights of the second single average wave, the third single average wave, the fourth single average wave, and the fifth single average wave. When the following conditions are met: sH② / sH③>constant G∩Camax=Ca②, it is determined to be a superficial pulse; When: sH④ / sH③> constant G 2 When ∩sH② / sH③<1, it is judged as a deep pulse; When: sH⑤ / sH④> constant G 2 When ∩Camax=Ca⑤, it is judged as a very deep pulse. Otherwise, the pulse is judged to be neither deep nor shallow; The pulse length is determined by comparing the single average wave c with the system average value and the ratio of the wave area above the waist to the wave area below the waist: When satisfied: When , it is judged as short pulse position; When satisfied: When , it is determined to be a long pulse position; Otherwise, the pulse is judged to be neither long nor short.

4. A wrist pulse spectrum fitting and analysis method as claimed in claim 1, characterized in that: This single average wave reference wave is compared with the system average value to count the 15-second peaks to determine the frequency of the pulse. When ν / τν<constant G^-1.25 is satisfied, it is judged as: slow When the constant G^-1.25≤ν / τν<constant G^-0.5 is satisfied, it is judged as: slightly slow When the constant G^1≤ν / τν<constant G^2.5 is satisfied, it is judged as: fast When the constant G^2.5≤ν / τν is satisfied, it is judged as: extremely fast When the constant G^-0.5≤ν / τν<constant G^1 is satisfied, it is judged as: neither fast nor slow; This method divides the rhythm into irregular and paused rhythms by comparing the wave height or wavelength within the sub-reference wave curve. Specifically: When satisfied: When , it is judged as: uneven; When satisfied: When , it is judged as: pause; Otherwise, it is judged as: the rhythm is even.

5. A wrist pulse spectrum fitting and analysis method as claimed in claim 1, characterized in that: The pulse shape is calculated from three dimensions: width, tension and fullness. The width is divided into wide and thin according to the ratio of the single average wave area to the 15-second peak count when comparing the single average wave benchmark wave with the system average. The tension is divided into tight and loose according to the y-axis value of the peak when comparing the single average wave benchmark wave with the system average. The fullness is divided into stretch and empty according to the y-axis value of the trough when comparing multiple single average waves.

6. A wrist pulse spectrum fitting and analysis method as claimed in claim 5, characterized in that: This single mean wave Compare the ratio of the single-average wave area to the 15-second peak count with the system average: When satisfied: When , it is judged as: wide; When satisfied: When it is judged as: fine; Otherwise it is judged as neither big nor small.

7. A wrist pulse spectrum fitting and analysis method as claimed in claim 5, characterized in that: This single mean wave Compare the peak y-axis value with the system average to determine the tension, specifically: When satisfied: When it is determined to be tight; when it meets: When it is judged as: soft; Otherwise it is judged as: neither loose nor tight.

8. A wrist pulse spectrum fitting and analysis method as claimed in claim 5, characterized in that: The fullness is determined by comparing the y-axis values ​​of the troughs between the single average waves ②, ③, and ④, including When satisfied: When it is judged as: slippery; When satisfied: When it is judged as: difficult; Otherwise it is judged as: neither slippery nor astringent.

9. A wrist pulse spectrum fitting and analysis method as claimed in claim 5, characterized in that: The pulse is calculated from three dimensions: fluency, strength and slope. The comparison between the single average wave reference wave and the system average value divides the fluency into smooth and difficult; the comparison between the single average wave reference wave and the system average value divides the strength into weak and strong by the ratio of the single average wave area to the single average wave circumference; the comparison between the peak x-axis value and the wavelength within the single average wave reference wave divides the slope into gentle and rapid; specifically: When the following conditions are met: ∑sH / τ∑sH<constant G^-0.5∩ΔsH<constant G^-6, it is judged as: powerless; When the following conditions are met: ∑sH / τ∑sH>constant G^0.5∩ΔsH<constant G^-6, it is judged as: powerful; Otherwise, it is determined to be balanced in strength; When satisfied: Judged as: slow; When satisfied: When it is judged as: urgent; Otherwise it is judged as calm and mild.

10. A wrist pulse spectrum fitting and analysis method, characterized in that: According to the judgment results of pulse position, pulse rate, pulse shape and pulse strength, the pulse conditions are divided into floating pulse, sinking pulse, hidden pulse, slow pulse, slow pulse, rapid pulse, fast pulse, slippery pulse, wiry pulse, weak pulse, strong pulse, short pulse, long pulse, thin pulse, string pulse, moist pulse, intermittent pulse, wiry pulse, scattered pulse, firm pulse, arterial pulse, weak pulse, subtle pulse, surging pulse, tight pulse, leathery pulse, knotted pulse and hurried pulse.

Images