CN100493445C - Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter - Google Patents
Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter Download PDFInfo
- Publication number
- CN100493445C CN100493445C CNB2005100613942A CN200510061394A CN100493445C CN 100493445 C CN100493445 C CN 100493445C CN B2005100613942 A CNB2005100613942 A CN B2005100613942A CN 200510061394 A CN200510061394 A CN 200510061394A CN 100493445 C CN100493445 C CN 100493445C
- Authority
- CN
- China
- Prior art keywords
- arteries
- veins
- pulse
- signal
- characteristic parameter
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
- 238000012360 testing method Methods 0.000 title claims description 19
- 238000000034 method Methods 0.000 claims abstract description 32
- 239000003814 drug Substances 0.000 claims abstract description 11
- 210000001367 artery Anatomy 0.000 claims description 144
- 210000003462 vein Anatomy 0.000 claims description 144
- 238000002372 labelling Methods 0.000 claims description 23
- 230000009466 transformation Effects 0.000 claims description 23
- 238000001514 detection method Methods 0.000 claims description 18
- 238000000354 decomposition reaction Methods 0.000 claims description 12
- 238000011430 maximum method Methods 0.000 claims description 12
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 238000003745 diagnosis Methods 0.000 claims description 8
- 238000000205 computational method Methods 0.000 claims description 5
- 210000000707 wrist Anatomy 0.000 claims description 4
- 238000012546 transfer Methods 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 14
- 238000012545 processing Methods 0.000 abstract description 5
- 230000008786 sensory perception of smell Effects 0.000 abstract 1
- 230000006870 function Effects 0.000 description 17
- 238000005516 engineering process Methods 0.000 description 9
- 230000004927 fusion Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 239000000872 buffer Substances 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 230000000452 restraining effect Effects 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 239000008280 blood Substances 0.000 description 1
- 210000004369 blood Anatomy 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 230000036285 pathological change Effects 0.000 description 1
- 231100000915 pathological change Toxicity 0.000 description 1
- 230000035479 physiological effects, processes and functions Effects 0.000 description 1
- 238000004451 qualitative analysis Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
- 210000001835 viscera Anatomy 0.000 description 1
Images
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
A method for automatically detecting the characteristic parameters of pulse state of the traditional Chinese medicine is based on the probability reasoning model of Bayes network. The modern signal processing technique and image recognizing technique are used to take the characteristics of tongue manifestation and pulse state for automatic analysis and judgement. After the result is integrated with the information obtained by ausculation, olfaction and interrogation, the intelligent computer-aided discrimination can be performed.
Description
Technical field
The present invention relates to the traditional Chinese medical science and quantize diagnostic field, is that a kind of realization sphygmogram characteristic parameters such as computer technology, image-recognizing method, signal processing technology of utilizing are extracted the method for detection automatically.This method can accurately detect the corner position of arteries and veins figure, and calculates the characteristic parameter of arteries and veins figure on this basis automatically.
Background technology
Pulse-taking is by checking and the examination method of analyzing pulse condition variation understanding human qi and blood running's state and internal organs physiology, pathological change, important evidence being provided for diagnosing the illness.The tradition pulse-taking relies on the doctor to push wrist and scratches arterial pulse and diagnose, and has subjective sensation and experience accumulation that its accuracy of very strong subjectivity and ambiguity and reliability depend on doctor individual, lacks objective, quantized diagnosis index.In today of computer technology, rapid development of information technology, traditional pulse wave spectrum mode can not adapt to the needs of modern medicine development.Arteries and veins figure index is manually read in the many employings of present existing arteries and veins map analysis method, carries out pulse condition identification more on this basis, is difficult to realize pulse-taking automatization.
Summary of the invention
The objective of the invention is to overcome the deficiency that above-mentioned prior art exists, a kind of Chinese medicine pulse characteristic parameter automatic testing method is provided.
The pulse condition collecting device that the inventive method is made up of pulse diagnosis sensor and pulse-tracing collection circuit obtains pulse signal, and transfers to computer, in computer, finish the input pulse signal removed make an uproar, flex point detection and the calculating of arteries and veins graph parameter etc.The present invention combines the advanced technology and the method for other fields such as wavelet analysis, image recognition with traditional Chinese medical science tradition pulse-taking theory, realize the automatic extraction of sphygmogram characteristic parameter.
The present invention is achieved through the following technical solutions:
1. gather arteries and veins figure signal:
Pulse diagnosis sensor by adjustable in pressure places patient's wrist to scratch tremulous pulse place acquisition pulse voltage signal, is converted to digital signal via pulse-tracing collection circuit (built-in signal amplifying circuit, A/D analog to digital conversion circuit), by USB interface input computer.
Original pulse condition figure can be the signal that collects in real time by pulse image sensor and pulse-tracing collection circuit, also can be by being saved in the signal in the hard disc of computer after pulse image sensor and the collection of pulse-tracing collection circuit.
2. arteries and veins figure signal removes and to make an uproar: pulse condition is input in the computer processor, gathers pulse signal under the different pressures by programme-control, arteries and veins figure signal is filtered made an uproar.
The signal of the arteries and veins figure that the present invention is applied to wavelet analysis technology removes and makes an uproar; be used for arteries and veins figure and remove the vanishing moment of the wavelet basis function of making an uproar more than or equal to 2; and the wavelet decomposition yardstick is got a=4; can select sym8 small echo, bior3.5 small echo for use; can make arteries and veins figure noise obtain good restraining, and also protect arteries and veins figure signal undistorted.
3. the pulse condition characteristic parameter detects: by detecting arteries and veins figure flex point, carry out the labelling of flex point, according to corner position, the arteries and veins figure that chooses arteries and veins figure amplitude maximum carries out the detection of pulse condition characteristic parameter.
The present invention adopts multiple wavelet transformation modulus maximum method to be used for arteries and veins figure flex point and detects, and wherein wavelet basis function adopts the multiple small echo that the first derivative of Gaussian function becomes through the Hilbert shift conversion.This method can accurately detect required arteries and veins figure flex point, and testing result is not subjected to the influence of wavelet scale, does not therefore need the problem of considering that yardstick is selected, has simplified testing process greatly.
Adopt the chain code method to determine arteries and veins figure crest number during the labelling of flex point, then in conjunction with maximum method labelling tidal wave crest and heavy Bo Bobofeng, and by coming labelling tidal wave starting point and heavily rich ripple starting point in conjunction with calculating arteries and veins figure master crest top angle.
According to the inventive method, the arteries and veins graph parameter that needs to detect comprises the high h of main wave amplitude
1, main ripple rise time t
1, tidal wave starting point height h
2, tidal wave panel height h
3, heavily rich ripple starting point height h
4, the heavily rich high h of wave amplitude
5, arteries and veins figure period T and arteries and veins area of pictural surface A.Arteries and veins graph parameter h
1, h
2, h
3, h
4, h
5Can directly calculate according to corner position.The present invention adopts wavelet transformation modulus maximum method to determine arteries and veins figure period T exactly, and judges whether T equated on each time period; For arteries and veins area of pictural surface A, the present invention adopts the region area computational methods of two dimensional image to calculate, and has realized no inclined to one side with the consistent preferably estimation to the arteries and veins area of pictural surface.
The present invention has following characteristics compared with the prior art:
1. the present invention is applied to wavelet analysis technology the signal of arteries and veins figure except that making an uproar, solved 2 problems of present existence: (1) is when gathering pulse signal, can be subjected to the influence of aspects such as instrument, human body and produce interference, the arteries and veins figure flex point that can make these interfere informations detects the difficulty-flase drop that becomes and measures too much flex point, and can't determine which flex point is useful flex point; (2) simultaneously, in removing interferential process, also to preserve the shape of former arteries and veins figure as far as possible, in order to avoid lose the information that diagnostic value is arranged.
2. the present invention adopts multiple wavelet transformation modulus maximum method to be used for the detection of arteries and veins figure flex point, and wherein wavelet basis function adopts the multiple small echo that the first derivative of Gaussian function becomes through the Hilbert shift conversion.This method can accurately detect required arteries and veins figure flex point, and testing result is not subjected to the influence of wavelet scale, does not therefore need the problem of considering that yardstick is selected, has simplified testing process greatly.
3. the present invention adopts the chain code method to determine arteries and veins figure crest number, then in conjunction with maximum method labelling tidal wave crest and heavy Bo Bobofeng, and by coming labelling tidal wave starting point and heavily rich ripple starting point in conjunction with calculating arteries and veins figure master crest top angle.Solved owing to have the situation of tidal wave and the fusion of main ripple or the situation of tidal wave and heavily rich ripple fusion in the pulse signal, be that the tidal wave starting point overlaps with main crest or the tidal wave starting point overlaps with heavily rich ripple starting point, therefore the arteries and veins figure that collects presents double-humped wave sometimes, sometimes present three spikes, make the become problem of difficulty of the labelling of flex point.
4. the present invention adopts wavelet transformation modulus maximum method to determine arteries and veins figure period T exactly, and judges whether T equated on each time period; For arteries and veins area of pictural surface A, the present invention adopts the region area computational methods of two dimensional image to calculate, and has realized no inclined to one side with the consistent preferably estimation to the arteries and veins area of pictural surface.
The signal of the arteries and veins figure that the present invention is applied to wavelet analysis technology removes and makes an uproar, it is exactly with the band filter of different center frequency signal to be done filtering that arteries and veins figure signal is made multi-scale wavelet transformation essence, therefore the component of those yardsticks of main reflection noise frequency is removed or does and to a certain degree decay, again inversion is done in the wavelet transformation combination of each yardstick of residue, being reduced into does not have the arteries and veins figure signal of making an uproar, and has promptly reached the purpose of de-noising.
Utilizing wavelet transformation that arteries and veins figure signal is filtered in the process of making an uproar, the selection of wavelet basis is very important, and different wavelet basis functions decompose arteries and veins figure signal, can give prominence to the signal characteristic of different characteristics, therefore filters the result that makes an uproar and also has very big difference.Through discovering, be used for arteries and veins figure and remove the wavelet basis function of making an uproar and should have good symmetry, its vanishing moment should be more than or equal to 2, and the wavelet decomposition yardstick when getting a=4 the filter effect of making an uproar best.All make arteries and veins figure noise obtain good restraining such as sym8 small echo, the slight ripple of bior3.5 etc., and protected arteries and veins figure signal undistorted simultaneously.
The present invention mainly is the automatic detection of utilization computer realization arteries and veins graph parameter, and referring to Fig. 1, the arteries and veins graph parameter that needs to detect comprises the high h of main wave amplitude
1, main ripple rise time t
1, tidal wave starting point height h
2, tidal wave panel height h
3, heavily rich ripple starting point height h
4, the heavily rich high h of wave amplitude
5, arteries and veins figure period T and arteries and veins area of pictural surface A.These CALCULATION OF PARAMETERS all are to be based upon on the basis that accurately detects arteries and veins figure flex point (shown in Fig. 2 " * " number).Wavelet transformation has the characteristics of multiresolution, can give prominence to the signal local feature preferably, is well suited for being used for the detection of singular points such as flex point.It is unsatisfactory that singular point in the past detects the effect that adopts real wavelet transformations, discovery after deliberation, real wavelet transformation to be used for the detection of arteries and veins figure flex point more.This mainly is because there is vibration in the mould that real wavelet transformation obtains, and therefore the number of detected flex point need be provided with threshold value to the modulus maximum of wavelet transformation far away more than useful flex point.But the setting of threshold value is a difficult point, if threshold value is too low, then still there is useless flex point in testing result, makes subsequent treatment become complicated; If threshold value is too high, then the flex point that obtains of screening is less, and some important flex points will be lost.And desire to make the threshold size of setting just to detect required useful flex point, and almost be the task that impossible realize, the flex point that no matter detects is too much or very few, all can't be used for the arteries and veins graph parameter and calculate.In order to address this problem, the present invention adopts multiple wavelet transformation modulus maximum method to be used for arteries and veins figure flex point and detects, and wherein wavelet basis function adopts the multiple small echo that the first derivative of Gaussian function becomes through the Hilbert shift conversion.This method can accurately detect required arteries and veins figure flex point, and testing result is not subjected to the influence of wavelet scale, does not therefore need the problem of considering that yardstick is selected, has simplified testing process greatly.
Owing to have the situation of tidal wave and main ripple fusion or the situation of tidal wave and heavily rich ripple fusion in the pulse signal, be that the tidal wave starting point overlaps with main crest or the tidal wave starting point overlaps with heavily rich ripple starting point, therefore the arteries and veins figure that collects presents double-humped wave sometimes, sometimes present three spikes, make the labelling of flex point become difficult.In order to address this problem, the present invention at first adopts the chain code method to determine arteries and veins figure crest number, then in conjunction with maximum method labelling tidal wave crest and heavy Bo Bobofeng, and by coming labelling tidal wave starting point and heavily rich ripple starting point in conjunction with calculating arteries and veins figure master crest top angle.
Arteries and veins graph parameter h
1, h
2, h
3, h
4, h
5Can directly calculate according to corner position.The present invention adopts wavelet transformation modulus maximum method to determine arteries and veins figure period T exactly, and judges whether T equated on each time period; For arteries and veins area of pictural surface A, the present invention adopts the region area computational methods of two dimensional image to calculate, and has realized no inclined to one side with the consistent preferably estimation to the arteries and veins area of pictural surface.
Description of drawings
Fig. 1 is this detection method application system structured flowchart;
Fig. 2 is a sphygmogram characteristic point example;
Fig. 3 is the main flow chart of this detection method;
Fig. 4 is the main program flow chart of this detection method;
Fig. 5 is that arteries and veins figure filters the subroutine flow chart of making an uproar;
Fig. 6 is an arteries and veins figure cycle detection subroutine flow chart;
Fig. 7 is that arteries and veins figure flex point detects subroutine flow chart;
Fig. 8 is that arteries and veins figure crest number detects subroutine flow chart;
Fig. 9 is an arteries and veins figure flex point labelling subroutine flow chart;
Figure 10 is an arteries and veins figure example;
Figure 11 is that Figure 10 filters the arteries and veins figure that obtains after making an uproar;
Figure 12 is the flex point testing result to Figure 11;
Figure 13 is the characteristic parameter testing result to the arteries and veins illustrated example.
The specific embodiment
The present invention is described further in conjunction with the embodiments.
The present invention carries out pulse-tracing collection by pulse diagnosis sensor and pulse-tracing collection circuit, and pulse condition is input in the computer processor by USB interface, by the pulse signal under the programme-control collection different pressures, the arteries and veins figure that chooses arteries and veins figure amplitude maximum carries out the pulse condition characteristic parameter and detects automatically.Referring to Fig. 1, Fig. 3.1 is pulse diagnosis sensor among Fig. 1, and 2 is the pulse-tracing collection circuit, and 3 is USB interface, and 4 is computer processor, and 5 is the output buffers district, and 6 is pulse analysis, and 7 is the parameter detecting result, and 8 is display, and 9 is printer.
Arteries and veins figure is carried out wavelet transformation, and the difference of transmission characteristic and feature is separated it when utilizing arteries and veins figure signal to carry out wavelet decomposition with noise under different scale.
Based on the arteries and veins figure signal denoising method step of wavelet transformation referring to Fig. 5:
(1) select certain wavelet function Ψ (x), and definite wavelet decomposition yardstick J, arteries and veins figure signal f (x) is carried out J layer wavelet decomposition, obtain corresponding coefficient of wavelet decomposition CA
j, CD
j, j=1...J;
(2) choose suitable yardstick a=S, yardstick S is decomposed the small echo detail coefficients CD that obtains down
S Put 0, obtain new coefficient of wavelet decomposition CA
S
(3) carry out wavelet inverse transformation, the wavelet coefficient under the yardstick S is reconstructed, promptly obtain filtering the arteries and veins figure F (x) after making an uproar;
(4) change wavelet basis, suitable wavelet basis is determined in repeating step (1)～(3), then on this basis, changes wavelet decomposition number of plies J, chooses suitable J.
1) for convenience detected flex point is carried out labelling, need at first detect the arteries and veins figure cycle, and arteries and veins figure is intercepted one-period handle.The cycle detection step is seen Fig. 6.
2) detect arteries and veins figure flex point based on multiple wavelet transformation.At first choose suitable scaling function and construct real wavelet function, and real wavelet function is carried out the Hilbert conversion convert it into multiple small echo.Arteries and veins figure signal after filter made an uproar carries out multiple wavelet transformation, and the zero crossing after the conversion is the modulus maximum point of the multiple wavelet transformation of arteries and veins figure.The curvature that WAVELET TRANSFORM MODULUS maximum point correspondence signal changes violent point, the i.e. flex point of signal.Flex point detects step and sees Fig. 7.
3) carry out the flex point labelling.After detecting arteries and veins figure flex point, need rise point, main crest, tidal wave starting point, tidal wave peak, heavily rich ripple starting point and heavy Bo Bobofeng to main ripple and carry out labelling.Wherein detected first flex point is labeled as main ripple and rises a little, and the labelling of other four characteristic points needs at first to determine the crest number of arteries and veins figure.In the present invention, arteries and veins figure crest number determines by the realization of chain code method.Chain code is to utilize a series of straightways that link to each other with length-specific and specific direction to represent the border of target.In order to realize convenience, the chain code value is got i={-3, and-2 ,-1,0,1,2,3}.Because do not have closed curve in the arteries and veins figure signal, then the direction shown in 3 ,-3 does not exist, so arteries and veins figure signal chain code L
dValue is i={-2 ,-1,0,1, and 2}.
Determine that according to the arteries and veins figure crest number of chain code method method is referring to Fig. 8:
(1) arteries and veins figure signal list is shown as i={-2 ,-1,0,1, the chain code string L shown in the 2} direction
d
(2) isolate from main ripple and rise o'clock to the subchain sign indicating number string L the 6th flex point
d';
(3) record L
d' in chain code transfer the position of negative to by positive number or zero, write down this position chain code simultaneously for just or zero number h
d, add up chain code is transferred to negative by positive number or zero number of times C at last
d
Threshold value T is set
Bd, if h
dT
BdAnd C
d=2, then arteries and veins figure is a double-humped wave; If h
dT
BdAnd C
d=3, then arteries and veins figure is three spikes
Consider that tidal wave crest and dicrotic wave crest in most cases are the maximum point of former arteries and veins figure, so the present invention adopts following method that it is carried out labelling, referring to Fig. 9:
(1) if arteries and veins figure is three spikes, then preceding 4 flex points are labeled as main ripple respectively and rise some P ', main crest P, tidal wave starting point E and dicrotic wave starting point F;
(2) maximum point of detection arteries and veins figure.If the current point of arteries and veins figure is x (n), its adjacent two points are x (n-1) and x (n+1), if x (n) satisfies x (n) 〉=x (n+1) and x (n) 〉=x (n-1), then x (n) is a maximum point.
(3) detect between arteries and veins figure CE whether have maximum point, if exist, then this maximum point is labeled as the tidal wave crest; Detect E and next main ripple and rise between the some A whether have arteries and veins figure maximum point, if exist, then this maximum point is labeled as the dicrotic wave crest.
The labelling of tidal wave starting point and heavily rich ripple starting point is based on priori: the summit angle is little and thin if main wave height is steep, then think the tidal wave starting point with heavily win the ripple starting point and overlap; If main ripple is wide type of circle or flat-and-wide type, think that then tidal wave and main ripple merge, therefore, the present invention determines main crest state by the method for asking for top angle angle.Detailed process is:
Arteries and veins figure crest number according to the chain code method determines that method is:
(1) calculates arteries and veins figure master crest top angle α;
(2) setting threshold T
α, if α〉and T
αThink that then the tidal wave starting point overlaps with main wave crest point, be that detected preceding 3 flex points are respectively main ripple liter point P ', main crest P (E) and dicrotic wave starting point F, otherwise, think that then the tidal wave starting point overlaps with heavily rich ripple starting point, promptly detected preceding 3 flex points are respectively main ripple and rise some P ', main crest P and dicrotic wave starting point F (E).
4) carry out calculation of parameter.After having determined that main ripple rises the position of point, main crest, tidal wave starting point, tidal wave peak, heavily rich ripple starting point and heavy Bo Bobofeng, h
1, h
2, h
3, h
4, h
5, t1, T can directly calculate.For arteries and veins area of pictural surface A, adopt the region area computational methods of two dimensional image to calculate, concrete grammar is as follows:
(1) the main ripple of two adjacent pulse waves of connection rises a little, forms a 2 dimensional region PS
1S
2
(2) regional PS
1S
2In the pixel number be arteries and veins area of pictural surface A.
Referring to Figure 10～13, the inventive method can detect the sphygmogram characteristic parameter automatically by computer, and is convenient and swift, and the result is accurate, helps the objectifying of pulse-taking, standardization.
Referring to Fig. 1, in pulse condition characteristic parameter automatic checkout system block diagram, pulse diagnosis sensor by adjustable in pressure places patient's wrist to scratch tremulous pulse place acquisition pulse voltage signal, be converted to digital signal via pulse-tracing collection circuit (built-in signal amplifying circuit, A/D analog to digital conversion circuit), by USB interface input computer, be convenient to operations such as Computer Processing, transmission; Computer Processing mainly is to obtain arteries and veins figure signal under the different pressures by USB interface, and arteries and veins figure and parameter detecting result after the processing output to buffer, is convenient to show and print.Display is an outut device, and human eye can be watched arteries and veins figure and pulse analysis result by display.Pulse analysis is that the arteries and veins figure that computer reads in is carried out qualitative and quantitative analysis, and the output analysis result.
Original pulse condition figure can be the signal that collects in real time by pulse image sensor and pulse-tracing collection circuit, also can be to realize collecting the signal that is kept in the hard disc of computer by pulse image sensor and pulse-tracing collection circuit.
The key issue that the present invention relates to comprise arteries and veins figure filter make an uproar, based on the arteries and veins figure flex point of multiple wavelet transformation detect, flex point labelling and calculation of parameter.
Adopt the sym8 small echo among the present invention, get decomposition scale a=4 and arteries and veins figure is filtered make an uproar.
In arteries and veins figure flex point detects, be scaling function, get its first derivative as wavelet basis function ψ (x), and it is become multiple small echo ψ (x)=(1+iH) ψ (x) by the Hilbert shift conversion, be used to detect flex point with the Gaussian function.
For convenience's sake, when arteries and veins figure cycle detection, still select for use the sym8 small echo to handle.
Pulse analysis is mainly realized by software.In computer, finish following program (mastery routine is seen Fig. 4):
1. read in the arteries and veins diagram data, enter the arteries and veins figure filter subprogram of making an uproar, adopt the sym8 small echo that arteries and veins figure is done 4 grades of wavelet decomposition, and at yardstick 4 with high frequency coefficient CD
4Put 0, with low frequency coefficient CA
4Be reconstructed the arteries and veins figure storage after making an uproar as filter.
2. enter arteries and veins figure cycle detection subprogram, referring to Fig. 6, length M=16 after the wavelet function sym8 discretization, decomposition scale L=1, with the start position in detected arteries and veins figure cycle be stored in u (1, i), i=1..., among the n, wherein n is the number in arteries and veins figure cycle.
3. enter arteries and veins figure flex point and detect subprogram, referring to Fig. 7, detected flex point is stored among the G (r), wherein r is a detected flex point number in the arteries and veins figure cycle.
4. enter arteries and veins figure crest number and detect subprogram,, get d=7 referring to Fig. 8, M=T/d, wherein T is the arteries and veins figure length of one-period, gets threshold value Thd=6, detected crest number is stored among the variable g.
5. enter the subprogram of arteries and veins figure flex point labelling, referring to Fig. 9, mark main ripple and rise point, main crest, tidal wave starting point, tidal wave peak, heavily rich ripple starting point and heavy Bo Bobofeng, the position of each flex point is stored in one-dimension array b (i), i=1 ..., in 6.
6. directly according to b (i), i=1 ..., the corner position calculating parameter h of storage in 6
1, h
2, h
3, h
4, h
5, t1, period T calculates in the arteries and veins figure of step 2 cycle detection subprogram, for the arteries and veins area of pictural surface, directly calculates two adjacent main ripples and rises the pixel number that the zone comprised that point and arteries and veins figure surround and get final product.
Pulse analysis result's demonstration.In order to make analysis result be convenient to understand and clinical practice, adopt numeral, literal and bitmap display analysis result.Figure 10～Figure 13 is the example that a pulse condition characteristic parameter detects, and Figure 10 is the arteries and veins diagram data, and Figure 11 filters the arteries and veins figure of the back storage of making an uproar for step 1, and Figure 12 is that the detected corner position of step 3 shows the arteries and veins graph parameter that Figure 13 calculates for step 6.
Need not further to elaborate, believe and adopt the disclosed content in front, those skilled in the art can use the present invention to greatest extent.Therefore, the embodiment of front is interpreted as only illustrating, but not limits the scope of the invention by any way.
Claims (7)
1. Chinese medicine pulse characteristic parameter automatic testing method, it is characterized in that: the pulse condition collecting device of being made up of pulse diagnosis sensor and pulse-tracing collection circuit obtains pulse signal, and transfer to computer, remove make an uproar, flex point detects and the arteries and veins graph parameter detects, specifically be achieved through the following technical solutions:
(1) gathers arteries and veins figure information
Pulse diagnosis sensor by adjustable in pressure places patient's wrist to scratch tremulous pulse place acquisition pulse voltage signal, is digital signal via the pulse-tracing collection circuit conversion, by USB interface input computer;
(2) arteries and veins figure signal removes and makes an uproar
Pulse signal is input in the computer processor, gathers different pressures pulse signal down, arteries and veins figure signal is removed make an uproar by programme-control, be used for arteries and veins figure except that the vanishing moment of the wavelet basis function of making an uproar more than or equal to 2, and the wavelet decomposition yardstick is got a=4;
(3) the pulse condition characteristic parameter detects
By detecting arteries and veins figure flex point, carry out the labelling of flex point, according to corner position, the arteries and veins figure that chooses arteries and veins figure amplitude maximum carries out the detection of pulse condition characteristic parameter.
2. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method is characterized in that: adopt multiple wavelet transformation modulus maximum method to detect arteries and veins figure flex point, wherein wavelet basis function adopts the multiple small echo that the first derivative of Gaussian function becomes through the Hilbert shift conversion.
3. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method, feature is: adopt the chain code method to determine arteries and veins figure crest number during the flex point labelling, then in conjunction with maximum method labelling tidal wave crest and heavy Bo Bobofeng, and by coming labelling tidal wave starting point and heavily rich ripple starting point in conjunction with calculating arteries and veins figure master crest top angle.
4. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method, it is characterized in that: the arteries and veins graph parameter of detection is by the high h of main wave amplitude
1, main ripple rise time t
1, tidal wave starting point height h
2, tidal wave panel height h
3, heavily rich ripple starting point height h
4, the heavily rich high h of wave amplitude
5, arteries and veins figure period T and arteries and veins area of pictural surface A form.
5. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method, it is characterized in that: the arteries and veins graph parameter can directly calculate according to corner position, adopt wavelet transformation modulus maximum method to determine arteries and veins figure period T exactly, and judge whether T equates on each time period, for arteries and veins area of pictural surface A, adopt the region area computational methods of two dimensional image to calculate.
6. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method, it is characterized in that: original pulse condition figure can be the signal that collects in real time by pulse image sensor and pulse-tracing collection circuit in the step (1), or by being saved in the signal in the hard disc of computer after pulse image sensor and the collection of pulse-tracing collection circuit.
7. the described Chinese medicine pulse characteristic parameter of claim 1 automatic testing method is characterized in that: wavelet basis function is selected sym8 small echo or bior3.5 small echo for use in the step (2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100613942A CN100493445C (en) | 2005-11-03 | 2005-11-03 | Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2005100613942A CN100493445C (en) | 2005-11-03 | 2005-11-03 | Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN1792319A CN1792319A (en) | 2006-06-28 |
| CN100493445C true CN100493445C (en) | 2009-06-03 |
Family
ID=36804101
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2005100613942A Expired - Fee Related CN100493445C (en) | 2005-11-03 | 2005-11-03 | Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100493445C (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101371779B (en) * | 2007-08-24 | 2011-08-03 | 柳竹 | Method for extracting traditional Chinese medicine pulse manifestation physiology information |
| CN101408912B (en) * | 2008-11-21 | 2010-06-09 | 天津师范大学 | Method for automatically extracting characteristic function of traditional Chinese medicine pulse manifestation |
| CN103027667B (en) * | 2011-09-30 | 2017-01-18 | Ge医疗系统环球技术有限公司 | Characteristic parameter extraction of pulse wave |
| CN103027669B (en) * | 2011-09-30 | 2016-09-14 | Ge医疗系统环球技术有限公司 | For the method and apparatus judging floating and sinking degree of pulse condition |
| CN102880812B (en) * | 2012-08-02 | 2016-08-03 | 江苏华生基因数据科技股份有限公司 | The method of solubility curve, device and electric terminal is processed based on wavelet analysis method |
| TWI551266B (en) * | 2013-12-30 | 2016-10-01 | 財團法人工業技術研究院 | Analyzing arterial pulse method and system thereof |
| CN103932686A (en) * | 2014-04-22 | 2014-07-23 | 北京印刷学院 | Method and device for extracting pulse condition signal |
| JP6910290B2 (en) | 2015-04-17 | 2021-07-28 | 太陽誘電株式会社 | Vibration waveform sensor and waveform analyzer |
| CN105067650B (en) * | 2015-08-03 | 2018-01-02 | 东南大学 | A kind of method that derivation detection spectrum signature peak is calculated using small echo |
| CN106419864A (en) * | 2015-08-05 | 2017-02-22 | 南开大学 | Cloud expert system for TCM diagnosis |
| CN107898443B (en) * | 2017-11-21 | 2020-11-24 | 深圳先进技术研究院 | Method and device for detecting counterpulsation wave and computer storage medium |
-
2005
- 2005-11-03 CN CNB2005100613942A patent/CN100493445C/en not_active Expired - Fee Related
Also Published As
| Publication number | Publication date |
|---|---|
| CN1792319A (en) | 2006-06-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100493445C (en) | Automatic testing method for traditional Chinese medical pulse manifestation characteristics parameter | |
| CN100418480C (en) | Heart disease automatic classification system based on heart sound analysis and heart sound segmentation method | |
| CN102512158B (en) | Electrocardiosignal quality evaluation method and device based on high-dimensional fuzzy recognition | |
| Lu et al. | Classification of single-channel EEG signals for epileptic seizures detection based on hybrid features | |
| CN107529646A (en) | A kind of non-contact type method for measuring heart rate and device based on Euler's image zoom | |
| US10368755B2 (en) | Apparatus and method for feature extraction and classification of fetal heart rate | |
| US20120184862A1 (en) | System and method for patient monitoring | |
| CN101919704B (en) | Heart sound signal positioning and segmenting method | |
| CN105286853A (en) | Disease detection method and device based on wearable device and wearable device | |
| Zou et al. | An ultra-low power QRS complex detection algorithm based on down-sampling wavelet transform | |
| CN107622259B (en) | T wave detection method, electrocardiogram data analysis method and device | |
| CN112869716B (en) | Pulse feature identification system and method based on two-channel convolutional neural network | |
| Gupta et al. | Higher order derivative-based integrated model for cuff-less blood pressure estimation and stratification using PPG signals | |
| CN113066502A (en) | Heart sound segmentation positioning method based on VMD and multi-wavelet | |
| Polat et al. | A novel data reduction method: Distance based data reduction and its application to classification of epileptiform EEG signals | |
| Schmidt et al. | No evidence of nonlinear or chaotic behavior of cardiovascular murmurs | |
| CN113729653A (en) | Human body pulse wave signal acquisition method | |
| Mitra et al. | Pattern defined heuristic rules and directional histogram based online ECG parameter extraction | |
| CN106419884B (en) | A kind of rate calculation method and system based on wavelet analysis | |
| CN117357080A (en) | Near infrared spectrum signal denoising method and device, terminal equipment and storage medium | |
| CN109620198B (en) | Cardiovascular index detection and model training method and device | |
| CN117017297A (en) | Method for establishing prediction and identification model of driver fatigue and application thereof | |
| CN116421144A (en) | ESES (ESES) -related epileptic encephalopathy prognosis evaluation model, training and using method thereof | |
| CN103099642A (en) | Ultrasonic blood flow signal quality real-time analysis method | |
| Nibhanupudi | Signal denoising using wavelets |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090603 Termination date: 20091203 |