TWI618054B - Computer program product and computer readable medium for determination of narrowing condition of fistula by detection of audio signals - Google Patents

Abstract

A program product capable of obtaining a narrowing degree of a tube by means of an audio technology, and a computer readable medium storing the program product, the program product being suitable for a computer device to receive in advance when the program product is loaded The spectrum signals from the normal fistula and the narrowed fistula are spectrally converted to obtain their respective characteristic spectra as normal samples and narrowed samples. The normal sample and the narrowed sample are then calculated by chaos synchronization theory, and a dynamic error is output. And using the attribution function to perform the fuzzy-color inference calculation of the dynamic error, and convert the result into hue (Hue, H), so that the color represented by the hue is presented on the display interface, which can represent the narrowing degree of the narrowed fistula, and The reliability is expressed by Saturation (S), which is convenient for doctors and patients to directly understand the results.

Description

A program product that can obtain a narrowing degree of the tube by audio technology and store the described Computer readable media for program products

The present invention relates to a program product capable of obtaining a narrowing degree of a tube by means of an audio technology, and a computer readable medium storing the program product, in particular, a blood flowing through a chaotic system in a normal fistula and a narrow fistula The difference of the audio feedback is numerically analyzed to judge the narrowing of the fistula, and the attribution function is used to perform the blur-color inference for the degree of blocking corresponding to the obtained dynamic error, and the inference result is hue and hue saturation. Presented to facilitate direct understanding of the results of screening.

Good dialysis vascular access can be said to be the lifeline of patients with kidney disease, which can be divided into autologous and artificial fistulas. Maintaining a good fistula pathway function can improve circulation efficiency during hemodialysis and avoid vascular lesions caused by long-term hemodialysis. Because the dialysis tube has been stimulated by the puncture needle for a long time, it is easy to cause vascular aneurysm, thrombus and intimal thickening and fibrosis inside the vascular passage, so that the arteriovenous fistula stenosis loses the function of normal blood flow, and the blood vessel narrows the blood. Flow is an early symptom of vascular access failure. When the vascular access fails, the physician needs to use a thrombectomy or Percutaneous Transluminal Angioplasty (PTA) to clear the blood clot or re-introducing a new fistula in the arteries and veins.

According to the National Kidney Foundation (NKF) Dialysis Therapy Quality Initiative Guidelines (DOQI), the current recommended methods for assessing the possible failure of vascular access are monitoring and monitoring the vascular access in both directions, once highly suspected Then arrange a angiographic examination.

Surveillance is to detect that the vascular access may be ineffective through a physical examination or an abnormality in the test data. A physical examination is a simple and quick way to evaluate arteriovenous fistulas and artificial blood vessels, including visual inspection to see if the arm is swollen, palpation of the tremor (Thrill), whether the pulse is weakened, and whether the auscultation tone (Bruit) is lost or high frequency occurs. If the arm is raised, the fistula is flat (Arm Elevation), and if the proximal end of the anastomosis is a few centimeters, it will make the distal pulse become stronger (Pulse Augmentation). Whether the fistula becomes difficult to stop bleeding, whether the dialysis index Kt/V is unreasonably reduced, etc. The advantages of the physical examination are easy to carry out and low in cost. The shortcoming is the lack of standard quantification as a reference. It is necessary for experienced people to monitor the positive predictive rate to increase. Therefore, it is predicted by human factors that the fistula stenosis is positive and there is a big difference in different research data (about 30-93%). .

There are three methods commonly used in monitoring: First, the static Venous Pressure is measured using a puncture needle. Second, the ultrasonic blood flow measurement is used to measure the blood flow of the fistula. When the blood flow of the arteriovenous fistula is less than 400-500ml/min or the blood flow of the artificial blood vessel is lower than 600ml/min, it is recommended to undergo angiography and perform percutaneous transection in severe stenosis. Transluminal angioplasty repair. Third, the use of Doppler ultrasound to measure the blood flow of the fistula and to measure the highest systolic blood flow rate (Peak Systolic Velocity, PSV) before and after the suspected stenosis, when the stenosis of the posterior stenosis of the front end of the PSV ratio is greater than 2 should be recommended For angiography. Among them, the Doppler ultrasonic measurement and angiography methods have a high prediction rate, and are also frequently used in recent years. However, the Doppler ultrasound cannot be performed in the dialysis unit in time. It is necessary to make an appointment at the time and place to cause the patient to be inconvenient, and the Puler ultrasonic measurement and angiography methods require expensive design. Prepare and hire a full-time technician or be operated by a radiologist. Monitoring is currently recommended for one to six months depending on the capabilities of each medical unit.

In order to prevent dialysis patients from urinating in the hospital, in addition to monitoring the relevant physiological parameters of the human body, and simultaneously monitoring the sound of the arteriovenous fistula with a simple operation instrument to detect lesions in the hemodialysis pathway at an early stage, the team of the present invention In the past, the "Recommendation Method for Vascular Access Function of Dialysis Tubes", No. I484939 of the Republic of China, and the New Device Patent Announcement No. M507738 of the Republic of China, "The Device for Judging the Narrowing of the Dialysis Tubes" were presented. How to determine the degree of narrowing of the fistula by comparing the characteristic spectrum of the sound signal, and the novel patent discloses that the method of narrowing the degree of audio judgment is applied to the measurement of body weight and blood pressure before dialysis in dialysis patients. It is easy to use for dialysis patients.

However, when the concept of the above two patents is actually commercialized, a preferred method is to integrate the device for narrowing the degree of audio judgment into an integrated system (ES) to measure the body weight and blood pressure. The carrier is used to reduce the size and cost of the product, which is beneficial to improve reliability and performance. In the foregoing invention patent case, it was proposed to apply the Chen-Lee chaotic system to measure the dynamic error of the spectrogram in the arterial anastomotic point, the venous anastomotic point or the middle section, so that Φ ₂ -Φ ₁ , Φ ₃ -Φ ₁ , Φ ₃ -Φ ₂ constitute the so-called three-dimensional "butterfly effect map" on three planes, and the dynamic error of the synthesis Corresponding to the vascular access obstruction degree DOS, used to assess the degree of narrowing of the fistula. However, the Chen-Lee system has three eigenvectors of Φ ₁ , Φ ₂ and Φ ₃ . If it is to be implemented in an embedded system, the degree of blocking will be increased by rapid screening, which will increase the calculation amount and calculation required for the screening algorithm. Time, increasing the complexity of the algorithm.

In addition, the foregoing invention patent case also proposes to screen the degree of obstruction of the fistula by direct look-up table method. However, if there are more audio frequency bands crossing, partial overlap, serious overlap, etc., it is often necessary To cross-align multiple feature bands, increase the difficulty for the physician to determine the level of blockage. The pre-invention patent case also proposes to use fuzzy inference to determine the narrowing level of the fistula, but it can not further know the degree of narrowing in each level, and does not disclose how to visually display the screening results, Let the doctor and patient quickly know the results of the screening and the obstruction.

Therefore, in order to improve the above-mentioned deficiencies, the team of the present invention has been working on a program product that can obtain a narrowing degree of the tube by the audio technology, and is suitable for a computer device, which first receives blood in a radio device. a normal fistula sound signal generated during the operation of the non-narrowed normal fistula, and a narrowed fistula sound signal generated when the blood is received in a narrowed fistula, and the computer device is connected to the radio device, and When the program product is loaded, the following methods are performed, including: A. spectrally converting the normal fistula sound signal to obtain a normal fistula characteristic spectrum as a normal sample; and performing the narrowing fistula sound signal on the spectrum Converting, obtaining a narrowed fistula characteristic spectrum as a narrowed sample; B. calculating both the normal sample and the narrowed sample by chaotic synchronization theory (CS), outputting a dynamic error; C. pre-established representative The three levels of obstruction of the degree of obstruction, and based on the attribution function of the fuzzy theory, plan different dynamic errors corresponding to the above three The respective degrees of occlusion of the blocking level; the computer device performs fuzzy inference calculation on the aforementioned dynamic error of the output, and obtains three attribution degrees μ _m1 , μ _m2 , μ _m3 corresponding to the three blocking levels, and the computer device The above three attribution degrees μ _m1 , μ _m2 , and μ _m3 are respectively converted into gray correlation degrees, and the gray correlation degrees are expressed as: ρ _mk =ξ exp[-ξ( μ _mk -1) ² ], k=1, 2 3, ξ is the identification coefficient, and then convert the three gray correlation degree into three primary colors, the three primary colors are expressed as: △ ρ _m = ρ _{m , max} - ρ _{m , min} , ρ _{m , min} =min[ ρ _{m 1} , ρ _{m 2} , ρ _{m 3} ], ρ _m,max =max[ ρ _{m 1} , ρ _{m 2} , ρ _{m 3} ], and then convert the three primary colors into a hue (Hue, H), the hue is expressed as: And presenting the color represented by the hue to a display interface to represent the degree of narrowing of the narrowed fistula.

Further, three sputum obstruction levels of DOS (clinical ratio calculated by clinical angiography) of less than 0.3, DOS of 0.3 to 0.7, and DOS of greater than 0.7 were prepared in advance.

Further, the attribution function is an exponential attribution function and is expressed as: Where Ψ _m is the dynamic error, and Mean _{m 1} , Mean _{m 2} and Mean _{m 3} are the average values of the dynamic error Ψ of the three manifold obstruction levels.

Further, the computer device calculates the brightness and the saturation by the maximum gray correlation degree ρ _{m ,max} and the minimum gray correlation degree ρ _{m ,min} , wherein the brightness is expressed as: v _m = ρ _{m ,max} , the saturation degree Expressed as: ( ρ _{m ,min} ≠ ρ _{m ,max'} ρ _{m ,max} ≠0), wherein the higher the value S _m and close to 1, the higher the reliability of the narrowing of the narrowed manifold obtained.

Further, the chaotic synchronization theory (CS) uses the Sprott chaotic system.

Further, the calculation of the Sprott chaotic system includes the following steps: the computer device inputs the normal sample and the narrowed sample into a main system and a servant system respectively; and the control unit calculates the error between the two by the computer device A signal for generating a dynamic error is generated; the control unit processes the signal of the dynamic error as a control signal, and inputs the signal to the servant system for controlling synchronization between the servant system and the main system.

Further, the mathematical model of the Sprott chaotic system is: the main system is expressed as: The servant system is expressed as: , where X = [ x ₁ , x ₂ , x ₃ ], Y = [ y ₁ , y ₂ , y ₃ ], X R ³ , Y R ³ , matrix elements a and b are system parameters, sign( ̇ ) is defined as a signal function, u is a controller to control the servant system to achieve synchronization with the main system; when calculating, unidirectional coupling is used to control The input signal is zero, and the error variable e = [( x ₁ - y ₁ ), ( x ₂ - y ₂ ), ( x ₃ - y ₃ )] = [ e ₁ , e ₂ , e ₃ ] ^T , The error system of the dynamic error is expressed as: The error system of the dynamic error is decoupled and expressed as:

Further, according to the Grunwald-Letnikov law, the error system after decoupling is modified into a fractional parameter, so that the error system is expressed as: And discretizing the error system to obtain: The dynamic error of the composite fractional order is: , i [1, n -2].

Further, discretizing the aforementioned error system, setting the error variable e ₁ [ i ]= x [ i ]- y [ i ], e ₂ [ i ]= x [ i +1]- y [ i +1], e ₃ [ i ]= x [ i +2]- y [ i +2], i =1,2,3,..., n -2, the equation of the dynamic error is expressed as Ψ=-1.2 e ₁ [ i ]- be ₂ [ i ]- ae ₃ [ i ]+2( sign ( x [ i ])- sign ( y [ i ]), x[i] is taken from the characteristic spectrum of the normal sample, y[i] The system is taken from the characteristic spectrum of the narrowed sample.

The present invention is also a computer readable medium storing a program product which can obtain a narrowing degree of the tube by means of audio technology, and stores the above-mentioned program product which can obtain the degree of narrowing of the tube by the audio technology.

According to the above technical features, the following effects can be achieved:

1. The dynamic error of the narrowed sample is converted into a hue, and the color is displayed on a display screen of various electronic devices, such as a tablet computer, for the doctor, the patient to quickly and intuitively obtain the result of the screening.

2. The color represented by the hue can be further obtained, and the degree of narrowing of the fistula can be further obtained. For example, in the embodiment, the red line represents DOS>0.7, the indication is that it is recommended to undergo PTA surgery repair; the green line represents DOS<0.3, the indication is Normal fistula; the blue line represents 0.3<DOS<0.7, indicating different degrees of fistula obstruction, that is, the color biased to light blue or blue-green represents 0.3<DOS<0.5, which is a low degree of obstruction, while the color is biased toward pink. Represents 0.5 < DOS < 0.7, blocking for a high degree.

3. The brightness and the saturation are obtained from the maximum gray correlation degree and the minimum gray correlation degree, and the higher the value of the saturation degree is close to 1, indicating that the reliability of the detection result is higher.

4. The invention utilizes the Sprott chaotic system to obtain the dynamic error of the narrowed sample and the normal sample, and reduces the calculation amount, the calculation time and the algorithm complexity, and is easy to implement in the embedded system, and compared with the known prior case, and The actual combination with the vehicle for measuring body weight and blood pressure can reduce the volume and cost of the product and improve reliability and performance.

5. The degree of blocking is obtained by dynamic inference through fuzzy inference. Compared with the previous case, multiple feature bands are crossed and compared, which reduces the chance of misjudgment.

(11) ‧‧‧ normal sample

(12) ‧‧‧narrowed samples

(2) ‧‧‧ Dynamic Errorψ

(31)‧‧‧Main system

(32) ‧ ‧ servant system

(4) ‧ ‧ database

(51)‧‧‧Affiliation

(52) ‧‧‧ Gray correlation

(53) ‧ ‧ three primary colors

(54) ‧ ‧ Hue

(55)‧‧‧Saturation

[First figure] is a schematic flow chart of an embodiment of the present invention.

[Second image] is a waveform diagram of a narrowed fistula sound signal of blood flowing in a narrowed fistula according to an embodiment of the present invention.

[Third image] is a characteristic spectrum diagram of a normal sample and a narrowed sample according to an embodiment of the present invention.

[Fourth Diagram] is a schematic diagram of the dynamic error obtained in the embodiment of the present invention.

[Fifth Graph] is a schematic flow chart of the blur-color inference of the embodiment of the present invention.

[Sixth Graph] is a graph of the gray correlation degree in the embodiment of the present invention.

[Seventh figure] In the embodiment of the present invention, the color of the hue is different, and the schematic diagram of the degree of narrowing of the obstruction of the fistula is illustrated, and the color presented is displayed on the display interface.

[Eighth image] is a schematic diagram of the HSV color space in the embodiment of the present invention.

In combination with the above technical features, the main functions of the present invention, which can obtain the degree of narrowing of the tube by the audio technology, and the computer readable medium storing the program product, will be clearly shown in the following embodiments.

The program product of the embodiment which can obtain the narrowing degree of the tube by the audio technology is applicable to a computer device.

Referring to the first, second, third, fourth and fifth figures, this embodiment firstly uses a sound receiving device to receive blood when flowing in an unnarrowed normal fistula. a normal fistula sound signal, and a narrowed fistula sound signal generated when the blood is received in a narrowed fistula tube, wherein the narrowed fistula sound signal, as shown in the second figure, may be a two-channel electronic stethoscope Takes a tremor sound signal from the vascular access. The computer device is connected to the radio device, and when the program product is loaded, the following methods are performed, including:

A. spectrally converting the normal fistula sound signal to obtain a normal fistula characteristic spectrum as a normal sample (11); spectrally converting the narrowed fistula sound signal to obtain a narrowed fistula characteristic spectrum as a narrowing Sample (12). In this embodiment, the normal fistula sound is made. The signal and the narrowed fistula sound signal are subjected to signal segmentation processing and fast Fourier transform to obtain the characteristic spectrum of the normal sample (11) and the narrowed sample (12) as shown in the third figure.

B. Calculating the normal sample (11) and the narrowed sample (12) by chaotic synchronization theory (CS), and outputting a dynamic error (2) as shown in the fourth figure, wherein the embodiment The middle system uses the Sprott chaotic system for calculation.

C. Pre-established three levels of obstruction levels representing the degree of obstruction of the fistula, and according to the attribution function of the fuzzy theory, plan different attributions of the dynamic error corresponding to the respective three obstruction levels; the computer device will output the aforementioned dynamic error (2) performing the calculation of the fuzzy inference, and obtaining the three attributions (51) corresponding to the three blocking levels, the computer device and converting the three attributions (51) into the gray correlation degree (52) respectively. And converting the three gray correlation degree (52) into three primary colors (53), and then converting the three primary colors (53) into a hue (54) (Hue, H), and the color represented by the hue (54), Presented in a display interface to represent the degree of narrowing of the narrowed fistula.

Referring to the first figure, in step B, when the computer device calculates the normal sample (11) and the narrowed sample (12) by using the Sprott chaotic system, the following process is performed: the computer The device inputs the normal sample (11) and the narrowed sample (12) into a main system (31) and a servant system (32); and the control unit calculates a difference between the computer device to generate a dynamic error. The control unit processes the signal of the dynamic error into a control signal, and then inputs the signal to the servant system (32) for controlling synchronization between the servant system (32) and the main system (31).

Wherein, the mathematical model of the Sprott chaotic system is: the main system (31) is expressed as: The servant system (32) is expressed as: Where X = [ x ₁ , x ₂ , x ₃ ], Y = [ y ₁ , y ₂ , y ₃ ], X R ³ , Y R ³ , matrix elements a and b are system parameters, sign( ̇) is defined as a signal function, and u is a controller for controlling synchronization of the servant system (32) with the main system (31). When calculating, use one-way coupling to make the control input signal zero, set the error variable e =[( x ₁ - y ₁ ), ( x ₂ - y ₂ ), ( x ₃ - y ₃ )]=[ e ₁ , e ₂ , e ₃ ] ^T , the error system of the dynamic error is expressed as: The error system of the dynamic error is decoupled and expressed as:

Wherein, if the calculation of the Sprott chaotic system is to be implemented in a discrete self-synchronizing error detector, the error variable e ₁ [ i ]= x [[ i ]- y [ i ], e ₂ [ i ]= x is set. [ i +1]- y [ i +1], e ₃ [ i ]= x [ i +2]- y [ i +2], i =1,2,3,..., n -2, then The equation of dynamic error can be discretized and expressed as Ψ=-1.2 e ₁ [ i ]- be ₂ [ i ]- ae ₃ [ i ]+2( sign ( x [ i ])- sign ( y [ i ]), where x[i] represents discrete sequence data, taken from the characteristic spectrum of the normal sample (11), and the normal sample (11) can be pre-stored in a database (4), and y[i] ] indicates discrete sequence data, taken from the characteristic spectrum of the narrowed sample (12), that is, the narrowed fistula sound signal flowing in the narrowed fistula tube, the characteristic spectrum obtained by the conversion, and n is the sampling The number of points. After the above discrete processing, the present invention facilitates the computational execution of the computer.

In the present embodiment, according to the Grunwald-Letnikov law, the error system after decoupling is modified into a fractional parameter, so that the error system is expressed as: , where the system parameters And discretizing the error system to implement the foregoing discrete fractional self-synchronization error detector, obtaining: By, the dynamic error of the composite fractional order is: , i [1, n -2]. After this simplification, only one feature vector Φ ₂ , compared with the previous case, reduces the calculation amount, calculation time and algorithm complexity, and is easier to implement in the embedded system. And has the advantages that: a single equation contains e ₂ and e ₃ errors; system parameters are easy to set, parameters a and b meet the constraints: a greater than 0 and b greater than 0, to ensure the stability of the dynamic error system; Φ ₁ and Φ ₂ For the fractional scale planning value, the three sets of measurement positions of the arterial anastomosis and the fistula, the middle of the fistula and the venous anastomosis, the anastomosis and the venous anastomosis can be tried to find the fractional dynamic error and vascular occlusion. Ratio, showing a better positive correlation, and can be used as an indicator to assess the level of blockage.

Then, referring to the first figure and the fifth figure, wherein the embodiment is in the foregoing step C, and the obstruction ratio calculated by the clinical angiography (DOS%) is used, and the obstruction degree of the fistula is set to three levels. The blocking level is DOS less than 0.3, DOS is between 0.3 and 0.7, and DOS is greater than 0.7, and the exponential type attribution function is used to plan the attribution of different dynamic errors corresponding to each blocking level, so that the attribution The function is expressed as: DOS<0.30: 0.30<DOS<0.70: DOS>0.70: Where Ψ _m is the dynamic error; Mean _{m 1} , Mean _{m 2} and Mean _{m 3} are the average values of the three dynamic obturator levels corresponding to the dynamic error ,, which can be analyzed by clinical data analysis (Big Data Analysis, BDA), for the population of autologous fistula and artificial fistula, the average value of the screening obstruction level, and the distribution range of the corresponding dynamic error 阻塞 of the obstruction level, and the clinical big data analysis to develop the screening for clinical screening Function planning; μ _m1 , μ _m2 , μ _m3 are the attribution of the attribution function, m is 1 or 2, which means that the two-channel electronic stethoscope is taken at two positions such as an anastomosis of the artery and the middle of the fistula, or Two positions in the middle of the fistula and the venous anastomosis, or the fistula sound signal at the two locations of the anastomosis and venous anastomosis. Moreover, when the fuzzy inference is performed through the foregoing attribution function, it can be inferred that the degree of narrowing of the fistula is the third blocking level.

Referring to the first, fifth, sixth and eighth figures, in the foregoing step C, a fuzzy-color inference is performed, and in detail, the computer device outputs the aforementioned dynamic error (2). The fuzzy inference calculation is performed to obtain three attribution degrees (51) corresponding to the above three blocking levels, namely μ _m1 , μ _m2 , μ _m3 , and then the above three attribution degrees (51) (μ _m1 , μ _M2 , μ _m3 ) are respectively converted into gray correlation degree (52), and then the three gray correlation degree (52) is converted into three primary colors (red, basket, green basic color) (53), and then the three primary colors (53) are mixed. To coordinate, the three primary colors (53) are converted into a color space, and the three primary colors (53) are converted into a hue (Hue, H) [(such as the hue (H) shown in the eighth figure], wherein the gray correlation degree represents For: ρ _mk =ξ exp[-ξ( μ _mk -1) ² ], k=1, 2, 3, ξ is the identification coefficient, which can be formulated by clinical big data analysis, by which, Referring to Figure 6, the recognition factor 可 can be adjusted to increase the recognition contrast.

As shown in the fifth figure, this embodiment performs ρ _{m , min} =min[ ρ _{m 1} , ρ _{m 2} , ρ _{m 3} ], which is the minimum gray correlation degree, and performs ρ _m,max =max[ ρ _{m 1} , ρ _{m 2} , ρ _{m 3} ], which is the maximum gray correlation.

As shown in the fifth and seventh figures, the three primary colors (53) are expressed as: Δ ρ _m = ρ _{m , max} - ρ _{m , min} ; the hue (54) is expressed as: . By contrast, the difference in color exhibited by the hue (54) can account for the degree of narrowing of the obstruction of the fistula. Therefore, as shown in the seventh figure, in the present embodiment, the red line can represent DOS>0.7, which is the degree of obstruction that the general physician recommends to undergo PTA surgery repair; and the green line represents DOS<0.3, indicating It is a normal fistula; and the blue line represents 0.3<DOS<0.7, indicating different degrees of fistula obstruction, wherein if the color is biased toward light blue or blue-green, the system represents 0.3<DOS<0.5, which is a low degree of obstruction. The color biased towards pink represents 0.5 < DOS < 0.7, which is blocked to a high degree. Displaying the color presented above on a display interface, and the display interface may be a display screen of the computer device, or may be a display screen of an electronic device (eg, a smart mobile device, a tablet computer, etc.). It is very intuitive to provide the physician and the patient with a narrower degree of narrowing of the fistula in a color rendering manner.

In addition, as shown in the fifth figure, the computer device calculates and obtains the brightness (Value, V) and the saturation (55) from the maximum gray correlation degree ρ _{m, max} and the minimum gray correlation degree ρ _{m , min} (Saturation, S ), wherein the brightness is expressed as: v _m = ρ _{m ,max} , and the degree of saturation (55) is expressed as: ( ρ _{m ,min} ≠ ρ _{m ,max'} ρ _{m ,max} ≠0), wherein the higher the value S _m and close to 1, the higher the reliability of the narrowing of the narrowed manifold obtained.

In view of the foregoing description of the embodiments, the operation and the use of the present invention and the effects of the present invention are fully understood, but the above described embodiments are merely preferred embodiments of the present invention, and the invention may not be limited thereto. Included within the scope of the present invention are the scope of the present invention.

Claims (7)

A program product that can obtain the degree of narrowing of the fistula by means of audio technology, and is suitable for a computer device, which is a normal fistula sound signal generated by receiving a blood in a non-narrowed normal fistula by a radio device, and receiving A narrowed fistula sound signal generated when blood flows in a narrowed fistula, and the computer device is connected to the radio device, and when the program product is loaded, the following methods are performed, including: The normal fistula sound signal is spectrally converted to obtain a normal fistula characteristic spectrum as a normal sample; the narrowed fistula sound signal is spectrally converted to obtain a narrowed fistula characteristic spectrum as a narrowed sample; B. Both the normal sample and the narrowed sample are calculated by Chaos Synchronization Theory (CS), and a dynamic error is output; C. Three levels of blocking levels representing the degree of obstruction of the fistula are pre-established, and are pre-planned according to the attribution function of the fuzzy theory. Different dynamic errors correspond to the respective degrees of attribution of the three blocking levels; the computer device performs the aforementioned dynamic error of the output The calculation of the fuzzy inference is obtained, and the three degrees of attribution μ _m1 , μ _m2 , μ _m3 corresponding to the above three blocking levels are obtained, and the computer device converts the above three attribution degrees μ _m1 , μ _m2 , μ _m3 into Gray correlation degree, the gray correlation degree is expressed as: ρ _mk =ξ exp[-ξ( μ _mk -1) ² ], k=1, 2, 3, ξ is the identification coefficient, and then the three gray correlation degree is converted into The three primary colors, the three primary colors are expressed as: _{△ ρm = ρ m, max -} ρ m, min, ρ m, mim = min [ρ m 1, ρ m 2, ρ m 3], ρ m, max = max [ρ m 1, ρ m 2, ρ m ₃ ], and then convert the three primary colors into a hue (Hue, H), the hue is expressed as: And representing the color represented by the hue in a display interface to represent the degree of narrowing of the narrowed manifold; wherein the chaotic synchronization theory (CS) uses a Sprott chaotic system, and the calculation of the Sprott chaotic system The method includes the following steps: the computer device inputs the normal sample and the narrowed sample into a main system and a servant system respectively; and the control unit of the computer device calculates a difference between the two to generate a dynamic error signal; the control The unit processes the signal of the dynamic error into a control signal, and inputs the signal to the servant system for controlling synchronization between the servant system and the main system; wherein the mathematical model of the Sprott chaotic system is: the main system is represented as : The servant system is expressed as: , where X = [ x ₁ , x ₂ , x ₃ ], Y = [ y ₁ , y ₂ , y ₃ ], X R ³ , Y R ³ , matrix elements a and b are system parameters, sign( ̇ ) is defined as a signal function, u is a controller to control the servant system to achieve synchronization with the main system; when calculating, unidirectional coupling is used to control The input signal is zero, and the error variable e = [( x ₁ - y ₁ ), ( x ₂ - y ₂ ), ( x ₃ - y ₃ )] = [ e ₁ , e ₂ , e ₃ ] ^T , The error system of the dynamic error is expressed as: The error system of the dynamic error is decoupled and expressed as: As described in the first paragraph of the patent application, the audio technology can be used to obtain the degree of narrowing of the program, wherein the DOS (clinical angiography calculation blocking ratio) is less than 0.3, DOS is between 0.3 and 0.7, and DOS is greater than The three fistula block levels of 0.7. The program product of the narrowing degree can be obtained by the audio technology according to the first aspect of the patent application, wherein the attribution function is an exponential attribution function and is expressed as: Where Ψ _m is the dynamic error, and Mean _{m 1} , Mean _{m 2} and Mean _{m 3} are the average values of the dynamic error Ψ of the three manifold obstruction levels. The program product of the narrowing degree of the tube can be obtained by the audio technology as described in claim 1, wherein the computer device is calculated by the maximum gray correlation degree ρ _{m , max} and the minimum gray correlation degree ρ _{m , min} Brightness and saturation, wherein the brightness is expressed as: v _m = ρ _{m ,max} , and the saturation is expressed as: ( ρ _{m , min} ≠ ρ _{m , max'} ρ _{m , max} ≠ ⁰⁾ , wherein the higher the value S _m and close to 1, the higher the reliability of the narrowing of the narrowed manifold obtained. The program product of the narrowing degree can be obtained by the audio technology according to the first aspect of the patent application, wherein the error system after decoupling is modified into a fractional parameter according to Grunwald-Letnikov's law, so that the error system is represented. for: Discretize the error system to obtain: The dynamic error of the composite fractional order is: , i [1, n -2]. As described in claim 1, the audio technology can be used to obtain a program product with a narrowing degree, wherein the error system is discretized and the error variable e ₁ [ i ]= x [ i ]- y [ i ] is set. , e ₂ [ i ]= x [ i +1]- y [ i +1], e ₃ [ i ]= x [ i +2]- y [ i +2], i =1,2,3,. .., n -2, the equation of the dynamic error is expressed as Ψ=-1.2 e ₁ [ i ]- be ₂ [ i ]- ae ₃ [ i ]+2( sign ( x [ i ])- sign ( y [ i ]), x[i] is taken from the characteristic spectrum of the normal sample, and y[i] is taken from the characteristic spectrum of the narrowed sample. A computer readable medium storing a program product capable of obtaining a narrowing degree of a tube by means of an audio technology, which is capable of obtaining a narrowing degree by using an audio technology as claimed in any one of claims 1 to 6. Program product.