Construction method and device of comprehensive disease assessment model and electronic equipment

Abstract

The invention relates to a method and device for constructing a comprehensive disease assessment model and electronic equipment, and the method comprises the steps: obtaining a training data set which comprises at least one of ultrasonic quantitative parameters, biochemical test parameters, physiological quantitative parameters and morphological detection quantitative parameters of at least two to-be-tested parts of a plurality of subjects, and evaluation results of the condition of each subject; training the machine learning model according to the training data set to obtain a comprehensive disease assessment model; the input data of the machine learning model at least comprises at least one of ultrasonic quantitative parameters, biochemical test parameters, physiological quantitative parameters and morphological detection quantitative parameters of at least two to-be-detected parts of each subject in the training data set, and the output data are assessment results of diseases of the corresponding subjects in the training data set; the ultrasonic quantitative parameter of each to-be-detected part is determined based on an echo signal corresponding to an ultrasonic signal transmitted to the to-be-detected part, and is used for evaluating the parameter of the state of the to-be-detected part.

Description

Technical Field

[0001] The present disclosure relates to artificial intelligence technology, and more specifically, to a method, device and electronic device for constructing a comprehensive disease assessment model. Background Art

[0002] Portal hypertension, hepatorenal syndrome, and hepatic encephalopathy are three complex medical conditions closely related to liver disease. They are not only related to the symptoms of the liver, but also have certain connections with the symptoms of multiple other organs through complex physiological and pathological mechanisms.

[0003] Portal hypertension is a series of clinical syndromes caused by increased pressure in the portal vein system. It is mainly caused by liver cirrhosis, liver fibrosis, etc., which leads to obstruction of portal vein blood flow, and then causes symptoms such as splenomegaly, hypersplenism, esophageal varices and ascites. These symptoms not only reflect the direct impact of increased pressure in the portal vein system on the liver and spleen, but also affect the normal function of organs such as the esophagus and stomach through the establishment and opening of collateral circulation. In severe cases, it may even cause massive bleeding in the upper gastrointestinal tract.

[0004] Hepatorenal syndrome is a kidney dysfunction caused by severe liver diseases such as cirrhosis. It is mainly manifested by symptoms of renal insufficiency such as oliguria, anuria, and azotemia. The occurrence of hepatorenal syndrome is closely related to the changes in renal hemodynamics caused by liver disease. At the same time, impaired kidney function will further aggravate liver disease, forming a vicious cycle. In addition, hepatorenal syndrome may also be accompanied by symptoms of other organs, such as digestive system symptoms and electrolyte disorders.

[0005] Hepatic encephalopathy is a syndrome of central nervous system dysfunction caused by severe liver disease. It is mainly manifested by symptoms such as impaired consciousness, abnormal behavior and coma. The occurrence of hepatic encephalopathy is related to multiple factors such as ammonia poisoning caused by liver disease, formation of false neurotransmitters, electrolyte and acid-base imbalance. These symptoms not only reflect the dysfunction of the central nervous system, but may also be accompanied by symptoms of other organs, such as digestive system symptoms, respiratory system symptoms, etc.

[0006] When diagnosing and treating diseases similar to the above syndromes, it is necessary to comprehensively consider the manifestations of multiple organ tissues of the patient. Currently, the diagnostic evaluation is relatively complicated. In view of this, it is very important to explore a more comprehensive and comprehensive non-invasive syndrome evaluation system.

[0007] Public Content

[0008] One object of the present disclosure is to provide a new technical solution for a method of constructing a comprehensive disease assessment model.

[0009] According to a first aspect of the present disclosure, a method for constructing a comprehensive disease assessment model is provided, comprising:

[0010] Acquire a training data set, wherein the training data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of a plurality of subjects, and an evaluation result of a symptom of each subject;

[0011] Training the machine learning model according to the training data set to obtain a comprehensive disease assessment model;

[0012] Among them, the input data of the machine learning model includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the evaluation result of the symptoms of each subject corresponding to the training data set; the ultrasound quantitative parameters of each part to be tested are parameters determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

[0013] Optionally, the ultrasonic quantitative parameter is at least one of the volume, nonlinear parameter, scattering characteristic parameter, attenuation characteristic parameter, tissue elasticity parameter, and motion parameter of the part to be measured.

[0014] Optionally, the nonlinear parameter is a parameter corresponding to a nonlinear effect generated when the ultrasonic signal propagates in the part to be measured.

[0015] The nonlinear parameter is determined based on the corresponding relationship between the second harmonic and the nonlinear parameter when the ultrasonic signal propagates in the part to be measured, and the attenuation coefficient when the ultrasonic signal propagates in the part to be measured.

[0016] Optionally, the nonlinear parameter is determined according to the following calculation formula:

[0017]

[0018] Where β represents the nonlinear parameter, ρ0 represents the density of the part to be measured, c0 represents the speed of ultrasound propagation in the part to be measured, f represents the emission frequency of ultrasound, p2(x) represents the amplitude of the second harmonic, x represents the distance of ultrasound propagation, p0 represents the amplitude of ultrasound at the emission source, α1 represents the attenuation coefficient of the fundamental wave, α2 represents the attenuation coefficient of the harmonic, and p2(x) / xp0 2 The value of is when x approaches 0.

[0019] Optionally, the scattering characteristic parameter is at least one of an average intensity parameter, a scattering peak value, a scatterer density, and a scatterer distribution characteristic.

[0020] Optionally, the parameter corresponding to the scatterer distribution feature is determined according to the following calculation formula:

[0021]

[0022] Wherein, k is the parameter corresponding to the scatterer distribution characteristic, β is the ratio of coherent scattering energy to diffuse scattering energy, μ represents the number of effective scatterers in the ultrasonic resolution unit, and the parameters μ and β are calculated based on the mean value of the intensity I of the envelope amplitude. The value obtained by X statistics and the value obtained by U statistics are calculated, wherein the value obtained by X statistics and the value obtained by U statistics are obtained based on XU statistics, wherein the calculation formulas corresponding to U statistics and X statistics are as follows:

[0023] U=E[lnI]-lnE[I]

[0024] X=E[IlnI] / E[I]-E[lnI]

[0025] The calculation of the parameters μ and β can be transformed into the solution of the following nonlinear equations,

[0026] U HK =U,X HK =X

[0027] Among them, U HK and X HK The corresponding calculation formula is as follows:

[0028]

[0029]

[0030] Among them, γ E is the Euler constant, Γ(·) represents the gamma function, is the double gamma function.

[0031] Optionally, the volume of the part to be measured is determined based on the transverse diameter, the anterior-posterior diameter and the superior-inferior diameter of the part to be measured determined by the echo signal; or,

[0032] The volume of the part to be measured is determined based on a three-dimensional image of the part to be measured.

[0033] Optionally, the training of the machine learning model according to the training data set to obtain a comprehensive disease assessment model includes:

[0034] Training the parameters of the machine learning model according to the training data set to obtain a trained machine learning model;

[0035] Acquire a test data set, wherein the test data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of a plurality of subjects and corresponding evaluation results of symptoms of each subject;

[0036] Using the test data set, evaluating the trained machine learning model to obtain an evaluation result;

[0037] When the evaluation indicators in the evaluation results meet the preset requirements, the trained machine learning model is used as a comprehensive disease evaluation model.

[0038] Optionally, the at least two sites to be tested are located in at least two tissues or organs associated with the subject's syndrome; or are located in one tissue or organ associated with the subject's syndrome.

[0039] Optionally, the syndrome includes at least one or more of portal hypertension, hepatorenal syndrome, and hepatic encephalopathy;

[0040] The tissues or organs associated with the syndrome include at least one or more of the liver, kidney, spleen, muscle, brain tissue, pancreas, and adipose tissue.

[0041] Optionally, the ultrasound quantitative parameters of each part to be measured of each subject are parameters determined based on one area of ​​the part to be measured, or the ultrasound quantitative parameters of each part to be measured of each subject are parameters determined based on different areas of the part to be measured.

[0042] According to a second aspect of the present disclosure, there is provided a device for constructing a comprehensive disease assessment model, comprising:

[0043] An acquisition module, used to acquire a training data set, wherein the training data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of multiple subjects, and an evaluation result of a symptom of each subject;

[0044] A training module is used to train a machine learning model according to the training data set to obtain a comprehensive symptom assessment model; wherein the input data of the machine learning model includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the assessment result of the symptoms of each subject corresponding to the training data set as output data; the ultrasound quantitative parameters of each part to be tested are determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

[0045] According to a third aspect of the present disclosure, an electronic device is provided, comprising a memory and a processor, wherein the memory stores a computer program, and the computer program is used to control the processor to operate so as to execute the method according to any one of the first aspects of the present disclosure.

[0046] According to a fourth aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the method described in any one of the first aspects is implemented.

[0047] According to a fifth aspect of the present disclosure, there is provided a comprehensive symptom assessment device, comprising:

[0048] An acquisition module, used to acquire at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of the subject;

[0049] An evaluation module is used to input at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters obtained for at least two test sites into a comprehensive disease evaluation model constructed by the method described in any one of the first aspects to obtain an evaluation result of the comprehensive disease of the subject.

[0050] The method for constructing a comprehensive disease evaluation model provided by the present invention constructs a comprehensive disease evaluation model through the correspondence between at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two tested parts of each subject and the evaluation results of the symptoms of each subject. On the one hand, since the ultrasound quantitative parameters and morphological detection quantitative parameters of each tested part are parameters of the state of the corresponding tested part, they have a high correlation with the evaluation results of the comprehensive disease. Although the biochemical test parameters and physiological quantitative parameters represent the overall health status of the organism, they are also correlated with the evaluation results of the tested parts to a certain extent. Therefore, the constructed comprehensive disease evaluation model has high accuracy and can provide strong support for the evaluation of the comprehensive disease. On the other hand, constructing the comprehensive disease evaluation model can realize the automatic evaluation of the comprehensive disease and improve the evaluation efficiency.

[0051] Features and advantages of the embodiments of the present specification will become apparent from the following detailed description of exemplary embodiments of the present specification with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the specification and, together with the description, serve to explain the principles of the embodiments of the specification.

[0053] Figure 1is a processing flow chart of a method for constructing a comprehensive disease assessment model according to an embodiment of the present invention.

[0054] Figure 2 is a processing flow chart of a method for constructing a comprehensive disease assessment model according to an embodiment of the present invention.

[0055] Figure 3 is a processing flow chart of a method for constructing a comprehensive disease assessment model according to an embodiment of the present invention.

[0056] Figure 4 It is a principle block diagram of a device for constructing a comprehensive disease assessment model according to an embodiment of the present invention.

[0057] Figure 5 is a hardware structure block diagram of an electronic device according to an embodiment of the present invention.

[0058] Figure 6 4 is a functional block diagram of a comprehensive symptom assessment device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0059] Various exemplary embodiments of the present specification will now be described in detail with reference to the accompanying drawings.

[0060] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the embodiments of the present specification and its application or uses.

[0061] It should be noted that like reference numerals and letters refer to similar items in the following figures, and therefore, once an item is defined in one figure, it need not be further discussed in subsequent figures.

[0062] In one embodiment of the present disclosure, a method for constructing a comprehensive disease assessment model is provided. Figure 1 As shown, a method for constructing a comprehensive disease assessment model provided in this embodiment may include the following steps S110 to S120.

[0063] Step S110, obtaining a training data set, the training data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, morphological detection quantitative parameters of at least two test sites of multiple subjects, and evaluation results of symptoms of each subject.

[0064] In some embodiments, at least two sites to be tested are located in at least two tissues or organs associated with the subject's syndrome; or are located in one tissue or organ associated with the subject's syndrome.

[0065] The syndrome includes at least one or more of portal hypertension, hepatorenal syndrome, and hepatic encephalopathy; the tissues or organs associated with the syndrome include at least one or more of the liver, kidney, spleen, muscle, brain tissue, pancreas, and adipose tissue.

[0066] Ultrasonic quantitative parameters are parameters determined based on the echo signal corresponding to the ultrasonic signal transmitted to the part to be tested, and are used to evaluate the state of the part to be tested. Before transmitting the ultrasonic signal to the part to be tested, the position of the part to be tested is first determined to ensure the accuracy of the ultrasonic signal transmission. In some embodiments, an ultrasonic image is acquired, and then the area corresponding to the part to be tested is determined from the ultrasonic image.

[0067] Specifically, before determining the area corresponding to the part to be measured from the ultrasound image, the ultrasound image is denoised, enhanced, and smoothed to improve the quality of the ultrasound image so as to more accurately determine the area corresponding to the part to be measured. The area corresponding to the part to be measured is determined from the ultrasound image using an image processing algorithm. Specifically, key features are extracted from the ultrasound image using an image processing algorithm. The key features include shape features, contour features, and texture features. The image processing algorithm is at least one of an edge detection algorithm, a corner detection algorithm, and a texture analysis algorithm. Based on the extracted key features, each area in the ultrasound image is classified using a classification algorithm to determine the area corresponding to the part to be measured. The classification algorithm is any one of an algorithm based on a support vector machine, an algorithm based on a decision tree, and a neural network model algorithm. A sampling frame is placed in the area corresponding to the determined part to be measured to ensure that the sampling frame covers the entire area and has clear boundaries.

[0068] In some embodiments, after determining the area corresponding to the part to be measured, an ultrasonic signal is transmitted to the area corresponding to the part to be measured, and the transmitted ultrasonic signal should avoid the blood vessels and connective tissues in the part to be measured, and then the ultrasonic echo signal is received. Radio frequency data is collected from the ultrasonic echo signal, and the ultrasonic quantitative parameters of the part to be measured are determined based on the radio frequency data. Before collecting radio frequency data from the ultrasonic echo signal, the ultrasonic echo signal is reconstructed to obtain a reconstructed ultrasonic echo signal to improve the quality of the ultrasonic echo signal and provide a basis for subsequent analysis of the ultrasonic echo signal. The reconstruction of the ultrasonic echo signal includes signal denoising and signal enhancement. The reconstructed ultrasonic echo signal is subjected to multi-dimensional analysis to collect radio frequency data, and then the ultrasonic quantitative parameters of the part to be measured are determined based on the radio frequency data. The multi-dimensional analysis includes time domain analysis (e.g., waveform analysis, statistical analysis), frequency domain analysis (e.g., spectrum analysis, power spectrum analysis) and time-frequency analysis (e.g., short-time Fourier transform, wavelet transform).

[0069] Biochemical test parameters refer to the content of specific chemical substances in blood, urine or other biological fluids. Specific chemical substances include enzymes, proteins, lipids, and electrolytes. Biochemical test parameters can reflect the body's metabolic status, organ function, and overall health status. Changes in biochemical test parameters can indirectly reflect changes in physiological functions and are an important reference for assessing health status, diagnosing diseases, and treating them. For example, blood sugar levels can reflect the state of sugar metabolism and are closely related to the diagnosis and treatment of diabetes; transaminase levels can reflect the health of the liver and are of great significance for the diagnosis of diseases such as hepatitis and cirrhosis. Biochemical test parameters include blood sugar, blood lipids, liver function indicators (such as transaminase), renal function indicators (such as urea nitrogen, creatinine), and electrolytes (such as sodium, potassium, chloride, calcium, and magnesium). Biochemical test parameters can be measured by laboratory equipment such as biochemical analyzers, and blood, urine, and other samples need to be collected for analysis.

[0070] Physiological quantitative parameters reflect the basic life activities of the body, such as the function of the cardiovascular system, temperature regulation, respiratory exchange, etc. Physiological quantitative parameters reflect the physiological function state of the body and are of great significance for the prevention, diagnosis and treatment of diseases. For example, heart rate monitoring can timely detect cardiovascular diseases such as arrhythmia; blood pressure measurement is crucial for the diagnosis and control of diseases such as hypertension. Physiological quantitative parameters include heart rate, blood pressure, body temperature, respiratory rate, and blood oxygen saturation. Physiological quantitative parameters can be measured by portable monitoring equipment or clinical examinations, such as electrocardiographs, sphygmomanometers, thermometers, etc.

[0071] It should be noted that there is a close correlation between biochemical test parameters and physiological quantitative parameters. Changes in physiological quantitative parameters may be caused by changes in biochemical test parameters, and changes in biochemical test parameters may be caused by changes in physiological quantitative parameters. For example, an increase in blood sugar levels may lead to changes in physiological quantitative parameters such as increased heart rate and blood pressure, and a decrease in heart function may also affect the metabolism of biochemical test parameters such as blood sugar. In short, biochemical test parameters and physiological quantitative parameters reflect the health status of organisms from the perspectives of chemical composition and physiological function respectively. In practical applications, the two complement each other and provide important basis for the prevention, diagnosis and treatment of diseases.

[0072] Morphological quantitative parameters include the size, shape, signal intensity of tissues or organs, and morphological characteristics of lesions. Morphological quantitative parameters can be detected by A-ultrasound, B-ultrasound, M-ultrasound, CT, MRI (Magnetic Resonance Imaging) or PET (Positron Emission Computed Tomography).

[0073] Step S120, training the machine learning model according to the training data set to obtain a tissue inflammation assessment model; wherein the input data of the machine learning model includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of the tested part of each subject in the training data set, and the output data of the machine learning model is the inflammation detection result of the tested part of each subject corresponding to the training data set; the ultrasound quantitative parameter is a parameter determined based on the echo signal corresponding to the ultrasonic signal emitted to the tested part, and is used to evaluate the state of the tested part.

[0074] In some embodiments, the assessment result of the condition of each subject is that the subject has the corresponding condition or does not have the corresponding condition.

[0075] Step S120, training the machine learning model according to the training data set to obtain a comprehensive disease assessment model; wherein the input data of the machine learning model includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the evaluation result of the disease of each subject corresponding to the training data set; the ultrasound quantitative parameters of each part to be tested are determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

[0076] Based on the training data set, there is a one-to-one correspondence between the ultrasound quantitative parameters, biochemical detection parameters, physiological quantitative parameters, morphological detection quantitative parameters of at least two tested parts of each subject and the evaluation results of the symptoms of each subject, and this one-to-one correspondence is also input into the machine learning model.

[0077] The method for constructing a comprehensive disease evaluation model provided by an embodiment of the present invention constructs a comprehensive disease evaluation model through the correspondence between at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two tested parts of each subject and the evaluation results of the symptoms of each subject. On the one hand, since the ultrasound quantitative parameters and morphological detection quantitative parameters of each tested part are parameters of the state of the corresponding tested part, they have a high correlation with the evaluation results of the comprehensive disease. Although the biochemical test parameters and physiological quantitative parameters represent the overall health status of the organism, they are also correlated with the evaluation results of the tested parts to a certain extent. Therefore, the constructed comprehensive disease evaluation model has high accuracy and can provide strong support for the evaluation of comprehensive diseases. On the other hand, constructing the comprehensive disease evaluation model can realize the automatic evaluation of comprehensive diseases and improve the evaluation efficiency.

[0078] In some embodiments, the ultrasonic quantitative parameter of each part to be measured is at least one of the volume, nonlinear parameter, scattering characteristic parameter, attenuation characteristic parameter and tissue elasticity parameter of the part to be measured.

[0079] When the ultrasonic quantitative parameters are multiple ones of the volume, nonlinear parameters, scattering characteristic parameters, attenuation characteristic parameters and tissue elasticity parameters of the part to be measured, it is possible to characterize the state of the part to be measured from a multi-dimensional perspective, thereby providing more comprehensive data for constructing an inflammation assessment model for the part to be measured, so that the constructed inflammation assessment model for the part to be measured has higher assessment accuracy.

[0080] In some embodiments, the volume of the part to be measured is determined based on the transverse diameter, anteroposterior diameter and superior-inferior diameter of the part to be measured determined by the echo signal. Specifically, the transverse diameter, anteroposterior diameter and superior-inferior diameter of the part to be measured are substituted into the ellipsoid volume calculation formula to obtain the volume of the part to be measured.

[0081] In some embodiments, the volume of the part to be measured is determined based on a three-dimensional ultrasound image of the part to be measured. A three-dimensional ultrasound image of the part to be measured is obtained by three-dimensional ultrasound imaging technology, and the volume of the part to be measured is calculated based on the three-dimensional ultrasound image. This method for determining the volume of the part to be measured can reduce human measurement errors and improve measurement accuracy.

[0082] Taking the liver as an example, when determining the volume of the liver, it is necessary to select appropriate standard sections, such as the maximum oblique section of the right liver lobe, the anteroposterior radial section of the right liver lobe, the sagittal longitudinal section of the left liver lobe, etc., to ensure the accuracy of the measurement results.

[0083] There is a close relationship between the volume of the part to be tested and its symptom assessment. This relationship is particularly significant in organs such as the liver, kidneys, and spleen, which are not only important components of the human body's physiological systems, but their volume changes are also often used as important indicators for syndromic disease diagnosis, condition assessment, and prognosis.

[0084] The long diameter of a normal liver is about 25 cm, the upper and lower diameter is about 15 cm, and the front-to-back diameter is about 16 cm. An increase in liver volume may be caused by a variety of reasons, such as fatty liver, hepatitis, liver congestion, hepatic schistosomiasis, etc. An enlarged liver may cause symptoms such as indigestion, nausea and vomiting, insomnia, dizziness, tinnitus, and memory loss. Physical examination, B-ultrasound, CT or MRI and other imaging examinations can be used to determine whether the liver is enlarged and the extent of the enlargement. A reduction in liver volume is usually related to decreased liver function, such as cirrhosis, chronic hepatitis, etc. A reduced liver may cause serious symptoms such as jaundice, ascites, and hepatic encephalopathy.

[0085] The size of normal kidneys also varies from person to person. Usually, the left kidney of an adult male is about 10.26cm long, 4.45cm wide, and 4.75cm thick, and the right kidney is about 10.16cm long, 4.21cm wide, and 4.54cm thick. The kidneys of adult women are slightly smaller. Enlarged kidney volume may be caused by diseases such as hydronephrosis, kidney inflammation, and tumors. Enlarged kidneys may cause symptoms such as low back pain, hematuria, and proteinuria. Imaging examinations such as B-ultrasound, CT, or MRI can accurately determine whether the kidneys are enlarged and the cause. Reduced kidney volume is usually related to chronic kidney disease, renal failure, etc. Reduced kidney volume may cause symptoms such as oliguria, anuria, hypertension, and anemia.

[0086] There are certain individual differences in the size of the spleen in normal adults, but generally speaking, its length is about 8 to 12 cm, width is about 6 to 8 cm, and thickness is about 3 to 4 cm. An increase in the size of the spleen may be caused by a variety of diseases, such as inflammatory diseases of the spleen, metastatic tumors, autoimmune diseases, and blood diseases. Splenomegaly may cause symptoms such as left upper abdominal pain, abdominal distension, anemia, jaundice, bleeding spots or ecchymosis. Physical examination, B-ultrasound, CT or MRI and other imaging examinations can be used to determine whether the spleen is enlarged and its cause. A reduction in the size of the spleen is relatively rare and may be caused by certain special diseases or conditions.

[0087] The nonlinear parameter is a parameter corresponding to the nonlinear effect generated by the propagation of the ultrasonic signal in the part to be measured. In some embodiments, the nonlinear parameter is determined based on the corresponding relationship between the second harmonic and the nonlinear parameter when the ultrasonic signal propagates in the part to be measured, and the attenuation coefficient of the ultrasonic signal when propagating in the part to be measured.

[0088] The mathematical relationship between the second harmonic and the nonlinear parameters is as follows:

[0089]

[0090] Among them, p2(x) represents the amplitude of the second harmonic, β represents the nonlinear parameter, and ρ0 represents the density of the part to be measured. represents the speed of ultrasound propagation in the part to be measured, x represents the distance of ultrasound propagation, f represents the emission frequency of ultrasound, and p0 represents the amplitude of ultrasound at the emission source. When ultrasound propagates in the part to be measured, it is necessary to consider the attenuation of ultrasound in the part to be measured, and make corresponding compensation based on the attenuation to improve the accuracy of nonlinear parameter determination. Combining the above mathematical relationship and the attenuation coefficient of the ultrasound signal when propagating in the part to be measured, the nonlinear parameter is obtained based on the following calculation formula:

[0091]

[0092] Among them, α1 represents the attenuation coefficient of the fundamental wave, α2 represents the attenuation coefficient of the harmonic, and when x approaches 0, p2(x) / xp0 2 Substitute the value of into the above calculation formula. When x approaches 0, p2(x) / xp0 2 The value of is determined by determining the second harmonic amplitude at different propagation distances and plotting p2(x) / xp0 2 The curve of x, and through linear deduction, we can get p2(x) / xp0 when x approaches 0 2 It should be noted that when the nonlinear parameter is obtained by using the above calculation formula, the value of x is determined according to the requirements.

[0093] In some embodiments, the scattering characteristic parameter is at least one of an average intensity parameter, a scattering peak value, a scatterer density, and a scatterer distribution feature.

[0094] The scattering characteristic parameter is a parameter used to characterize the energy distribution in various directions of the ultrasonic signal after scattering when propagating in the part to be measured. The scattering characteristic parameter can be at least one of a homodyne K distribution parameter and a nakagami distribution parameter.

[0095] The present invention discloses a probability statistical model of ultrasonic backscattered signal of homodyne K distribution, which is represented by the following probability density function:

[0096]

[0097] Where A represents the amplitude of the ultrasonic backscatter signal envelope, f(A) represents the probability density function of the envelope amplitude, x represents the integral variable, J0(·) is the zero-order first-kind Bessel function, and s 2 represents the energy of the coherent scattered signal, σ 2 represents the energy of the diffuse scattering signal, μ represents the number of effective scatterers in the ultrasonic resolution unit. Define parameter k = s / σ, which represents the ratio of the coherent scattering signal to the diffuse scattering signal.

[0098] Based on the calculation formula corresponding to the above probability density function, the corresponding discrete form can be obtained as follows:

[0099]

[0100] Where Γ(·) represents the gamma function, K p represents the p-order modified Bessel function of the second kind.

[0101] Furthermore, based on the above definition of parameter k, the applicant has noticed the influence of the difference between coherent scattering energy and diffuse scattering energy. In order to further improve the estimation accuracy of parameter k, the present invention also provides an improved evaluation algorithm, which is as follows:

[0102] Define parameter β as the coherent scattering energy s 2 and diffuse scattering energy 2σ 2 The ratio of β = s 2 / (2σ 2 ). Specifically, according to the mean value of the intensity I of the envelope amplitude The values ​​obtained by X statistics and U statistics are used to calculate the parameters μ and β. The values ​​obtained by X statistics and U statistics are statistically analyzed based on the XU statistics method. The corresponding calculation formulas for U statistics and X statistics are as follows:

[0103] U=E[lnI]-lnE[I]

[0104] X=E[IlnI] / E[I]-E[lnI]

[0105] The estimation of parameters μ and β can be transformed into the solution of the following nonlinear equations,

[0106] U HK =U,X HK =X

[0107] Among them, U HK and X HK The corresponding calculation formula is as follows:

[0108]

[0109]

[0110] Among them, γ E is the Euler constant, which is generally taken as 0.577. is the double gamma function.

[0111] Solve the parameters μ and β according to the formula, and then estimate k based on the parameters μ and β:

[0112]

[0113] The improved homodyne K distribution model of the present invention can more accurately evaluate the scatterer distribution characteristic parameter k of the homodyne K distribution.

[0114] The tissue elasticity parameter is a state parameter of the part to be tested. Taking the liver as an example, the tissue elasticity parameter can be obtained using a liver detection system. The liver detection system includes an elasticity detection probe and a control component that is communicatively connected to the elasticity detection probe. The elasticity detection probe includes a pressure measurement component. The elasticity detection probe is used to obtain the current pressure applied to the part to be tested based on the pressure measurement component, and allows the current pressure to be adjusted under the control of the control component so that the current pressure meets the preset requirements. The control component is used to obtain the current pressure applied by the elasticity detection probe to the part to be tested, and identify the current pressure, and control the elasticity detection probe to adjust the current pressure according to the identification result. The elasticity detection probe is also used to trigger vibration excitation to generate shear waves in the part to be tested when the current pressure meets the preset requirements, and emit ultrasonic signals to track the propagation process of the shear waves in the part to be tested. The control component is also used to generate ultrasonic data based on the echo signal carrying the propagation process information in the part to be tested, and process the ultrasonic data to determine the elasticity parameters corresponding to the part to be tested as tissue elasticity parameters.

[0115] The attenuation characteristic parameter is the propagation characteristic parameter of the ultrasonic signal in the part to be measured.

[0116] The motion parameters are the displacement and strain of the measured area of ​​the measured part caused by the force. The force is external pressure or active contraction and relaxation.

[0117] In some embodiments, the ultrasound quantitative parameters of each tested part of each subject are parameters determined based on one area of ​​the tested part, or the ultrasound quantitative parameters of each tested part of each subject are parameters determined based on different areas of the tested part.

[0118] The ultrasonic quantitative parameters of each part to be measured of each subject are parameters determined based on different areas of the part to be measured, which can ensure the accuracy of the determination of the ultrasonic quantitative parameters of the part to be measured. When the ultrasonic quantitative parameters of each part to be measured of each subject are parameters determined based on different areas of the part to be measured, multiple transmitters and multiple receivers are used to realize parallel transmission of ultrasonic signals and parallel reception of ultrasonic echo signals. Multiple transmitters are used to transmit radio frequency signals at the same time to cover a wider area and realize multi-channel and multi-frequency signal acquisition. This parallel transmission method can greatly shorten the signal acquisition time and improve the acquisition efficiency. Equipped with multiple receivers, each receiver is responsible for receiving the signal of a specific transmitter. In this way, synchronous detection can be performed on different areas of the part to be measured at the same time, which improves the speed and efficiency of signal acquisition.

[0119] It should be noted that in order to ensure that each transmitter and receiver does not interfere with each other, an intelligent frequency allocation and management algorithm is used. This algorithm can dynamically adjust the transmission frequency of each transmitter and the receiving frequency of each receiver to ensure the accuracy and stability of signal acquisition. In order to ensure the synchronization of parallel transmission and parallel reception, high-precision clock synchronization technology and calibration algorithms are used. These technologies can ensure that the time error between each transmitter and receiver is within the microsecond level, thereby ensuring the accuracy and consistency of signal acquisition.

[0120] In some embodiments, step S120 specifically includes: training the parameters of the machine learning model according to the training data set to obtain a trained machine learning model; obtaining a test data set, the test data set including at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two tested parts of multiple subjects and the corresponding evaluation results of the symptoms of each subject; using the test data set to evaluate the trained machine learning model to obtain an evaluation result; when the evaluation indicators in the evaluation result meet the preset requirements, the trained machine learning model is used as a comprehensive symptom evaluation model.

[0121] In some embodiments, before using a training data set to train a machine learning model, the data in the training data set is cleaned to remove outliers and missing values ​​to ensure the training accuracy of the subsequent machine learning model.

[0122] In some embodiments, before using the training data set to train the machine learning model, the ultrasound quantitative parameters of each part to be measured are normalized to ensure that the ultrasound quantitative parameters have the same scale to ensure the training accuracy of the subsequent machine learning model.

[0123] In some embodiments, when there are multiple ultrasound quantitative parameters of the subject's part to be tested, a feature selection algorithm is used to screen out ultrasound quantitative parameters whose correlation with the evaluation of the syndrome exceeds a preset threshold value, so as to serve as data in the training data set. Since the selected ultrasound quantitative parameters have a high correlation with the evaluation of the syndrome, the prediction accuracy of the machine learning model is high.

[0124] The screened ultrasound quantitative parameter is at least one of the volume, nonlinear parameter, scattering characteristic parameter, attenuation characteristic parameter and tissue elasticity parameter of the part to be measured.

[0125] The feature selection algorithm may be any of the following algorithms: recursive feature elimination algorithm, tree model-based feature selection algorithm, and dimension reduction algorithm. The dimension reduction algorithm is PCA (Principal Components Analysis) and t-SNE (t-distributed Stochastic Neighbor Embedding).

[0126] In some embodiments, the machine learning model can be any one of the following: a linear regression model, a logistic regression model, a support vector machine, a random forest model, a gradient boosting tree model.

[0127] In some embodiments, when training a machine learning model using data in a training data set, a grid search method or a random search method is used to optimize the parameters of the machine learning model to improve the accuracy and robustness of the machine learning model.

[0128] In some embodiments, when evaluating the trained machine learning model using a test data set, at least one evaluation indicator among the AUC value (Area Under the Curve, ROC curve area value), accuracy, and recall rate is used to evaluate the performance of the trained machine learning model.

[0129] In one example, taking the assessment of portal hypertension as an example, the construction of the assessment model of portal hypertension is specifically described. For details, see Figure 2 Steps S210 to S220 are shown. First, in step S210, a training data set is obtained, and the training data set includes ultrasonic quantitative parameters of the liver, ultrasonic quantitative parameters of the spleen, and evaluation results of portal hypertension of each subject of multiple subjects. In step S220, the ultrasonic quantitative parameters of the liver and the ultrasonic quantitative parameters of the spleen of each subject in the training data set are used as input data, and the evaluation results of portal hypertension of each subject in the training data set are used as output data to train the machine learning model to obtain an evaluation model of portal hypertension. The ultrasonic quantitative parameters of the liver are parameters determined based on the echo signals corresponding to the ultrasonic signals emitted to the liver, and are used to evaluate the state of the liver. The ultrasonic quantitative parameters of the spleen are parameters determined based on the echo signals corresponding to the ultrasonic signals emitted to the spleen, and are used to evaluate the state of the spleen.

[0130] In one example, the construction of the evaluation model for hepatorenal syndrome is specifically described by taking the evaluation of hepatorenal syndrome as an example. Figure 3Steps S310 to S320 are shown. First, in step S310, a training data set is obtained, and the training data set includes ultrasonic quantitative parameters of the liver, ultrasonic quantitative parameters of the kidney, and the evaluation results of hepatorenal syndrome of each subject of multiple subjects. In step S310, the ultrasonic quantitative parameters of the liver and the ultrasonic quantitative parameters of the kidney of each subject in the training data set are used as input data, and the evaluation results of hepatorenal syndrome of each subject in the training data set are used as output data to train the machine learning model to obtain an evaluation model for hepatorenal syndrome. The ultrasonic quantitative parameters of the liver are parameters determined based on the echo signals corresponding to the ultrasonic signals emitted to the liver, and are used to evaluate the state of the liver. The ultrasonic quantitative parameters of the kidney are parameters determined based on the echo signals corresponding to the ultrasonic signals emitted to the kidney, and are used to evaluate the state of the kidney.

[0131] The present invention also provides a device for constructing a comprehensive disease assessment model for implementing any of the above methods. Figure 4 FIG. 4 is a block diagram showing a structure of a device for constructing a comprehensive disease assessment model according to an embodiment of the present invention. Figure 4 As shown, the device for constructing a complex disease assessment model may include an acquisition module 410 and a training module 420 .

[0132] The acquisition module 410 is used to acquire a training data set, which includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two test sites of multiple subjects, and the evaluation results of the symptoms of each subject.

[0133] The training module 420 is used to train the machine learning model according to the training data set to obtain a comprehensive disease assessment model; wherein the input data of the machine learning model includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the assessment result of the disease of each subject corresponding to the training data set as the output data; the ultrasound quantitative parameters of each part to be tested are determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

[0134] In some embodiments, the ultrasound quantitative parameter is at least one of the volume, nonlinear parameters, scattering characteristic parameters, attenuation characteristic parameters, tissue elasticity parameters, and motion parameters of the part to be measured.

[0135] In some embodiments, the nonlinear parameter is a parameter corresponding to the nonlinear effect generated by the propagation of the ultrasonic signal in the measured part, and the nonlinear parameter is determined based on the correspondence between the second harmonic and the nonlinear parameter when the ultrasonic signal propagates in the measured part, and the attenuation coefficient of the ultrasonic signal when propagating in the measured part.

[0136] In some embodiments, the scattering characteristic parameter is at least one of an average intensity parameter, a scattering peak value, a scatterer density, and a scatterer distribution feature.

[0137] In some embodiments, the volume of the part to be measured is determined based on the transverse diameter, the anterior-posterior diameter and the superior-inferior diameter of the part to be measured determined by the echo signal; or, the volume of the part to be measured is determined based on a three-dimensional image of the part to be measured.

[0138] In some embodiments, the training module 420 is also used to: train the parameters of the machine learning model according to the training data set to obtain a trained machine learning model; obtain a test data set, the test data set includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two test parts of multiple subjects and the corresponding evaluation results of the symptoms of each subject; use the test data set to evaluate the trained machine learning model to obtain an evaluation result; when the evaluation indicators in the evaluation result meet the preset requirements, use the trained machine learning model as a comprehensive symptom evaluation model.

[0139] In some embodiments, at least two sites to be tested are located in at least two tissues or organs associated with the subject's syndrome; or are located in one tissue or organ associated with the subject's syndrome.

[0140] In some embodiments, the syndrome includes at least one or more of portal hypertension, hepatorenal syndrome, and hepatic encephalopathy; the tissues or organs associated with the syndrome include at least one or more of the liver, kidneys, spleen, muscle, brain tissue, pancreas, and adipose tissue.

[0141] In some embodiments, the ultrasound quantitative parameters of each tested part of each subject are parameters determined based on one area of ​​the tested part, or the ultrasound quantitative parameters of each tested part of each subject are parameters determined based on different areas of the tested part.

[0142] The present invention also provides an electronic device for implementing any of the above method embodiments. Figure 5 FIG. 5 shows a hardware structure block diagram of an electronic device 500 according to an embodiment of the present invention. Figure 5 As shown, the electronic device 500 includes a processor 510 and a memory 520 for storing instructions executable by the processor 510. The processor 510 is configured to implement the method according to any embodiment of the present disclosure when executing the instructions stored in the memory 520.

[0143] The processor 510 is used to execute computer instructions, which can be written in an instruction set of an architecture such as x86, Arm, RISC, MIPS, SSE, etc. The memory 520 includes, for example, ROM (read-only memory), RAM (random access memory), a non-volatile memory such as a hard disk, etc., which are not limited here.

[0144] An embodiment of the present invention further provides a storage medium on which computer instructions are stored. When the computer instructions are executed by a processor, the steps of the method provided in any of the above embodiments are implemented.

[0145] An embodiment of the present invention further provides a comprehensive symptom assessment device, which performs assessment based on the comprehensive symptom assessment model constructed in any of the above embodiments. Figure 6 FIG. 2 shows a structural block diagram of a comprehensive disease evaluation device according to an embodiment of the present invention. Figure 6 As shown, the complex symptom evaluation device may include an acquisition module 610 and an evaluation module 620 .

[0146] The acquisition module 610 is used to acquire at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of the subject.

[0147] The related introduction of the ultrasonic quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of the tested part can refer to any of the above embodiments, and will not be elaborated here.

[0148] The evaluation module 620 is used to input at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two tested parts into the comprehensive disease evaluation model constructed in any of the above embodiments to obtain the evaluation result of the comprehensive disease of the subject.

[0149] The training process of the comprehensive disease assessment model can refer to any of the above embodiments, and will not be described in detail here.

[0150] In one embodiment, a tissue inflammation assessment device is used to assess portal hypertension, and the tissue inflammation assessment device includes an acquisition module and an assessment module.

[0151] The acquisition module is used to acquire the ultrasonic quantitative parameters of the liver, the ultrasonic quantitative parameters of the spleen, the biochemical test parameters, the physiological quantitative parameters, and the morphological detection quantitative parameters of the subject. The evaluation module is used to input the acquired ultrasonic quantitative parameters of the liver, the ultrasonic quantitative parameters of the spleen, the biochemical test parameters, the physiological quantitative parameters, and the morphological detection quantitative parameters into the comprehensive disease evaluation model to obtain the evaluation result of the portal hypertension of the subject.

[0152] In one embodiment, the tissue inflammation assessment device is used to assess hepatorenal syndrome, and the syndrome assessment device comprises an acquisition module and an assessment module.

[0153] The acquisition module is used to acquire the ultrasonic quantitative parameters of the liver, the ultrasonic quantitative parameters of the kidney, the biochemical test parameters, the physiological quantitative parameters, and the morphological detection quantitative parameters of the subject. The evaluation module is used to input the acquired ultrasonic quantitative parameters of the liver, the ultrasonic quantitative parameters of the kidney, the biochemical test parameters, the physiological quantitative parameters, and the morphological detection quantitative parameters into the comprehensive disease evaluation model to obtain the evaluation result of the hepatorenal syndrome of the subject.

[0154] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. For the device embodiment, its related parts can be referred to the partial description of the method embodiment.

[0155] The above is a description of a specific embodiment of the specification. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0156] The embodiments of the present specification may be systems, methods and / or computer program products. The computer program product may include a computer-readable storage medium carrying computer instructions for causing a processor to implement various aspects of the embodiments of the present specification.

[0157] A computer-readable storage medium may be a tangible device that can hold and store computer instructions used by a computer instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which computer instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0158] The computer instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device through a network layer, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network layer may include copper transmission cables, optical fiber transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. The network layer adapter card or network layer interface in each computing / processing device receives the computer instructions from the network layer and forwards the computer instructions for storage in a computer-readable storage medium in each computing / processing device.

[0159] The flowcharts and block diagrams in the accompanying drawings show the possible architecture, functions and operations of the systems, methods and computer program products according to the multiple embodiments of this specification. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a computer instruction, and a part of a module, a program segment or a computer instruction contains one or more executable computer instructions for realizing the specified logical function. In some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented by a dedicated hardware-based system that performs the specified function or action, or can be implemented by a combination of dedicated hardware and computer instructions. It is well known to those skilled in the art that it is equivalent to implement it by hardware, implement it by software, and implement it by combining software and hardware.

[0160] The embodiments of the present specification have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art without departing from the scope of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.

Claims

1. A method for constructing a comprehensive disease assessment model, characterized in that: include: Acquire a training data set, wherein the training data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of a plurality of subjects, and an evaluation result of a symptom of each subject; Training the machine learning model according to the training data set to obtain a comprehensive disease assessment model; Among them, the input data of the machine learning model includes at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the evaluation result of the symptoms of each subject corresponding to the training data set; the ultrasound quantitative parameters of each part to be tested are parameters determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

2. The method according to claim 1, characterized in that The ultrasonic quantitative parameter is at least one of the volume, nonlinear parameter, scattering characteristic parameter, attenuation characteristic parameter, tissue elasticity parameter, and motion parameter of the part to be measured.

3. The method according to claim 2, characterized in that The nonlinear parameter is a parameter corresponding to the nonlinear effect generated by the propagation of the ultrasonic signal in the part to be measured. The nonlinear parameter is determined based on the corresponding relationship between the second harmonic and the nonlinear parameter when the ultrasonic signal propagates in the part to be measured, and the attenuation coefficient when the ultrasonic signal propagates in the part to be measured.

4. The method according to claim 3, characterized in that The nonlinear parameters are determined according to the following calculation formula: Where β represents the nonlinear parameter, ρ0 represents the density of the part to be measured, c0 represents the speed of ultrasound propagation in the part to be measured, f represents the emission frequency of ultrasound, p2(x) represents the amplitude of the second harmonic, x represents the distance of ultrasound propagation, p0 represents the amplitude of ultrasound at the emission source, α1 represents the attenuation coefficient of the fundamental wave, α2 represents the attenuation coefficient of the harmonic, and p2(x) / xp0 2 The value of is when x approaches 0.

5. The method according to claim 2, characterized in that: The scattering characteristic parameter is at least one of an average intensity parameter, a scattering peak value, a scatterer density, and a scatterer distribution feature.

6. The method according to claim 5, characterized in that The parameters corresponding to the scatterer distribution characteristics are determined according to the following calculation formula: Wherein, k is the parameter corresponding to the scatterer distribution characteristic, β is the ratio of coherent scattering energy to diffuse scattering energy, μ represents the number of effective scatterers in the ultrasonic resolution unit, and the parameters μ and β are calculated based on the mean value of the intensity I of the envelope amplitude. The value obtained by X statistics and the value obtained by U statistics are calculated, wherein the value obtained by X statistics and the value obtained by U statistics are obtained based on XU statistics, wherein the calculation formulas corresponding to U statistics and X statistics are as follows: U=E[lnI]-lnE[I] X=E[IlnI] / E[I]-E[lnI] The calculation of the parameters μ and β can be transformed into the solution of the following nonlinear equations, Among them, U HK and X HK The corresponding calculation formula is as follows: Among them, γ E is the Euler constant, Γ(·) represents the gamma function, is the double gamma function.

7. The method according to claim 2, characterized in that The volume of the part to be measured is determined based on the transverse diameter, the anterior-posterior diameter and the superior-inferior diameter of the part to be measured determined by the echo signal; or, The volume of the part to be measured is determined based on a three-dimensional image of the part to be measured.

8. The method according to claim 1, characterized in that The step of training the machine learning model according to the training data set to obtain a comprehensive disease assessment model comprises: Training the parameters of the machine learning model according to the training data set to obtain a trained machine learning model; Acquire a test data set, wherein the test data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of a plurality of subjects and corresponding evaluation results of symptoms of each subject; Using the test data set, evaluating the trained machine learning model to obtain an evaluation result; When the evaluation indicators in the evaluation results meet the preset requirements, the trained machine learning model is used as a comprehensive disease evaluation model.

9. The method according to claim 1, characterized in that: The at least two sites to be tested are located in at least two tissues or organs associated with the subject's syndrome; or are located in one tissue or organ associated with the subject's syndrome.

10. The method according to claim 9, characterized in that The syndrome includes at least one or more of portal hypertension, hepatorenal syndrome, and hepatic encephalopathy; The tissues or organs associated with the syndrome include at least one or more of the liver, kidney, spleen, muscle, brain tissue, pancreas, and adipose tissue.

11. The method according to claim 1, characterized in that: The ultrasound quantitative parameters of each part to be measured of each subject are parameters determined based on one area of ​​the part to be measured, or the ultrasound quantitative parameters of each part to be measured of each subject are parameters determined based on different areas of the part to be measured.

12. A device for constructing a comprehensive disease assessment model, characterized in that: include: An acquisition module, used to acquire a training data set, wherein the training data set includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of multiple subjects, and an evaluation result of a symptom of each subject; A training module is used to train a machine learning model according to the training data set to obtain a comprehensive symptom assessment model; wherein the input data of the machine learning model includes at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two parts to be tested of each subject in the training data set, and the output data of the machine learning model is the evaluation result of the symptoms of each subject corresponding to the training data set as output data; the ultrasound quantitative parameters of each part to be tested are determined based on the echo signal corresponding to the ultrasonic signal emitted to the part to be tested, and are used to evaluate the state of the part to be tested.

13. An electronic device, characterized in that: The method comprises a memory and a processor, wherein the memory stores a computer program, and the computer program is used to control the processor to operate so as to perform the method according to any one of claims 1 to 11.

14. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 11 is implemented.

15. A comprehensive disease assessment device, characterized in that: include: An acquisition module, used to acquire at least one of ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two to-be-tested parts of the subject; An evaluation module is used to input at least one of the ultrasound quantitative parameters, biochemical test parameters, physiological quantitative parameters, and morphological detection quantitative parameters of at least two tested parts into a comprehensive disease evaluation model constructed by the method according to any one of claims 1 to 11, so as to obtain an evaluation result of the comprehensive disease of the subject.

Images