1. A traditional Chinese medicine formula recommendation method based on case clinical phenotype association degree is characterized by comprising the following steps:

acquiring case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

searching a plurality of homogenization cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the traditional Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

calculating case clinical phenotype association degrees between the symptom information of the target patient and the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogenization cases;

screening a plurality of homogenization cases with similarity greater than a preset threshold value from the plurality of homogenization cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

selecting one reference case from the screened multiple reference cases as a target reference case, and acquiring a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

and calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the final traditional Chinese medicine prescription of the target patient.

2. The method as claimed in claim 1, wherein the step of calculating the case clinical phenotype association between the symptom information of the target patient and the symptom information corresponding to the chinese medical diagnosis result of each homogeneous case to obtain the clinical case phenotype associations between the target patient and the homogeneous cases comprises:

extracting a first feature vector of the symptom information of the target patient;

extracting a second feature vector of symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case;

and calculating the case clinical phenotype association degree between the extracted first feature vector and each second feature vector to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogeneous cases.

3. The method of claim 2, wherein the first feature vector is expressed as:

V_ZJ＝∈Z_j1×∈Z_j2×…×∈Z_jt

wherein Z is_jA plurality of types of physiological abnormality parameters representative of a target patient; e is Z_jk represents the symptom Z_jThe category of the attribute k; t represents the attribute number of the physiological abnormal parameters of a plurality of types; v_ZJIndicating the symptom Z_jThe feature vector of (2);

the second feature vector is represented as:

wherein, Z'_jA plurality of types of physiological abnormality parameters representing homogenized cases;represents symptom Z'_jThe category of the attribute k; t represents the number of attributes of multiple types of physiological abnormality parameters, W_Z′JRepresents symptom Z'_jThe feature vector of (2).

4. The method of claim 3, wherein the specific structure of the clinical phenotype association between the target patient and the homogeneous cases is:

calculating the correlation coefficient of clinical phenotype of the case, and expressing the correlation coefficient as:

wherein n represents target patient symptom information; m represents symptom information of a certain symptomatology case; n n.andgate m represents the number of common symptom information between the symptom information of the target patient and the symptom information of a certain symptomatology case; α (n, m) represents a case clinical phenotype association degree coefficient of the target patient with a certain synclastic case;

calculating a case clinical phenotype association weight value expressed as:

when n and m are equal to 0, beta (n, m) is equal to 0.1

When n and m are equal to 0, beta (n, m) is equal to 1

When n is more than 0 and m is less than 10,

wherein β (n, m) represents a case clinical phenotype association weight value between the target patient symptom information and some homogenization case symptom information;

adjusting the correlation coefficient of the clinical phenotype of the case according to the correlation weight of the clinical phenotype of the case to obtain the correlation coefficient of the clinical phenotype of the case, which is expressed as:

Clinc.Sim(n，m)＝α(n，m)×β(n，m)

sim (n, m) represents the case clinical phenotype association between the target patient and the homogeneous case.

5. The method for recommending a Chinese medicinal composition based on clinical phenotype association of cases as claimed in claim 1, wherein said selecting a reference case from the screened plurality of reference cases as the target reference case comprises:

acquiring multiple types of physiological abnormal parameters of target patient symptom information and weight coefficients thereof;

calculating a first physiological abnormality evaluation value of the target patient according to the acquired physiological abnormality parameters and the weighting coefficients of the physiological abnormality parameters of the symptom information of the target patient to obtain the first physiological abnormality evaluation value;

acquiring multiple types of physiological abnormal parameters and weight coefficients of the physiological abnormal parameters of each reference case symptom information;

calculating a second physiological abnormal evaluation value of the reference case according to the acquired physiological abnormal parameters of the reference case symptom information and the weighting coefficient thereof to obtain a plurality of second physiological abnormal evaluation values;

and screening out a reference case closest to the physiological abnormality degree of the target patient from the plurality of reference cases as the target reference case according to the first physiological abnormality evaluation value and the plurality of second physiological abnormality evaluation values.

6. The method for recommending a Chinese medicinal composition based on clinical phenotype association of a case as claimed in claim 5, wherein the calibrating the Chinese medicinal composition of the target reference case according to the symptom information of the target patient to obtain the Chinese medicinal prescription of the final target patient comprises:

comparing difference information between each physiological abnormality parameter in the target patient symptom information and a corresponding physiological abnormality parameter in the target reference case symptom information;

and updating the corresponding traditional Chinese medicine flavor in the traditional Chinese medicine prescription of the reference case according to each difference information to obtain the target traditional Chinese medicine prescription.

7. The method for recommending a Chinese medicinal composition based on the correlation between clinical phenotypes of cases as claimed in claim 1, wherein selecting a reference case from the screened plurality of reference cases as the target reference case further comprises:

acquiring first medical history information of a target patient;

acquiring second medical history information of each reference case;

according to the acquired first medical history information and each second medical history information, one reference case with the smallest medical history difference degree is selected from the multiple reference cases to serve as a target reference case.

8. The method of claim 7, wherein the method of recommending traditional Chinese medicine prescriptions based on the correlation between clinical phenotypes of patients comprises calibrating a traditional Chinese medicine prescription of a target reference case according to symptom information of a target patient to obtain a final traditional Chinese medicine prescription of the target patient, and further comprises:

and calibrating the traditional Chinese medicine prescription of the reference case according to the symptom information of the target patient and the difference degree of the medical history between the target patient and the target reference case to obtain the target traditional Chinese medicine prescription.

9. A traditional Chinese medicine formula recommendation system based on case clinical phenotype association degree is characterized by comprising:

an acquisition module for acquiring case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

the searching module is used for searching a plurality of homogeneous cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

the calculation module is used for calculating the case clinical phenotype association degree between the symptom information of the target patient and the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogenization cases;

the screening module is used for screening a plurality of homogenization cases with similarity larger than a preset threshold value from the plurality of homogenization cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

the selection module is used for selecting one reference case from the screened multiple reference cases as a target reference case and acquiring a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

and the calibration module is used for calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the traditional Chinese medicine prescription of the final target patient.

10. A traditional Chinese medicine prescription recommendation system based on case clinical phenotype association degree, which comprises a processor, a memory and a storage medium, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the traditional Chinese medicine prescription recommendation method based on case clinical phenotype association degree according to any one of claims 1 to 8 is executed to obtain the traditional Chinese medicine prescription of a target patient.

Background

Traditional Chinese medicine is a basic constitutional unit of a traditional Chinese medicine clinical prescription and is a key for providing diagnosis and treatment schemes for patients by traditional Chinese doctors. The main functional indications of the traditional Chinese medicine and the individual symptoms of patients are important bases and targets for clinical prescription medication, and are important supports for supporting syndrome differentiation and treatment of the traditional Chinese medicine. With the development of informatization technology and artificial intelligence, how to introduce advanced intelligent recommendation technology into the field of traditional Chinese medicine and promote the development of traditional medicine has become a focus of industrial attention and a research hotspot.

In addition, it is worth noting that the clinical medical record of TCM is an important vector for the clinical experience and academic thinking of TCM. Therefore, the clinical decision support system of traditional Chinese medicine is established by summarizing and organizing the medication information of the traditional Chinese medical records and taking a large number of traditional Chinese medical records as the basis, and is a research field with urgent clinical requirements. Compared with western medicine, the traditional Chinese medicine is more flexible, and attaches importance to accumulation of experience and summary of rules, while the traditional Chinese medicine clinical case is a unique rule summary mode, and a good case is greatly helpful for traditional Chinese medicine diagnosis and treatment. Therefore, the medical record is also used as a valuable wealth in the field of traditional Chinese medicine, a large amount of medical record documents related to famous old traditional Chinese medicine exist, and the classical medical records have important significance for syndrome differentiation and medication of clinicians.

At present, the traditional Chinese medicine recommendation method research based on medical records and medical record reasoning mainly comprises the development of an intelligent auxiliary decision-making system for traditional Chinese medicine clinical treatment, such as the traditional Chinese medicine information research institute of Chinese academy of science of traditional Chinese medicine. However, in the prior art, a specific implementation method is not proposed for how to screen typical cases similar to the symptom types of target patients from a large number of cases, how to generate a traditional Chinese medicine prescription of the target patients by using prescription data of the typical cases, and how to implement recommendation of the traditional Chinese medicine prescription, and this also needs to be developed and solved by technical personnel in the field.

However, no effective technical solution has been formed to solve the above problems.

Disclosure of Invention

The invention provides a traditional Chinese medicine formula recommendation method and system based on case clinical phenotype association degree, aiming at solving the technical problems of case reasoning and traditional Chinese medicine formula optimization. Firstly, screening a homogenization case from a preset database according to the age, the sex and the diagnosis result of a target patient; secondly, judging the syndrome of viscera dysfunction of internal organs of the body according to the dialectical analysis of the traditional Chinese medicine, and calculating the similarity between a target patient and a homogenization case by combining physiological test parameters and abnormal main feelings or behavioral manifestations of the patient to obtain specific parameters of the clinical phenotype association degree of the case; thirdly, obtaining a target reference case through a threshold value of the clinical phenotype association degree of the case; and finally, calibrating according to the traditional Chinese medicine formula of the target reference case to obtain the traditional Chinese medicine formula.

In order to achieve the purpose, the invention adopts the following technical scheme:

a traditional Chinese medicine formula recommendation method based on case clinical phenotype association degree comprises the following steps:

acquiring case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

searching a plurality of homogenization cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the traditional Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

calculating case clinical phenotype association degrees between the symptom information of the target patient and the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogenization cases;

screening a plurality of homogenization cases with similarity greater than a preset threshold value from the plurality of homogenization cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

selecting one reference case from the screened multiple reference cases as a target reference case, and acquiring a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

and calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the final traditional Chinese medicine prescription of the target patient.

Further, the calculating a case clinical phenotype association degree between the symptom information of the target patient and the symptom information corresponding to the chinese medical diagnosis result of each of the homogeneous cases to obtain a plurality of case clinical phenotype association degrees between the target patient and the plurality of homogeneous cases specifically includes:

extracting a first feature vector of the symptom information of the target patient;

extracting a second feature vector of symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case;

and calculating the case clinical phenotype association degree between the extracted first feature vector and each second feature vector to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogeneous cases.

Further, the first feature vector is represented as:

V_ZJ＝∈Z_j1×∈Z_j2×…×∈Z_jt

wherein Z is_jA plurality of types of physiological abnormality parameters representative of a target patient; e is Z_jk represents the symptom Z_jThe category of the attribute k; t represents the attribute number of the physiological abnormal parameters of a plurality of types; v_ZJIndicating the symptom Z_jThe feature vector of (2);

the second feature vector is represented as:

wherein, Z'_jShows homogeneityPhysiological abnormality parameters of multiple types of cases;represents symptom Z'_jThe category of the attribute k; t represents the number of attributes of multiple types of physiological abnormality parameters, W_Z′JRepresents symptom Z'_jThe feature vector of (2).

Further, the specific structure of the multiple case clinical phenotype association between the target patient and the multiple homogeneous cases is:

calculating the correlation coefficient of clinical phenotype of the case, and expressing the correlation coefficient as:

wherein n represents target patient symptom information; m represents symptom information of a certain symptomatology case; n n.andgate m represents the number of common symptom information between the symptom information of the target patient and the symptom information of a certain symptomatology case; α (n, m) represents a case clinical phenotype association degree coefficient of the target patient with a certain synclastic case;

calculating a case clinical phenotype association weight value expressed as:

when n and m are equal to 0, beta (n, m) is equal to 0.1

When n and m are equal to 0, beta (n, m) is equal to 1

When n is more than 0 and m is less than 10,

wherein β (n, m) represents a case clinical phenotype association weight value between the target patient symptom information and some homogenization case symptom information;

adjusting the correlation coefficient of the clinical phenotype of the case according to the correlation weight of the clinical phenotype of the case to obtain the correlation coefficient of the clinical phenotype of the case, which is expressed as:

Clinc.Sim(n,m)＝α(n,m)×β(n,m)

sim (n, m) represents the case clinical phenotype association between the target patient and the homogeneous case.

Further, the selecting one reference case from the screened multiple reference cases as the target reference case specifically includes:

acquiring multiple types of physiological abnormal parameters of target patient symptom information and weight coefficients thereof;

calculating a first physiological abnormality evaluation value of the target patient according to the acquired physiological abnormality parameters and the weighting coefficients of the physiological abnormality parameters of the symptom information of the target patient to obtain the first physiological abnormality evaluation value;

acquiring multiple types of physiological abnormal parameters and weight coefficients of the physiological abnormal parameters of each reference case symptom information;

calculating a second physiological abnormal evaluation value of the reference case according to the acquired physiological abnormal parameters of the reference case symptom information and the weighting coefficient thereof to obtain a plurality of second physiological abnormal evaluation values;

and screening out a reference case closest to the physiological abnormality degree of the target patient from the plurality of reference cases as the target reference case according to the first physiological abnormality evaluation value and the plurality of second physiological abnormality evaluation values.

Further, the calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the final traditional Chinese medicine prescription of the target patient specifically includes:

comparing difference information between each physiological abnormality parameter in the target patient symptom information and a corresponding physiological abnormality parameter in the target reference case symptom information;

and updating the corresponding traditional Chinese medicine flavor in the traditional Chinese medicine prescription of the reference case according to each difference information to obtain the target traditional Chinese medicine prescription.

Further, selecting one reference case from the screened multiple reference cases as a target reference case specifically includes:

acquiring first medical history information of a target patient;

acquiring second medical history information of each reference case;

according to the acquired first medical history information and each second medical history information, one reference case with the smallest medical history difference degree is selected from the multiple reference cases to serve as a target reference case.

Further, the calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the final traditional Chinese medicine prescription of the target patient specifically includes:

and calibrating the traditional Chinese medicine prescription of the reference case according to the symptom information of the target patient and the difference degree of the medical history between the target patient and the target reference case to obtain the target traditional Chinese medicine prescription.

Correspondingly, a traditional Chinese medicine formula recommendation system based on the case clinical phenotype association degree is also provided, and comprises:

an acquisition module for acquiring case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

the searching module is used for searching a plurality of homogeneous cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

the calculation module is used for calculating the case clinical phenotype association degree between the symptom information of the target patient and the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogenization cases;

the screening module is used for screening a plurality of homogenization cases with similarity larger than a preset threshold value from the plurality of homogenization cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

the selection module is used for selecting one reference case from the screened multiple reference cases as a target reference case and acquiring a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

and the calibration module is used for calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the traditional Chinese medicine prescription of the final target patient.

Correspondingly, the traditional Chinese medicine formula recommending system based on the case clinical phenotype association degree comprises a processor, a memory and a storage medium, wherein the memory stores computer readable instructions, and when the computer readable instructions are executed by the processor, the traditional Chinese medicine formula recommending method based on the case clinical phenotype association degree is operated to obtain the traditional Chinese medicine formula of the target patient.

Compared with the prior art, the invention not only can realize the rapid generation of the traditional Chinese medicine prescription of the target patient on the basis of the classic traditional Chinese medicine medical record, but also can improve the prescription efficiency of the traditional Chinese medicine by combining the dialectical result of the traditional Chinese medicine and the symptom information of the patient, and provides reference for the dialectical medication of a doctor, thereby ensuring that the traditional Chinese medicine prescription has higher applicability with the target patient and improving the treatment efficiency and the suitability.

Drawings

FIG. 1 is a flowchart of a method for recommending a Chinese medicinal composition based on the correlation between clinical phenotypes of cases according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of a patient information entry interface shown in an embodiment of the present invention;

FIG. 3 is a flowchart illustrating a technique for correlating clinical phenotypes of cases according to an embodiment of the present invention;

FIG. 4 is a flow diagram illustrating a technique for adjusting the weight of the association of clinical phenotypes of cases according to an embodiment of the invention;

fig. 5 is a schematic diagram of an interface for obtaining a homogenized case, shown in an embodiment of the present invention;

FIG. 6 is a schematic diagram of a target case threshold setting interface shown in an embodiment of the present invention;

FIG. 7 is a schematic diagram of a symptom information weight setting interface shown in an embodiment of the present invention;

FIG. 8 is a flowchart illustrating a technique for obtaining a prescription of a target patient;

FIG. 9 is a schematic structural diagram of a system for recommending a Chinese medicinal composition based on the correlation between clinical phenotypes of cases according to an embodiment of the present invention;

fig. 10 is a schematic structural diagram of a traditional Chinese medicine prescription recommending apparatus based on the correlation degree of clinical phenotype of a case according to an embodiment of the present invention.

Detailed Description

The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.

The invention aims to provide a traditional Chinese medicine formula recommendation method and system based on case clinical phenotype association degree aiming at the defects of the prior art.

Example one

The embodiment provides a traditional Chinese medicine formula recommendation method based on case clinical phenotype association degree, as shown in fig. 1, comprising the steps of:

s1, acquiring case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

s2, searching a plurality of homogenization cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

s3, calculating case clinical phenotype association degrees between the symptom information of the target patient and the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogenization cases;

s4, screening a plurality of homogenization cases with similarity larger than a preset threshold value from the plurality of homogenization cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

s5, selecting one reference case from the screened multiple reference cases as a target reference case, and acquiring a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

s6, calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the traditional Chinese medicine prescription of the final target patient.

In step S1, case information corresponding to the target patient is acquired; the case information includes age, sex, diagnosis result of traditional Chinese medicine and corresponding symptom information.

The diagnosis result of the traditional Chinese medicine refers to the disease names obtained based on the basic diagnosis of the traditional Chinese medicine, such as pulse impediment caused by qi deficiency and blood stasis of spleen and kidney, external infection of wind-cold exterior excess, internal heat syndrome, yangming fu-organ excess syndrome, heat accumulation collateral flow syndrome, spleen restriction syndrome and the like; the symptom information refers to subjective feelings of a target patient after the target patient suffers from a disease and various physiological abnormality parameters corresponding to the subjective feelings, and for example, in the case of exogenous wind-cold, physiological abnormalities corresponding to the subjective feelings are mainly reflected in body temperature, cough, heart rate and the like. Therefore, the corresponding physiological abnormal parameters include body temperature parameters, cough parameters (cough frequency and cough sound volume), heart rate parameters, and the like, and in addition, the basic information of the patient can be imported through the electronic medical record data, and the specific system operation interface is shown in fig. 2.

In step S2, a plurality of homogeneous cases with the same gender as the target patient and in the same age group are searched from a preset database according to the age, gender and the diagnosis result of the traditional Chinese medicine; wherein the homogenization case is the same as the target patient's diagnosis in TCM.

The sex refers to male and female, and 3 years can be adopted as one age group; the preset database is a database established based on actual diagnosis cases in the last 10 years or 20 years; or a typical case database established according to the case literature of the famous old Chinese medicine.

It should be noted that the age groups can be set according to actual needs, for example, two years, 4 years, and 5 years can be set as one age group, and these schemes are all within the protection scope of the present invention.

In step S3, a case clinical phenotype association degree between the symptom information of the target patient and the symptom information corresponding to the chinese medical science diagnosis result of each homogeneous case is calculated, and a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogeneous cases are obtained.

In this embodiment, feature extraction is performed on the symptom information to obtain a corresponding feature vector, and then the case clinical phenotype association degree is calculated based on the extracted feature vector. The clinical phenotype association degree of the cases is the similarity degree of clinical phenotypes among the cases, and is the similarity degree of the visceral dysfunction, physiological examination parameters and abnormal main feelings or behavioral manifestations of patients, which are judged according to the dialectics of traditional Chinese medicine.

As shown in fig. 3, the method specifically includes:

s31, extracting a first feature vector of the symptom information of the target patient; wherein the symptom information includes a plurality of types of physiological abnormality parameters, the first feature vector is expressed as:

V_ZJ＝∈Z_j1×∈Z_j2×…×∈Z_jt

wherein Z is_jA plurality of types of physiological abnormality parameters representative of a target patient; e is Z_jk represents the symptom Z_jThe category of the attribute k; t represents the attribute number of the physiological abnormal parameters of a plurality of types; v_ZJIndicating the symptom Z_jThe feature vector of (2);

s32, extracting a second feature vector of the symptom information corresponding to the traditional Chinese medicine diagnosis result of each homogenization case; wherein the second feature vector is represented as:

wherein, Z'_jA plurality of types of physiological abnormality parameters representing homogenized cases;represents symptom Z'_jThe category of the attribute k; t represents the number of attributes of multiple types of physiological abnormality parameters, W_Z′JRepresents symptom Z'_jThe feature vector of (2);

in this embodiment, the first feature vector and the second feature vector may be calculated by using a convolutional neural network commonly used in the prior art to perform extraction, or by using other extraction methods, which is not specifically limited in this embodiment.

And S33, calculating the case clinical phenotype association degree between the extracted first feature vector and each second feature vector to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogeneous cases.

As shown in fig. 4, specifically:

s331, calculating a case clinical phenotype association degree coefficient, which is expressed as:

wherein n represents target patient symptom information; m represents symptom information of a certain symptomatology case; n n.andgate m represents the number of common symptom information between the symptom information of the target patient and the symptom information of a certain symptomatology case; α (n, m) represents a case clinical phenotype association degree coefficient of the target patient with a certain synclastic case;

s332, calculating a clinical phenotype association degree weight value of the case, which is expressed as:

when n and m are equal to 0, beta (n, m) is equal to 0.1

When n and m are equal to 0, beta (n, m) is equal to 1

When n is more than 0 and m is less than 10,

wherein β (n, m) represents a case clinical phenotype association weight value between the target patient symptom information and some homogenization case symptom information;

s333, adjusting the correlation coefficient of the clinical phenotype of the case according to the correlation weight of the clinical phenotype of the case to obtain the correlation coefficient of the clinical phenotype of the case, which is expressed as:

Clinc.Sim(n,m)＝α(n,m)×β(n,m)

sim (n, m) represents the case clinical phenotype association between the target patient and the homogeneous case.

In step S4, a plurality of homogeneous cases with similarity greater than a preset threshold are selected from the plurality of homogeneous cases as reference cases according to the obtained clinical phenotype association of the plurality of cases.

The threshold value is set according to the disease type, the disease duration of the patient and the severity of the complication, and the initial threshold value is 0.65. The number of reference cases selected should be plural. The system operation in which homogeneous case screening is performed is schematically shown in fig. 5 and the threshold setting is shown in fig. 6.

In step S5, one reference case is selected from the screened plurality of reference cases as a target reference case, and a chinese medicine prescription of the target reference case corresponding to the chinese medicine diagnosis result is obtained.

The basis for selecting the reference case as the target reference case may be selected based on the physiological abnormality evaluation value (e.g., steps a1-a5), based on the medical history information (e.g., steps B1-B3), or based on a combination of the physiological abnormality evaluation value and the medical history information, which is not particularly limited in this embodiment.

The screening based on the physiological abnormality evaluation value specifically includes:

A1. acquiring multiple types of physiological abnormal parameters of target patient symptom information and weight coefficients thereof;

A2. calculating and obtaining a first physiological abnormality evaluation value of the target patient according to the acquired multiple types of physiological abnormality parameters of the symptom information of the target patient and the weight coefficient corresponding to each type of physiological abnormality parameter;

for example: for a specific diagnosis result of traditional Chinese medicine, the basic value of the weight coefficients of four physiological abnormality parameters A1, A2, A3 and A4 is set to be 0.65, the disease condition and the rehabilitation process of a patient are comprehensively considered, and the weight coefficients are respectively adjusted to be B1, B2, B3 and B4, so that the first physiological abnormality evaluation value S of the patient is obtained, namely A1 × B1+ A2 × B2+ A3 × B3+ A4 × B4.

A3. Acquiring multiple types of physiological abnormal parameters and weight coefficients of the physiological abnormal parameters of each reference case symptom information;

A4. calculating a second physiological abnormal evaluation value of the reference case according to the multiple types of physiological abnormal parameters of the acquired reference case symptom information and the weight coefficient corresponding to each type of physiological abnormal parameter to obtain multiple second physiological abnormal evaluation values;

it is worth noting that: the setting of the physiological abnormality parameters needs to consider the examination and examination information of the target patient and the reference case, and the calculation method of the second physiological abnormality evaluation value is similar to that of the first physiological abnormality evaluation value, which is not described herein.

A5. And screening out a reference case closest to the physiological abnormality degree of the target patient from the plurality of reference cases as the target reference case according to the first physiological abnormality evaluation value and the plurality of second physiological abnormality evaluation values.

The specific screening method comprises the following steps: and (4) respectively subtracting the plurality of second physiological abnormal evaluation values from the first physiological evaluation value, selecting the second physiological evaluation value with the minimum difference, and taking the reference case corresponding to the selected second physiological evaluation value with the minimum difference as the target reference case.

Screening based on the medical history information specifically comprises the following steps:

B1. acquiring first medical history information of a target patient;

B2. acquiring second medical history information of each reference case;

B3. according to the acquired first medical history information and each second medical history information, one reference case with the smallest medical history difference degree is selected from the multiple reference cases to serve as a target reference case.

The medical history difference degree can be obtained by adopting a feature extraction algorithm to extract a feature vector of symptom information contained in each piece of medical history information, and then calculating the difference degree between the feature vector of the symptom information contained in each piece of second medical history information and the feature vector of the symptom information contained in the first medical history information.

In step S6, the chinese medicine prescription of the target reference case is calibrated according to the symptom information of the target patient, and the chinese medicine prescription of the final target patient is obtained.

Because the symptom information is represented by combination of multiple physiological abnormal parameters, the reference traditional Chinese medicine formula comprises multiple traditional Chinese medicine flavors, and each traditional Chinese medicine flavor usually corresponds to at least one physiological abnormal parameter, the traditional Chinese medicine flavors can be increased and decreased based on the difference information of the physiological abnormal parameters between the target patient and the target reference case, and the optimization of the prescription is realized.

In some embodiments, the Chinese herbal composition comprises a plurality of Chinese herbal medicines, each of which corresponds to at least one of the physiological abnormality parameters.

If the screening based on the physiological abnormality evaluation value is adopted in step S5, step S6 specifically includes:

s61, comparing difference information between each physiological abnormal parameter in the target patient symptom information and the corresponding physiological abnormal parameter in the target reference case symptom information;

s62, updating (i.e. increasing and decreasing) the corresponding traditional Chinese medicine flavor in the traditional Chinese medicine prescription of the reference case according to each difference information to obtain the target traditional Chinese medicine prescription.

The increase and decrease is not simply a linear relationship, but a clinical experience value in which the disease degree and parameter abnormality of the patient are comprehensively considered.

If the step S5 is to be filtered based on the medical history information, the step S6 specifically includes:

s63, calibrating the traditional Chinese medicine prescription of the reference case according to the symptom information of the target patient and the difference degree of the medical history between the target patient and the target reference case to obtain the target traditional Chinese medicine prescription.

The degree of difference in medical history is different from the degree of correction of the traditional Chinese medicine prescription in the reference case, for example, the target traditional Chinese medicine prescription comprises two traditional Chinese medicine combinations with efficacies of tonifying medicine (traditional Chinese medicine for strengthening physique) and exterior-syndrome-removing medicine (traditional Chinese medicine for removing exogenous pathogenic factors), if the difference in medical history shows that the physique and immunity of the target patient relative to the target reference case are stronger, the dosage of the tonifying traditional Chinese medicine can be properly reduced, the dosage of the exterior-removing medicine can be properly increased, and vice versa. A system diagram of the target patient symptom information weight setting is shown in fig. 7.

If the step S5 is to be screened based on the combination of the physiological abnormality evaluation value and the medical history information, the step S6 specifically includes, as shown in fig. 8:

s61, comparing difference information between each physiological abnormal parameter in the target patient symptom information and the corresponding physiological abnormal parameter in the target reference case symptom information;

s62, updating (namely increasing and decreasing) the corresponding traditional Chinese medicine flavor in the traditional Chinese medicine prescription of the reference case according to each difference information;

s63, calibrating the traditional Chinese medicine prescription of the reference case according to the symptom information of the target patient and the difference degree of the medical history between the target patient and the target reference case to obtain the target traditional Chinese medicine prescription.

The rapid generation of the target traditional Chinese medicine prescription is realized in the embodiment, the traditional Chinese medicine prescription efficiency can be improved by combining traditional Chinese medicine diagnosis and patient symptom information, reference is provided for a clinical dialectical prescription of a doctor, and the target traditional Chinese medicine prescription and the target patient have higher applicability.

Example two

The difference between the traditional Chinese medicine formula recommendation method based on the case clinical phenotype association degree provided by the embodiment and the embodiment I is that:

in this embodiment, the clinical phenotype association of the case can be obtained by cosine similarity, which specifically includes:

extracting a first feature vector of symptom information of a target patient;

extracting a second feature vector of the symptom information of each homogeneous case;

and calculating cosine similarity between the first characteristic vector of the symptom information of the target patient and the second characteristic vector of the symptom information of each homogenization case according to the first characteristic vector of the symptom information of the target patient to obtain a plurality of cosine similarities, thereby obtaining the clinical phenotype association degree of the cases between the two cases. The cosine similarity calculation method belongs to the prior art, and therefore, repeated description is not performed.

EXAMPLE III

The difference between the traditional Chinese medicine formula recommendation method based on the case clinical phenotype association degree provided by the embodiment and the embodiment I is that:

in this embodiment, the clinical phenotype association of the case can be obtained by similarity based on euclidean distance, which specifically includes:

extracting a first feature vector of the symptom information of the target patient;

extracting a second feature vector of the symptom information of each homogeneous case;

calculating the Euclidean distance between the first feature vector of the target patient and the second feature vector of each homogenization case;

and calculating the similarity of the symptom information of the target patient and the symptom information of each homogenization case based on the Euclidean distance according to each Euclidean distance to obtain a plurality of similarities based on the Euclidean distance.

In the present embodiment, the calculation method of the euclidean distance is prior art, and therefore, the description will not be repeated. The similarity based on the Euclidean distance is positively correlated with the Euclidean distance, so that the similarity based on the Euclidean distance can be directly compared based on the difference value of the Euclidean distance, and the clinical phenotype association degree of the case can be obtained.

Example four

The traditional Chinese medicine formula recommendation system based on the case clinical phenotype association degree provided by the embodiment, as shown in fig. 9, includes:

an acquisition module 11, configured to acquire case information corresponding to a target patient; wherein the case information comprises age, sex, Chinese medicine diagnosis result and corresponding symptom information;

the searching module 12 is used for searching a plurality of homogeneous cases which have the same sex as the target patient and are in the same age group from a preset database according to the age, the sex and the traditional Chinese medicine diagnosis result; wherein the homogenization case has the same Chinese medicine diagnosis result as the target patient;

the calculation module 13 is configured to calculate a case clinical phenotype association degree between the symptom information of the target patient and the symptom information corresponding to the chinese medical diagnosis result of each homogeneous case, to obtain a plurality of case clinical phenotype association degrees between the target patient and a plurality of homogeneous cases;

the screening module 14 is configured to screen a plurality of homogeneous cases with similarity greater than a preset threshold from the plurality of homogeneous cases as reference cases according to the obtained clinical phenotype association degrees of the plurality of cases;

a selection module 15, configured to select one reference case from the screened multiple reference cases as a target reference case, and obtain a traditional Chinese medicine formula of the target reference case corresponding to a traditional Chinese medicine diagnosis result;

and the calibration module 16 is used for calibrating the traditional Chinese medicine prescription of the target reference case according to the symptom information of the target patient to obtain the traditional Chinese medicine prescription of the final target patient.

It should be noted that the traditional Chinese medicine formula recommendation system based on the case clinical phenotype association degree provided in this embodiment is similar to the embodiment, and is not repeated herein.

In the embodiment, the traditional Chinese medicine prescription of the target patient can be quickly generated on the basis of the classical traditional Chinese medicine case, the traditional Chinese medicine prescription efficiency can be improved by combining the traditional Chinese medicine dialectical result and the patient symptom information, and a reference is provided for dialectical medication of doctors, so that the traditional Chinese medicine prescription and the target patient have higher applicability, and the treatment efficiency and the suitability are improved.

EXAMPLE five

The embodiment provides a traditional Chinese medicine prescription recommending system based on case clinical phenotype association degree, as shown in fig. 10, which includes a processor 1, a memory 2 and a storage medium, wherein the processor 1 and the memory 2 are interconnected and communicated with each other through a communication bus 3 and/or other connection mechanisms (not shown), the memory 2 stores computer-readable instructions executable by the processor 1, and when a computing device runs, the processor 1 executes the computer program to execute a traditional Chinese medicine prescription recommending method based on case clinical phenotype association degree in any optional implementation manner of the embodiment, so as to obtain a traditional Chinese medicine prescription of a target patient.

The computer program, when executed by a processor, performs the method of any of the alternative implementations of the embodiments described above. The storage medium may be implemented by any type of volatile or nonvolatile storage device or combination thereof, such as a Static Random Access Memory (SRAM), an Electrically Erasable Programmable Read-Only Memory (EEPROM), an Erasable Programmable Read-Only Memory (EPROM), a Programmable Read-Only Memory (PROM), a Read-Only Memory (ROM), a magnetic Memory, a flash Memory, a magnetic disk, or an optical disk.

In the embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above-described apparatus embodiments are also merely illustrative, and for example, the division of the units is only one logical division, and there may be other divisions in actual implementation, and for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection of devices or units through some communication interfaces, and may be in an electrical, mechanical or other form.

In addition, units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment.

Furthermore, the functional modules in the embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions.

Compared with the prior art, the invention realizes the rapid generation of the target traditional Chinese medicine prescription, not only can improve the traditional Chinese medicine prescription efficiency by combining the traditional Chinese medicine diagnosis and the patient symptom information and provide reference for the clinical dialectical prescription of doctors, but also can accurately evaluate the prognosis and curative effect of the target patient according to the treatment effect of the target reference case, so that the target traditional Chinese medicine prescription and the target patient have higher suitability.

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.