Method for generating traditional Chinese medicine personalized nursing scheme for preventive treatment of diseases

Abstract

The invention relates to the technical field of traditional Chinese medicine nursing or health improvement, in particular to a method for generating a traditional Chinese medicine personalized nursing scheme for preventive treatment of diseases. The method comprises the following steps: firstly, screening out all initial reference individuals of a current individual based on a similar condition between corresponding physique information of the individuals; further determining all concerned symptoms and corresponding concerned coefficients of the current individual according to the universality and change condition of each symptom of the current individual in the initial reference individual; screening out all reference individuals according to the attention coefficient of the common attention symptom of the current individual and each initial reference individual; and finally, generating a personalized nursing scheme of the current individual based on the nursing schemes of all the reference individuals. According to the method and the device, the reference individuals capable of providing similar nursing references are accurately screened out from the historical disease prevention individuals based on the similar physical conditions of the individuals and the similar conditions of the symptoms which are mainly concerned during nursing, so that the generation effect of the personalized scheme of the current individual is improved.

Description

Technical Field

[0001] The present invention relates to the field of traditional Chinese medicine nursing or health improvement technology, and in particular to a method for generating a personalized traditional Chinese medicine nursing plan for preventing diseases. Background Art

[0002] As people's awareness of health and wellness grows, more and more people are focusing on "preventive treatment." Preventive treatment, rooted in Traditional Chinese Medicine (TCM), emphasizes prevention, aiming to prevent illness through conditioning and intervention. This approach embodies a more comprehensive and long-term approach to health management. By identifying individual constitutions and diagnosing symptoms, TCM develops personalized conditioning and care plans for each individual, effectively preventing disease and maintaining health.

[0003] In the existing technology, the rehabilitation recommendation method based on similar cases is used to improve the diagnosis and treatment efficiency, that is, historical cases with similar symptoms to the current individual are matched, and the effective diagnosis and treatment plans of the matched cases provide a certain reference, thereby assisting relevant medical personnel in formulating the diagnosis and treatment plan for the current individual; however, for individuals who are not yet ill, they are in a relatively healthy state, and the various symptom indicators are relatively not obvious, which makes it difficult to match them with similar historical cases, so that the reference value of the corresponding nursing plans of the matched historical cases is low, which in turn leads to poor generation of personalized nursing plans. Summary of the Invention

[0004] In order to solve the technical problem of poor production effect of TCM personalized nursing plans for individuals who are not yet ill in the prior art, the present invention aims to provide a method for generating TCM personalized nursing plans for individuals who are not yet ill. The technical solution adopted is as follows:

[0005] A method for generating a personalized TCM nursing plan for preventing disease, the method comprising:

[0006] Obtain each individual's historical records of preventive treatment, as well as the current individual's physical condition information and all symptoms; the preventive treatment records include at least physical condition information, nursing plans, symptoms, and corresponding change information;

[0007] Based on the similarity between the current individual and the corresponding physical information of each historical individual, a physical similarity parameter between the current individual and each historical individual is obtained; all initial reference individuals are screened out from all historical individuals based on the physical similarity parameter; based on the prevalence of each symptom of the current individual in all the initial reference individuals and the change information of the same symptom in each initial reference individual, an attention coefficient of each symptom of the current individual is obtained; and based on the attention coefficient, all symptoms of interest are screened out from all symptoms of the current individual;

[0008] According to the attention coefficient of the attention symptom shared by the current individual and each of the initial reference individuals, the symptom similarity parameters between the corresponding individuals are obtained; according to the symptom similarity parameters, all reference individuals are screened out from all the initial reference individuals; and according to the care plans of all reference individuals, a personalized care plan for the current individual is generated.

[0009] Furthermore, the constitution information at least includes a conversion score for each constitution obtained based on a preset TCM constitution classification table, and self-test scores corresponding to all constitution self-test questions for each constitution.

[0010] Furthermore, the method for obtaining the similarity parameters of physical constitution includes:

[0011] The self-test scores corresponding to all the self-test questions for each individual under each constitution are used as sequence elements, and are sorted according to the sequence numbers of the constitution self-test questions to construct a representative constitution self-test sequence; based on the similarity between the current individual and each historical individual, the constitution similarity sub-parameter under the corresponding constitution is obtained;

[0012] The conversion score of the current individual under each constitution is used as the constitution tendency weight, and the constitution similarity sub-parameters under the corresponding constitution between the current individual and each historical individual are weighted and averaged using the constitution tendency weight. The weighted average result is normalized to obtain the constitution similarity parameters between the current individual and each historical individual.

[0013] Furthermore, the screening method of the initial reference individual includes:

[0014] All historical individuals corresponding to the time when the physical similarity parameter is greater than a preset parameter threshold are used as all initial reference individuals for the current individual.

[0015] Furthermore, the method for obtaining the symptoms and corresponding change information includes:

[0016] The preventive treatment record also includes medical records of historical individuals at different consultation periods; based on natural language processing technology, all symptom corresponding entities in the medical records of each historical individual at each consultation period, as well as symptom change entities corresponding to each symptom, are obtained, and the sentiment score of each symptom change entity is obtained based on the sentiment dictionary;

[0017] The emotion scores of the symptom change entities corresponding to each symptom at different consultation periods are sorted in chronological order to fit a change curve, and the change curve is used as the change information of the corresponding symptom.

[0018] Furthermore, the method for obtaining the attention coefficient includes:

[0019] Taking any symptom of the current individual as the target symptom, the occurrence frequency of the target symptom in all the initial reference individuals is used as the universal parameter of the target symptom; and the negative correlation mapping result of the universal parameter is used as the special attention parameter of the target symptom;

[0020] The curve subsegment between the starting point and the minimum value of each change curve is used as the deterioration subsegment of the corresponding symptom; according to the severity of the change of each deterioration subsegment, the deterioration parameter of the corresponding symptom is obtained; and the deterioration parameter of the target symptom corresponding to the deterioration subsegment of each initial reference individual is combined to obtain the deterioration focus parameter of the target symptom;

[0021] The special attention parameter and the deterioration attention parameter of the target symptom are integrated to obtain the attention coefficient of the target symptom.

[0022] Furthermore, the method for obtaining the deterioration parameter includes:

[0023] Between the starting point and the end point of each deterioration subsegment, the score difference between the corresponding emotion scores is used as the numerator, the time interval between the corresponding consultation periods is used as the denominator, and the fraction ratio is used as the deterioration parameter.

[0024] Furthermore, the method for obtaining the symptoms of interest includes:

[0025] Among all the symptoms of the current individual, all symptoms whose attention coefficients are greater than a preset coefficient threshold are regarded as attention symptoms.

[0026] Furthermore, the method for obtaining the symptom similarity parameter includes:

[0027] Take any initial reference individual as the target individual, and take the concern symptoms included in all the symptoms of the target individual as the intersection symptoms; take the cumulative sum of the concern coefficients of all intersection symptoms as the numerator, take the cumulative sum of the concern coefficients of all the concern symptoms of the current individual as the denominator, and take the fraction ratio as the symptom similarity parameter between the current individual and the target individual.

[0028] Furthermore, the method for obtaining the reference individual includes:

[0029] All the initial reference individuals whose symptom similarity parameters are greater than a preset similarity threshold are used as reference individuals.

[0030] The present invention has the following beneficial effects:

[0031] The present invention first obtains the preventive treatment record of each historical individual, as well as the physical constitution information and all symptoms of the current individual, in preparation for the subsequent evaluation of physical constitution similarity and symptom similarity to screen reference individuals for the current individual, and also prepares for the subsequent generation of preventive treatment personalized care for the current individual based on the preventive treatment care of the reference individuals; then, based on the similarity of the physical constitution information, the physical constitution similarity parameters between the current individual and each historical individual are obtained, so that all initial reference individuals can be screened out from all historical individuals, in preparation for the subsequent further fine screening of the reference group of the current individual in combination with individual symptoms; then, based on the prevalence of each symptom of the current individual in all initial reference individuals and the change information of the same symptom of each initial reference individual, the attention coefficient of each symptom of the current individual is obtained, so that all the symptoms of concern that need to be focused on when subsequently formulating personalized care are screened out from all the symptoms of the current individual; further, based on the attention coefficient of the symptoms of concern shared by the current individual and each initial reference individual, the symptom similarity parameters between the corresponding individuals are obtained, so that all reference individuals can be screened out from all initial reference individuals. The care plan of the reference individual will provide an accurate reference for the personalized care of the current individual, thereby generating a personalized care plan for the current individual. Based on the similarity of physical constitutions between individuals and the similarity of symptoms that are of primary concern during nursing care, the present invention accurately screens out reference individuals who can provide similar nursing references from individuals with historical preventive care, thereby improving the effect of generating personalized plans for the current individual. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 A flow chart of a method for generating a personalized TCM nursing plan for preventing disease, provided by one embodiment of the present invention;

[0034] Figure 2 A flow chart of a method for obtaining an attention coefficient provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0035] To further illustrate the technical means and efficacy of the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a method for generating a personalized Traditional Chinese Medicine nursing plan for preventing illness according to the present invention, including its specific implementation, structure, features, and efficacy. In the following description, references to different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0036] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0037] The following describes in detail a method for generating a personalized TCM nursing plan for preventing illness provided by the present invention in conjunction with the accompanying drawings.

[0038] See also Figure 1 , which shows a flow chart of a method for generating a personalized TCM nursing plan for preventing disease, provided by one embodiment of the present invention, specifically comprising:

[0039] Step S1, obtain the disease prevention record of each historical individual, as well as the physical information and all symptoms of the current individual; the disease prevention record at least includes physical information, nursing plan, symptoms and corresponding change information.

[0040] To generate a personalized care plan for the current individual to be treated for a disease, one embodiment of the present invention first obtains the disease treatment records of each individual who has received disease treatment care from the TCM electronic medical record system, including the individual's physical constitution information, medical records from different consultation periods, and the personalized care plan formulated by the TCM practitioner for that individual. At the same time, the disease treatment file of the current individual during the current consultation process is obtained, which includes the individual's physical constitution information and current medical records.

[0041] Among them, physical constitution information helps to screen historical individuals to provide preliminary reference for the current individual; medical records include the symptom records of each historical individual during the consultation by traditional Chinese medicine, such as the individual's self-report record, the traditional Chinese medicine's record of their observation, listening, questioning and palpation, etc., to provide data support for the subsequent determination of individual symptoms; personalized care plans include diet improvement plans, exercise plans, medication plans and acupuncture measures formulated by traditional Chinese medicine, etc., to provide a reference for the subsequent generation of personalized care plans for the current individual.

[0042] In a preferred embodiment of the present invention, the constitution information at least includes the conversion score for each constitution obtained based on the preset TCM constitution classification table, and the self-test scores corresponding to all constitution self-test questions under each constitution; wherein, the preset TCM constitution classification table is the existing "TCM Constitution Classification and Determination Self-Test Table", which contains a total of 9 self-test tables for constitutions. The individual fills in each self-test table based on his or her current physical signs and feelings, and obtains the grade score corresponding to each constitution self-test question in each self-test table, that is, the self-test score, and finally calculates the conversion score based on the existing scoring rules.

[0043] It should be noted that the "Self-Test Form for Classification and Determination of Traditional Chinese Medicine Constitution" was released by the China Association of Traditional Chinese Medicine in 2009. It and the method of obtaining the conversion score are already public knowledge and will not be repeated here.

[0044] In another embodiment of the present invention, considering that different individuals have a certain degree of subjectivity in their description and understanding of frequency, this may lead to a reduction in the reference value of the self-test score, thereby affecting the physical fitness judgment; therefore, when an individual fills out a self-test form, the answer to the physical fitness self-test question can be modified to a specific frequency, such as once a week, twice a week, etc., and all frequencies of all historical individuals are further counted, and the frequency value range is determined to be a preset number of level intervals, and then the self-test score of each historical individual under the filling frequency of this question can be determined; wherein the preset number is the same as the number of options or level settings in the self-test form.

[0045] Preferably, in one embodiment of the present invention, considering that natural language processing (NLP) technology can extract relevant symptoms and their emotional description information from a large amount of text data in medical records, thereby helping to evaluate symptom change information; based on this, the method for obtaining symptoms and corresponding change information includes:

[0046] The preventive treatment records also include the medical records of historical individuals at different consultation periods. Based on natural language processing technology, all symptom-corresponding entities in the medical records of each historical individual at each consultation period, as well as the symptom change entities corresponding to each symptom, are obtained. The sentiment score of each symptom change entity is obtained based on the sentiment dictionary.

[0047] The sentiment scores of the symptom change entities corresponding to each symptom at different consultation periods are sorted in chronological order to fit the change curve, and the change curve is used as the change information of the corresponding symptom.

[0048] It should be noted that obtaining entities based on natural language processing technology and obtaining sentiment scores based on sentiment dictionaries are both well-known technologies. Here we only briefly describe the acquisition process: first, the text information corresponding to the medical record is preprocessed, such as using the jieba word segmentation tool for word segmentation, and removing stop words and punctuation marks after word segmentation; then all entities are identified and classified through named entity recognition technology (NER). Specifically, the pre-trained BERT model or spaCy library can be used to identify entities, and the symptoms and symptom change corresponding entities are obtained; then, the sentiment score of the corresponding segmentation of each symptom change entity is obtained based on the NTUSD dictionary, where a positive score indicates positivity and a negative score indicates negativity. The larger the absolute value, the higher the sentiment intensity.

[0049] In other embodiments of the present invention, the implementer may also use a pre-built dictionary to match and obtain all symptom segmentations and symptom change segmentations in the medical record, or may obtain the sentiment scores of the corresponding symptom change segmentations based on other sentiment dictionaries. These are all existing technologies and will not be described in detail.

[0050] In one embodiment of the present invention, the change curve is fitted based on the least square method, which is already a prior art and will not be described in detail.

[0051] Step S2, based on the similarity between the current individual and the corresponding physical information of each historical individual, obtain the physical similarity parameters between the current individual and each historical individual; screen out all initial reference individuals from all historical individuals based on the physical similarity parameters; based on the prevalence of each symptom of the current individual in all initial reference individuals, and the change information of the same symptom corresponding to each initial reference individual, obtain the attention coefficient of each symptom of the current individual; screen out all symptoms of concern from all symptoms of the current individual based on the attention coefficient.

[0052] Taking into account that individuals with similar physical constitutions can provide certain nursing references to each other, the embodiment of the present invention first evaluates the physical similarity parameters between historical individuals and current individuals based on the similarity between physical constitution information, and then screens out all initial reference individuals; the comprehensive physical manifestations of the initial reference individuals are relatively similar to those of the current individual, and serve as the initial screening reference group for the current individual, preparing for further fine screening of the reference group for the current individual in combination with individual symptoms.

[0053] Preferably, in one embodiment of the present invention, although traditional Chinese medicine theory divides constitutions into many categories, an individual's constitution is not a single type, but rather has tendencies towards other constitutions based on a primary constitution; and considering that an individual's conversion score under each constitution preliminarily reflects their tendencies towards various constitutions, a certain reference proportion information can be provided among the numerous constitution types; at the same time, all self-test scores in the self-test table under each constitution reflect the specific physical manifestations of the individual under the corresponding constitution, which can help evaluate the detailed constitution deviations of different individuals under the corresponding constitution, and prepare for the subsequent measurement of constitution similarity between different individuals; therefore, the method for obtaining constitution similarity parameters includes:

[0054] The self-test scores corresponding to all the self-test questions for each individual under each constitution are used as sequence elements, and are sorted according to the sequence numbers of the constitution self-test questions to construct a representative constitution self-test sequence. Based on the similarity between the current individual and each historical individual under each constitution, the constitution similarity sub-parameters under the corresponding constitution are obtained.

[0055] The conversion score of the current individual under each constitution is used as the constitution tendency weight. The constitution tendency weight is used to weightedly average the constitution similarity sub-parameters under the corresponding constitution between the current individual and each historical individual. The weighted average result is normalized to obtain the constitution similarity parameter between the current individual and each historical individual.

[0056] As an example, we first construct a representative sequence of physical self-assessment for each individual under each constitution. The representative sequence of physical self-assessment reflects the specific physical signs of the individual under that constitution. Then, under the same constitution, the DTW similarity between the current individual and the corresponding physical self-assessment representative sequence of each historical individual is taken, and the DTW similarity is used as the constitution similarity sub-parameter. Further, the conversion score of the current individual under each constitution is used to correspond to the constitution similarity sub-parameter. The conversion score is used as the constitution tendency weight to provide a reference tendency for each constitution, which can help comprehensively evaluate the similarity between the constitution information. The weighted average result is then linearly normalized to obtain the constitution similarity coefficient between the current individual and each historical individual.

[0057] It should be noted that in other examples, the implementer may also directly use the sum of the absolute values ​​of the differences between the sequence elements with the same sequence number between two representative sequences of physical self-test as the physical deviation, and then perform a negative correlation mapping on the physical deviation, such as using it as the x in the exponential function exp(-x) with the natural constant e as the base, and use the exponential function value as the physical similarity sub-parameter; it is also possible to use vectors instead of sequences to construct corresponding physical self-test representative vectors, and further obtain the cosine similarity or Euclidean norm between the vectors to measure similarity, thereby obtaining the physical similarity sub-parameter; the acquisition and weighted averaging of this method and DTW similarity are both existing technologies and will not be repeated here;

[0058] In a preferred embodiment of the present invention, all historical individuals corresponding to physical similarity parameters greater than a preset parameter threshold are used as all initial reference individuals of the current individual; wherein the preset parameter threshold is set to 0.65, and the implementer can also customize it.

[0059] Considering that, from the perspective of Traditional Chinese Medicine, groups with similar constitutions may exhibit similarities in certain physiological functions and disease susceptibility, for example, those with Qi deficiency generally have lower immunity and are more susceptible to colds; while those with Yang deficiency may generally be more sensitive to colds and have slower metabolism; and considering that personalized care for individuals requires not only focusing on their constitution but also their main symptoms, if an individual's symptoms are more specific and prone to worsening compared to people with similar constitutions, the development of a personalized care plan should focus more on improving and alleviating these symptoms;

[0060] Based on this, the embodiment of the present invention first obtains the attention coefficient of each symptom of the current individual based on the prevalence of each symptom of the current individual in all initial reference individuals, and the change information of the same symptom corresponding to each initial reference individual; and then screens out all the attention symptoms from all the symptoms of the current individual according to the attention coefficient, so as to further finely screen out all reference individuals from all initial reference individuals in combination with the attention symptoms of the current individual.

[0061] Preferably, in one embodiment of the present invention, the method for obtaining the attention coefficient includes:

[0062] See also Figure 2 , which shows a flow chart of a method for obtaining an attention coefficient provided by an embodiment of the present invention, specifically comprising:

[0063] In step S201, any symptom of the current individual is taken as the target symptom, and the occurrence frequency of the target symptom in all initial reference individuals is taken as the universal parameter of the target symptom; the negative correlation mapping result of the universal parameter is taken as the special attention parameter of the target symptom.

[0064] Considering that the higher the frequency of a symptom of the current individual occurs in the initial reference individual, the more common the symptom is, and the need for special attention to it will be relatively lower.

[0065] As an example, first determine a target symptom, take the total number of initial reference individuals with the target symptom as the numerator, and take the total number of all initial reference individuals as the denominator. The fractional ratio reflects the frequency of occurrence of the target symptom in the initial reference individuals, and the fractional ratio is used as the universal parameter of the target symptom; the greater the frequency of occurrence, the larger the universal parameter, which means that the symptom is more universal, so the universal parameter is negatively mapped and adjusted logically, such as as the x in the exponential function exp(-x) with the natural constant e as the base, and the exponential function value is used as the special attention parameter of the target symptom.

[0066] By changing the target symptom, you can obtain special attention parameters for each symptom of the current individual.

[0067] In step S202, the curve sub-segment between the starting point and the minimum value of each change curve is used as the deterioration sub-segment of the corresponding symptom; according to the dramatic change of each deterioration sub-segment, the deterioration parameter of the corresponding symptom is obtained; and the deterioration parameter of the target symptom corresponding to the deterioration sub-segment of each initial reference individual is combined to obtain the deterioration attention parameter of the target symptom.

[0068] Considering that negative scores in the emotional score indicate negative emotions, it means that the symptom may be bothering the individual or causing depression. The change curve of the emotional score can indirectly help evaluate the deterioration of the individual's symptoms and thus determine the degree of attention to it.

[0069] As an example, first obtain the minimum value in the change curve. The minimum value reflects the extreme state of the individual's negative emotions during different TCM consultations, and indirectly reflects the worst possibility of the symptom. The curve subsegment between the starting point and the minimum value of the change curve is regarded as the deterioration subsegment.

[0070] Further, the deterioration parameters of the symptoms corresponding to each deterioration sub-segment are obtained. The deterioration parameters reflect the severity of the deterioration of the corresponding disease, and indirectly reflect the necessary degree of attention to the disease; then the deterioration parameters of the deterioration sub-segments corresponding to the target symptoms of all initial reference individuals are accumulated, and the accumulated sum is used as the deterioration attention parameter of the target symptom; by changing the target symptom, the deterioration attention parameter of each symptom of the current individual can be obtained.

[0071] In a preferred embodiment of the present invention, the difference in the emotional scores corresponding to the starting point and the end point of the deterioration sub-segment reflects the magnitude of the decline in the deterioration sub-segment; and the time interval between the consultation period corresponding to the starting point and the end point of the deterioration sub-segment reflects the deterioration speed of the deterioration sub-segment; the greater the magnitude of the decline in the deterioration sub-segment and the shorter the decline time, the more severe the deterioration trend of the symptom is and the greater its attention value; therefore, the method for obtaining the deterioration parameter includes:

[0072] Between the starting point and the end point of each deterioration subsegment, the score difference between the corresponding emotion scores is used as the numerator, the time interval between the corresponding consultation periods is used as the denominator, and the fraction ratio is used as the deterioration parameter.

[0073] Step S203: The special attention parameter and the deterioration attention parameter of the target symptom are integrated to obtain the attention coefficient of the target symptom.

[0074] As an example, considering that the special attention parameter and the deterioration attention parameter reflect the necessity of paying attention to the target symptom from the perspective of the particularity of the symptoms and the deterioration trend, the special attention parameter and the deterioration attention parameter of the target symptom are multiplied and fused, and then the product is linearly normalized to obtain the attention coefficient of the target symptom.

[0075] In other examples, the two can be combined through other basic mathematical operations such as addition or weighted summation.

[0076] Preferably, in one embodiment of the present invention, among all the symptoms of the current individual, all symptoms with a concern coefficient greater than a preset coefficient threshold are regarded as concern symptoms; wherein the preset coefficient threshold is set to 0.60, which can also be customized by the implementer.

[0077] Step S3, based on the attention coefficient of the common attention symptoms of the current individual and each initial reference individual, obtain the symptom similarity parameters between the corresponding individuals; screen all reference individuals from all initial reference individuals based on the symptom similarity parameters; generate a personalized care plan for the current individual based on the care plans of all reference individuals.

[0078] Considering that Traditional Chinese Medicine will comprehensively assess and formulate nursing plans based on individual constitutions and symptoms, people with similar constitutions and symptoms will have similar nursing plans. When formulating personalized nursing plans, more attention should be paid to improving and alleviating symptoms.

[0079] Based on this, the embodiment of the present invention further combines the attention coefficient of each symptom of interest with the initial reference individuals after preliminary screening to analyze the commonality of the current individual's symptoms of interest in the initial reference individuals, thereby evaluating the symptom similarity parameters between the current individual and each initial reference individual;

[0080] Among them, the symptom similarity parameter does not reflect the overall symptom similarity between individuals, but the symptom similarity from the perspective of attention symptoms; the attention coefficient provides a certain reference value for evaluating the common situation of attention symptoms between individuals. The larger the attention coefficient and the symptom is shared by the current individual and the initial reference individual, the more similar the symptoms between the current individual and the initial reference individual are, and the greater the reference value provided by the nursing plan of the initial reference individual for the formulation of the nursing plan of the current individual.

[0081] Preferably, in one embodiment of the present invention, the method for obtaining symptom similarity parameters includes:

[0082] Take any initial reference individual as the target individual, and take the concern symptoms included in all the symptoms of the target individual as the intersection symptoms; take the cumulative sum of the concern coefficients of all intersection symptoms as the numerator, take the cumulative sum of the concern coefficients of all concern symptoms of the current individual as the denominator, and take the fraction ratio as the symptom similarity parameter between the current individual and the target individual.

[0083] By changing the target individual, the symptom similarity parameter between the current individual and each initial reference individual can be obtained; since the fractional ratio cannot be greater than 1, the value range of the symptom similarity parameter is 0-1.

[0084] Among them, the intersection symptoms are the common concern symptoms between the current individual and the corresponding initial reference individual. The larger the fraction ratio, the more common concern symptoms and the larger the concern coefficient. It further shows that the larger the symptom similarity parameter between the current individual and the corresponding initial reference individual, the greater the subsequent nursing reference value provided by the initial reference individual.

[0085] After obtaining the symptom similarity parameters, the final reference individuals can be further screened out from all the initial reference individuals.

[0086] Preferably, in one embodiment of the present invention, all initial reference individuals whose symptom similarity parameters are greater than a preset similarity threshold are used as reference individuals; wherein the preset similarity threshold is set to 0.6, which can also be customized by the implementer.

[0087] After screening out all reference individuals for the current individual from all historical individuals, a personalized care plan for the current individual can be generated based on the care plans of all reference individuals.

[0088] In one embodiment of the present invention, an ant colony algorithm is used to automatically generate a personalized care plan for the current individual. Specifically, the symptom similarity coefficient between the current individual and each reference individual is multiplied by a preset value such as 5 or directly mapped to [3,5] to amplify the symptom similarity coefficient. The amplified value is then used as the expected heuristic factor of the ant colony algorithm, and the care plans of all reference individuals are input into the ant colony algorithm, which automatically outputs the personalized care plan for the current individual. It should be noted that the ant colony algorithm is an existing technology well known to those skilled in the art and will not be described in detail.

[0089] In other embodiments of the present invention, historical individuals with age and lifestyle more similar to the current individual may be further screened out from all reference individuals, and the care plan for such individuals may be used as the care plan for the current individual.

[0090] It should be noted that the generated personalized care plan needs to be further combined with the identification of relevant Chinese medicine experts.

[0091] In summary, the present invention first obtains the records of each historical individual's treatment for disease prevention, and the current individual's physical constitution information and all symptoms; then, based on the similarity between the corresponding physical constitution information of the individuals, all the initial reference individuals of the current individual are screened out from all historical individuals; further, based on the prevalence and changes of each symptom of the current individual in the initial reference individuals, all the symptoms of concern of the current individual are determined; then, based on the attention coefficient of the symptoms of concern shared by the current individual and each initial reference individual, all reference individuals are screened out from all the initial reference individuals; finally, a personalized nursing plan for the current individual is generated based on the nursing plans of all reference individuals. Based on the similarity of the physical constitutions between individuals and the similarity of the symptoms that are of primary concern during nursing, the present invention accurately screens out reference individuals who can provide similar nursing references from historical individuals who have been treated for disease prevention, thereby improving the generation effect of personalized plans for the current individual.

[0092] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0093] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A method for generating a personalized TCM nursing plan for preventing disease, characterized in that: The method comprises: Obtain each individual's historical records of preventive treatment, as well as the current individual's physical condition information and all symptoms; the preventive treatment records include at least physical condition information, nursing plans, symptoms, and corresponding change information; Based on the similarity between the current individual and the corresponding physical information of each historical individual, a physical similarity parameter between the current individual and each historical individual is obtained; all initial reference individuals are screened out from all historical individuals based on the physical similarity parameter; based on the prevalence of each symptom of the current individual in all the initial reference individuals and the change information of the same symptom in each initial reference individual, an attention coefficient of each symptom of the current individual is obtained; and based on the attention coefficient, all symptoms of interest are screened out from all symptoms of the current individual; According to the attention coefficient of the attention symptom shared by the current individual and each of the initial reference individuals, the symptom similarity parameters between the corresponding individuals are obtained; according to the symptom similarity parameters, all reference individuals are screened out from all the initial reference individuals; and according to the care plans of all reference individuals, a personalized care plan for the current individual is generated.

2. A method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The constitution information at least includes the conversion score for each constitution obtained based on a preset TCM constitution classification table, and the self-test scores corresponding to all constitution self-test questions for each constitution.

3. A method for generating a personalized TCM nursing plan for preventing disease according to claim 2, characterized in that: The method for obtaining the similarity parameters of physical constitution includes: The self-test scores corresponding to all the self-test questions for each individual under each constitution are used as sequence elements, and are sorted according to the sequence numbers of the constitution self-test questions to construct a representative constitution self-test sequence; based on the similarity between the current individual and each historical individual, the constitution similarity sub-parameter under the corresponding constitution is obtained; The conversion score of the current individual under each constitution is used as the constitution tendency weight, and the constitution similarity sub-parameters under the corresponding constitution between the current individual and each historical individual are weighted and averaged using the constitution tendency weight. The weighted average result is normalized to obtain the constitution similarity parameters between the current individual and each historical individual.

4. The method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The screening method of the initial reference individual comprises: All historical individuals corresponding to the time when the physical similarity parameter is greater than a preset parameter threshold are used as all initial reference individuals for the current individual.

5. The method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The method for obtaining the symptoms and corresponding change information includes: The preventive treatment record also includes medical records of historical individuals at different consultation periods; based on natural language processing technology, all symptom corresponding entities in the medical records of each historical individual at each consultation period, as well as symptom change entities corresponding to each symptom, are obtained, and the sentiment score of each symptom change entity is obtained based on the sentiment dictionary; The emotion scores of the symptom change entities corresponding to each symptom at different consultation periods are sorted in chronological order to fit a change curve, and the change curve is used as the change information of the corresponding symptom.

6. A method for generating a personalized TCM nursing plan for preventing disease according to claim 5, characterized in that: The method for obtaining the attention coefficient includes: Taking any symptom of the current individual as the target symptom, the occurrence frequency of the target symptom in all the initial reference individuals is used as the universal parameter of the target symptom; and the negative correlation mapping result of the universal parameter is used as the special attention parameter of the target symptom; The curve subsegment between the starting point and the minimum value of each change curve is used as the deterioration subsegment of the corresponding symptom; according to the severity of the change of each deterioration subsegment, the deterioration parameter of the corresponding symptom is obtained; and the deterioration parameter of the target symptom corresponding to the deterioration subsegment of each initial reference individual is combined to obtain the deterioration focus parameter of the target symptom; The special attention parameter and the deterioration attention parameter of the target symptom are integrated to obtain the attention coefficient of the target symptom.

7. The method for generating a personalized TCM nursing plan for preventing disease according to claim 6, characterized in that: The method for obtaining the deterioration parameter includes: Between the starting point and the end point of each deterioration subsegment, the score difference between the corresponding emotion scores is used as the numerator, the time interval between the corresponding consultation periods is used as the denominator, and the fraction ratio is used as the deterioration parameter.

8. The method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The method for obtaining the symptoms of concern includes: Among all the symptoms of the current individual, all symptoms whose attention coefficients are greater than a preset coefficient threshold are regarded as attention symptoms.

9. The method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The method for obtaining the symptom similarity parameter includes: Take any initial reference individual as the target individual, and take the concern symptoms included in all the symptoms of the target individual as the intersection symptoms; take the cumulative sum of the concern coefficients of all intersection symptoms as the numerator, take the cumulative sum of the concern coefficients of all the concern symptoms of the current individual as the denominator, and take the fraction ratio as the symptom similarity parameter between the current individual and the target individual.

10. The method for generating a personalized TCM nursing plan for preventing disease according to claim 1, characterized in that: The method for obtaining the reference individual includes: All the initial reference individuals whose symptom similarity parameters are greater than a preset similarity threshold are used as reference individuals.

Images