A system, device and medium for disease classification based on symptom characteristics
Through a disease classification system based on symptom characteristics, using SinHash algorithm and TF-IDF technology, the problem of inaccurate identification of diseases in the existing technology is solved, the accuracy and efficiency of symptom screening are improved, and personalized health management solutions are provided.
Patent Information
- Application Number
- CN202111258506.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-10-27
- Publication Date
- 2024-11-26
- Estimated Expiration
- 2041-10-27
AI Technical Summary
The prior art cannot accurately identify the disease through literal descriptions of symptoms, resulting in misdiagnosis and increased anxiety.
A disease classification system based on symptom characteristics was designed, including disease symptoms collection unit, expansion unit and screening unit. Through SinHash algorithm and TF-IDF technology, a disease symptoms screening model is established to achieve accurate screening of symptoms to identify symptoms and Disease recommendation.
Through text decomposition and screening, the accuracy and efficiency of disease screening are improved, the interference of irrelevant information is reduced, and personalized health management solutions are provided.
Smart Images

Figure CN113990513B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of disease diagnosis, and in particular to a system, device and medium for disease classification based on symptom characteristics. Background Art
[0002] In modern society, the symptoms of many diseases are very similar, but their causes may be different, and the corresponding treatment plans may also be very different. Many people judge what disease they have by simply searching for some of their own symptoms on the Internet, which is not only unscientific, but also easy to cause panic and anxiety. Therefore, a scientific system is needed to know the corresponding disease through the textual description of the symptoms, which is a technical problem that needs to be solved urgently. Summary of the invention
[0003] In view of this, the present invention proposes a system, device and medium for disease classification based on symptom characteristics, which are used to solve the problem that the corresponding disease cannot be known through the textual description of the symptoms.
[0004] The technical solution of the present invention is achieved in this way:
[0005] In a first aspect, the present invention discloses a system for disease classification based on symptom characteristics, the system comprising:
[0006] Disease symptom collection unit: used to collect various disease names and a plurality of corresponding symptom descriptive texts, and store all symptom descriptive texts corresponding to the disease names in the disease symptom database in the form of a symptom word library of the disease name;
[0007] Disease symptom expansion unit: For disease names that already exist in the disease symptom database, add keywords to the corresponding symptom vocabulary to expand the symptom vocabulary; for disease names that do not exist in the disease symptom database, create a new category and use Internet crawler search to create the corresponding symptom vocabulary and store it in the disease symptom database;
[0008] Disease symptom screening model: The disease symptom screening model is established through the SinHash algorithm to screen the corresponding diseases for the identified symptoms;
[0009] Disease symptom screening unit: Screen the corresponding diseases for the identified symptoms through the disease symptom screening model, and recommend the possible diseases and health problems of the examinee as well as personalized health management plans based on the screening results.
[0010] The present invention uses the above system to screen the subject for corresponding diseases based on the subject's disease symptoms, and recommend possible diseases and health problems as well as personalized health management plans to the subject.
[0011] Based on the above technical solution, preferably, the disease symptom collection unit specifically includes:
[0012] Collection subunit: collect various disease names and corresponding symptom descriptive texts;
[0013] Processing subunit: extract keywords from the descriptive text, perform part-of-speech tagging and term extraction on the keywords, store the keywords extracted from the symptom descriptive text into the symptom vocabulary of the disease name, establish a mapping relationship between the disease name and the symptom vocabulary, and store it in the disease symptom database.
[0014] The present invention realizes the collection of disease names and corresponding symptoms through the above-mentioned units, provides sufficient comparative data for disease screening, and establishes a mapping relationship between disease names and symptom thesaurus, which is convenient for the later screening of the symptoms of the subjects.
[0015] Based on the above technical solution, preferably, in the processing subunit, extracting keywords from the descriptive text specifically includes:
[0016] The multiple descriptive texts are automatically segmented, the word frequency and weight of each segmentation in the descriptive text are calculated, all the segmentations are sorted according to the word frequency and weight of each segmentation, and multiple segmentations ranked higher than a preset first threshold are used as keywords of the descriptive text. The keywords of each descriptive text are used as a keyword set, and there are multiple keyword sets in the symptom vocabulary of a disease name.
[0017] The present invention establishes a symptom word library by extracting keywords from descriptive texts, thereby reducing interference from irrelevant information in the descriptive texts and improving disease screening efficiency.
[0018] On the basis of the above technical solution, preferably, calculating the word frequency and weight of each participle in the descriptive text, and sorting all participles according to the word frequency and weight of each participle specifically includes:
[0019] Calculate the word frequency TF of each participle in the descriptive text, where the word frequency TF = the total number of times a participle appears in the article / the total number of participles in the article;
[0020] Calculate the weighted IDF of each participle, where the weighted IDF=log(the total number of symptom descriptive texts of the disease name / (the number of documents containing the participle+1));
[0021] Multiply the word frequency TF and weighted IDF of each word segment to obtain the TF-IDF values of multiple word segments, that is, TF-IDF=TF*IDF. According to the rule of TF-IDF values from high to low, arrange the multiple word segments in order, and select the multiple word segments whose ranking is higher than the preset first threshold as the keywords of the descriptive text.
[0022] The present invention calculates the word frequency and weight of each segmented word in the descriptive text, obtains the top-ranked segmented words as the keywords of the descriptive text, extracts the important parts of the symptom descriptive text, highlights the features, and improves the efficiency of subsequent disease screening.
[0023] Based on the above technical solution, preferably, the disease symptom expansion unit specifically includes:
[0024] Passive expansion subunit: For disease names that already exist in the disease symptom database, the system administrator has the authority to add keywords to the corresponding symptom vocabulary, and the symptom vocabulary is expanded by the system administrator adding keywords to the symptom vocabulary;
[0025] Active expansion subunit: For disease names that do not exist in the disease symptom database, the disease name and the corresponding symptom vocabulary are newly created in the database. The symptom vocabulary uses an Internet crawler to search for the corresponding symptom descriptive text, and extracts keywords from it to create its corresponding symptom vocabulary and store it in the disease symptom database.
[0026] The present invention realizes dynamic expansion of the symptom vocabulary through two methods, active expansion and passive expansion, and ensures the comprehensiveness of the symptom vocabulary.
[0027] Based on the above technical solution, preferably, the disease symptom screening unit specifically includes:
[0028] The user input subunit is used to provide a symptom input interface for the subject, obtain the symptoms input by the subject in the input interface, and perform word segmentation processing to obtain multiple keywords of the subject's symptoms as symptoms to be identified;
[0029] Symptom screening subunit: used to obtain the symptoms to be identified, and input the symptoms to be identified into the disease symptom screening model, obtain the screening results, and recommend the possible diseases and health problems of the subject and personalized health management plans based on the screening results.
[0030] The present invention obtains the symptoms input by the user and performs word segmentation processing on them to obtain keywords, and uses the keywords as search content to perform disease screening, thereby improving the efficiency of disease screening based on disease symptoms.
[0031] Based on the above technical solution, preferably, the disease symptom screening model specifically includes:
[0032] Each keyword in the symptom to be identified is used as the smallest unit of the symptom to be identified, and the fingerprint of each symptom vocabulary in the disease symptom database is calculated by the SimHash algorithm to generate a SimHash table;
[0033] Select a symptom to be identified, calculate the cluster to which the symptom to be identified belongs, and record it in the SimHash table, search the SimHash table according to the fingerprint of the symptom to be identified, and obtain a symptom vocabulary set S0 similar to the symptom to be identified;
[0034] Using formula
[0035]
[0036] Calculate the similarity value between each symptom lexicon in the symptom lexicon set S0 and the symptom to be identified, where sim(A,B) is the similarity value, |A∩B| represents the number of identical keywords in the symptom A to be identified and the symptom lexicon B, |A| represents the total number of keywords in the symptom A to be identified, and |B| represents the total number of keywords in the symptom lexicon B. The symptom lexicon whose similarity value with the symptom to be identified is greater than the preset similarity is taken as the corresponding lexicon set S1, and the disease names corresponding to all symptom lexicons in the corresponding lexicon set S1 are taken as the screening results.
[0037] The present invention constructs a disease screening model through SimHash, takes keywords as the key to the screening mechanism, and greatly improves the screening speed of diseases.
[0038] On the basis of the above technical solution, preferably, searching the SimHash table according to the fingerprint of the symptom to be identified to obtain a symptom vocabulary set S0 similar to the symptom to be identified specifically includes:
[0039] SimHash is used to calculate the Hamming distance between the current symptom fingerprint to be identified and each symptom vocabulary fingerprint in the disease symptom database, and the symptom vocabulary whose Hamming distance to the symptom to be identified is less than the preset distance is taken as the symptom vocabulary set S0 similar to the symptom to be identified.
[0040] A second aspect of the present invention discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus; wherein the processor, memory and communication interface communicate with each other via the bus; the memory stores a system program for disease classification based on symptom characteristics that can be executed by the processor, and the system program for disease classification based on symptom characteristics is configured to implement a system for disease classification based on symptom characteristics as described in the first aspect of the present invention.
[0041] According to a third aspect of the present invention, a computer-readable storage medium is disclosed, on which is stored a system program for classifying diseases based on symptom characteristics. When the system program for classifying diseases based on symptom characteristics is executed, the system for classifying diseases based on symptom characteristics as described in the first aspect of the present invention is implemented.
[0042] The system, device and medium for disease classification based on symptom characteristics of the present invention have the following beneficial effects compared with the prior art:
[0043] (1) By decomposing and screening the text, the technical effect of screening out the corresponding disease can be achieved only through the textual description of the symptoms.
[0044] (2) For the symptom descriptive texts corresponding to the identified symptoms and disease names, the keyword extraction method is used for screening, which reduces the interference of other language information and improves the accuracy of disease screening. The SimHash algorithm is then used to establish a disease screening model, which is combined with the keyword screening method to improve the efficiency of disease screening. BRIEF DESCRIPTION OF THE DRAWINGS
[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 creative work.
[0046] Figure 1 This is a system unit structure diagram for disease classification based on symptom characteristics of the present invention. DETAILED DESCRIPTION
[0047] The following will be combined with the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0048] Example
[0049] The system unit structure of a disease classification based on symptom characteristics of the present invention is shown in Figure 1 , the unit functions are described as follows:
[0050] In the disease symptom collection unit, the collection subunit collects various disease names and the corresponding symptom descriptive texts; the processing subunit extracts keywords from the descriptive text, performs part-of-speech tagging and term extraction on the keywords, stores the keywords extracted from the descriptive text into the symptom vocabulary of the disease name, and establishes a mapping relationship between the disease name and the symptom vocabulary, and stores it in the disease symptom database.
[0051] It should be understood that extracting keywords from the descriptive text includes the following steps:
[0052] Automatically segment multiple descriptive texts corresponding to a disease name, calculate the word frequency and weight of each segmentation in the descriptive text, sort all the segmentations according to the word frequency and weight of each segmentation, and use multiple segmentations with a ranking higher than a preset first threshold as keywords of the descriptive text. The keywords of each descriptive text are used as a keyword set, and there are multiple keyword sets in the symptom vocabulary of a disease name.
[0053] It should be understood that calculating the frequency and weight of each segmentation in the descriptive text and sorting all segmentations according to the frequency and weight of each segmentation specifically includes the following steps:
[0054] Step 1, calculate the word frequency TF of each participle in the descriptive text, where the word frequency (TF) = the total number of times a participle appears in the article / the total number of participles in the article;
[0055] Step 2, calculate the weighted IDF of each participle, the weight (IDF) = log (the total number of symptom descriptive texts of the disease name / (the number of documents containing the participle + 1));
[0056] Step 3, multiply the word frequency TF and weighted IDF of each word segment to obtain the TF-IDF values of multiple word segments, that is, TF-IDF = TF*IDF. According to the rule of TF-IDF values from high to low, arrange the multiple word segments in order, and select the multiple word segments whose ranking is higher than the preset first threshold as the keywords of the descriptive text.
[0057] For example, the disease name: platelet-type pseudo-von Willebrand disease, corresponds to multiple disease symptom descriptive texts: "The main manifestations of the disease are mild to moderate skin and mucous membrane bleeding, such as gingival bleeding, nose bleeding, bleeding after tooth extraction, bleeding after tonsillectomy or other surgical operations, and bleeding worsens after taking aspirin or other antiplatelet drugs, but some patients do not bleed even after surgical operations", etc. After segmenting each descriptive text, multiple segmentations are obtained. The segmentations of the above descriptive text are: "skin and mucous membrane", "bleeding", "gingiva", "surgery", "aspirin". "Antiplatelet drugs", etc., calculate the word frequency and weight of multiple word segments, obtain the TF-IDF value of each word segment, take the word segment with a TF-IDF value higher than 5 from high to low as the keyword, obtain the keywords of the descriptive text, and store them in the symptom vocabulary of platelet-type pseudo-von Willebrand disease in the form of a set. Taking the above descriptive text as an example, its keyword set is {bleeding, surgical operation, after, more than, skin and mucous membrane}, establish a mapping relationship between the symptom vocabulary and platelet-type pseudo-von Willebrand disease, and store the mapping relationship and the symptom vocabulary in the disease symptom database.
[0058] The disease symptom expansion unit automatically runs to dynamically expand the disease symptom database. Go to step 3.
[0059] It should be understood that the disease symptom expansion unit has two modes: active expansion and passive expansion. The passive expansion subunit provides the system administrator with the function of adding keywords to the symptom vocabulary for passive expansion; the active expansion subunit creates a new disease name and a corresponding symptom vocabulary in the database for a disease name that does not exist in the disease symptom database. The symptom vocabulary uses an Internet crawler to search for the corresponding symptom descriptive text, extracts keywords from it, creates the corresponding symptom vocabulary, and stores it in the disease symptom database.
[0060] For example, the system administrator can add keywords such as "aspirin" and "antiplatelet drugs" to the symptom vocabulary of platelet-type pseudo-von Willebrand disease through the passive expansion subunit, and passively expand the symptom vocabulary of platelet-type pseudo-von Willebrand disease; the active expansion subunit automatically searches online to see if some disease names exist in the disease symptom database. If not, it will be actively expanded. Assuming that "pediatric hemophilia" does not exist in the disease symptom database, the active expansion subunit will automatically create a new symptom vocabulary corresponding to "pediatric hemophilia", search for the corresponding symptom descriptive text on the designated authoritative website through an Internet crawler, extract keywords from it, and store them in the corresponding symptom vocabulary. Finally, the disease name and symptom vocabulary are stored in the disease symptom database. It is worth mentioning that the method of extracting keywords still uses the above-mentioned method of segmenting words and calculating word frequency and weight.
[0061] In the disease symptom screening unit, the user input subunit obtains the symptoms input by the subject in the input interface, and performs word segmentation processing to obtain multiple keywords as symptoms to be identified; the symptom screening subunit obtains the symptoms to be identified and inputs the symptoms to be identified into the disease symptom screening model.
[0062] For example, the subject inputs "there is a situation of continuous bleeding, which occurs after taking aspirin", and the user input sub-unit performs word segmentation on this sentence to obtain the keywords "continuous bleeding" and "aspirin", and uses them as the symptoms to be identified. The symptom screening sub-unit obtains the symptoms to be identified and inputs them into the disease symptom screening model.
[0063] It should be understood that the disease symptom screening model performs disease screening through the SimHash algorithm in the following steps:
[0064] Each keyword in the symptom to be identified is used as the smallest unit of the symptom to be identified, and the fingerprint of each symptom vocabulary in the disease symptom database is calculated by the SimHash algorithm to generate a SimHash table;
[0065] Select a symptom to be identified, calculate the cluster to which the symptom to be identified belongs, and record it in the SimHash table, search the SimHash table according to the fingerprint of the symptom to be identified, and obtain a symptom vocabulary set S0 similar to the symptom to be identified;
[0066] Using formula
[0067]
[0068] Calculate the similarity value between each symptom lexicon in the symptom lexicon set S0 and the symptom to be identified, where sim(A,B) is the similarity value, |A∩B| represents the number of identical keywords in the symptom A to be identified and the symptom lexicon B, |A| represents the total number of keywords in the symptom A to be identified, and |B| represents the total number of keywords in the symptom lexicon B. The symptom lexicon whose similarity value with the symptom to be identified is greater than the preset similarity is taken as the corresponding lexicon set S1, and the disease names corresponding to all symptom lexicons in the corresponding lexicon set S1 are taken as the screening results.
[0069] It should be understood that the SimHash table is searched according to the fingerprint of the symptom to be identified to obtain the symptom vocabulary set S0 similar to the symptom to be identified. The specific steps are as follows:
[0070] SimHash is used to calculate the Hamming distance between the current symptom fingerprint to be identified and each symptom vocabulary fingerprint in the disease symptom database, and the symptom vocabulary whose Hamming distance to the symptom to be identified is less than the preset distance is taken as the symptom vocabulary set S0 similar to the symptom to be identified.
[0071] For example, taking the above-mentioned symptoms to be identified as an example, the keywords "continuous bleeding" and "aspirin" are used as the smallest units of the symptoms to be identified, and the symptom word library fingerprint of each disease name in the disease symptom database is calculated by the SimHash algorithm to generate a SimHash table; the cluster to which the symptoms to be identified belong is calculated, and it is recorded in the SimHash table, and the Hamming distance between the current symptom fingerprint to be identified and each symptom word library fingerprint in the disease symptom database is calculated by SimHash, and the symptom word library with a Hamming distance less than the preset distance to the symptom to be identified is used as the symptom word library set S0 similar to the symptom to be identified, which is {platelet-type pseudo-von Willebrand disease, childhood hemophilia, acquired hemophilia, myeloid anemia, von Willebrand disease} in this example; and then the formula is used
[0072]
[0073] Calculate the similarity value between each symptom lexicon in the symptom lexicon set S0 and the symptom to be identified, and use the symptom lexicon whose similarity value with the symptom to be identified is greater than the preset similarity as the corresponding lexicon set S1, and use the disease names corresponding to all symptom lexicons in the corresponding lexicon set S1 as the screening results. In this example, S1 is {platelet-type pseudo-von Willebrand disease, vascular hemophilia}, which is the final screening result.
[0074] In the fifth step, the symptom screening subunit in the disease symptom screening unit obtains the screening results and recommends the possible diseases and health problems of the subject as well as personalized health management plans based on the screening results.
[0075] In the above example, the symptom screening subunit recommends the possible diseases and health problems as well as personalized health management plans to the subject based on the final screening results, namely, platelet-type pseudo-von Willebrand disease and von Willebrand disease.
[0076] The present invention also discloses an electronic device, comprising: at least one processor, at least one memory, a communication interface and a bus; wherein the processor, memory and communication interface communicate with each other via the bus; the memory stores a system program for disease classification based on symptom characteristics that can be executed by the processor, and the system program for disease classification based on symptom characteristics is configured to implement a system for disease classification based on symptom characteristics as described in the first aspect of the present invention.
[0077] The present invention also discloses a computer-readable storage medium, on which is stored a system program for disease classification based on symptom characteristics. When the system program for disease classification based on symptom characteristics is executed, a system for disease classification based on symptom characteristics as described in the first aspect of the present invention is implemented.
[0078] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.
Claims
1. A system for disease classification based on symptom characteristics, characterized in that: include: Disease symptom collection unit: used to collect various disease names and a plurality of corresponding symptom descriptive texts, and store all symptom descriptive texts corresponding to the disease names in the disease symptom database in the form of a symptom word library of the disease name; Disease symptom expansion unit: For disease names that already exist in the disease symptom database, add keywords to the corresponding symptom vocabulary to expand the symptom vocabulary; for disease names that do not exist in the disease symptom database, create a new category and use Internet crawler search to create the corresponding symptom vocabulary and store it in the disease symptom database; Disease symptom screening model: A disease symptom screening model is established through the SimHash algorithm to screen for corresponding diseases based on the symptoms to be identified; Disease symptom screening unit: Screen the corresponding diseases for the symptoms to be identified through the disease symptom screening model, and recommend the possible diseases and health problems of the subjects and personalized health management plans based on the screening results; The disease symptom collection unit specifically includes: Collection subunit: collect various disease names and corresponding symptom descriptive texts; Processing subunit: extracting keywords from the descriptive text, and performing part-of-speech tagging and term extraction on the keywords, storing the keywords extracted from the symptom descriptive text into a symptom word library of the disease name, and establishing a mapping relationship between the disease name and the symptom word library, and storing the keywords into a disease symptom database; In the processing subunit, extracting keywords from the descriptive text specifically includes: Automatically segment the multiple descriptive texts, calculate the word frequency and weight of each segmentation in the descriptive texts, sort all the segmentations according to the word frequency and weight of each segmentation, and use multiple segmentations ranked higher than a preset first threshold as keywords of the descriptive text, each keyword of the descriptive text as a keyword set, and there are multiple keyword sets in a symptom word library of a disease name; The step of calculating the word frequency and weight of each word in the descriptive text and sorting all words according to the word frequency and weight of each word specifically includes: Calculate the word frequency TF of each participle in the descriptive text, where the word frequency TF = the total number of times a participle appears in the article / the total number of participles in the article; Calculate the weighted IDF of each word, where the weighted IDF=log(the total number of symptom descriptive texts of the disease name / (the number of documents containing the word+1)); Multiply the word frequency TF and weighted IDF of each word segment to obtain the TF-IDF values of multiple word segments, that is, TF-IDF=TF*IDF. According to the rule of TF-IDF values from high to low, arrange the multiple word segments in order, and select the multiple word segments whose ranking is higher than the preset first threshold as the keywords of the descriptive text; The disease symptom expansion unit specifically includes: Passive expansion subunit: for disease names that already exist in the disease symptom database, the symptom lexicon is expanded by the system administrator adding keywords to the symptom lexicon; Active expansion subunit: for a disease name that does not exist in the disease symptom database, the disease name and the corresponding symptom word library are newly created in the database, the symptom word library uses an Internet crawler to search for the corresponding symptom descriptive text, and extracts keywords from it to create its corresponding symptom word library and store it in the disease symptom database; The disease symptom screening unit specifically includes: The user input subunit is used to provide a symptom input interface for the subject, obtain the symptoms input by the subject in the input interface, and perform word segmentation processing to obtain multiple keywords of the subject's symptoms as symptoms to be identified; Symptom screening subunit: used to obtain the symptoms to be identified, input the symptoms to be identified into the disease symptom screening model, obtain the screening results, and recommend the possible diseases and health problems of the subject and personalized health management plan based on the screening results; The disease symptom screening model specifically includes: Each keyword in the symptom to be identified is used as the smallest unit of the symptom to be identified, and the fingerprint of each symptom vocabulary in the disease symptom database is calculated by the SimHash algorithm to generate a SimHash table; Select a symptom to be identified, calculate the cluster to which the symptom to be identified belongs, and record it in the SimHash table, search the SimHash table according to the fingerprint of the symptom to be identified, and obtain a symptom vocabulary set S0 similar to the symptom to be identified; Using formula ; Calculate the similarity value between each symptom word library in the symptom word library set S0 and the symptom to be identified, where sim(A,B) is the similarity value, |A∩B| represents the number of identical keywords in the symptom to be identified A and the symptom word library B, |A| represents the total number of keywords in the symptom to be identified A, and |B| represents the total number of keywords in the symptom word library B. The symptom word library having a similarity value greater than a preset similarity with the symptom to be identified is taken as the corresponding word library set S1, and the disease names corresponding to all symptom words in the corresponding word library set S1 are taken as the screening results; The method of searching the SimHash table according to the fingerprint of the symptom to be identified to obtain a symptom vocabulary set S0 similar to the symptom to be identified specifically includes: SimHash is used to calculate the Hamming distance between the current symptom fingerprint to be identified and each symptom vocabulary fingerprint in the disease symptom database, and the symptom vocabulary whose Hamming distance to the symptom to be identified is less than the preset distance is taken as the symptom vocabulary set S0 similar to the symptom to be identified.
2. An electronic device, characterized in that: It includes at least one processor, at least one memory, a communication interface and a bus; wherein the processor, memory and communication interface communicate with each other through the bus; the memory stores a system program for disease classification based on symptom characteristics that can be executed by the processor, and the system program for disease classification based on symptom characteristics is configured to implement the unit function of a system for disease classification based on symptom characteristics as claimed in claim 1.
3. A computer-readable storage medium, characterized in that: The storage medium stores a system program for disease classification based on symptom characteristics, which, when executed, implements the unit functions of a system for disease classification based on symptom characteristics as claimed in claim 1.
