Screening method, device, storage medium and electronic device for typhoid and paratyphoid

Abstract

The application discloses a typhoid and paratyphoid screening method, device, storage medium and electronic device. The typhoid and paratyphoid screening method comprises the following steps: acquiring case information of typhoid and paratyphoid; wherein the case information at least comprises pathogenic characteristics, disease stages and clinical types of typhoid and paratyphoid; classifying the disease stages and the clinical types according to the pathogenic characteristics by using a kNN algorithm; training an association rule model of typhoid and paratyphoid based on the classification results by using an Apr i or i algorithm; and inputting to-be-identified case information into the association rule model to obtain a typhoid or paratyphoid report. The application solves the technical problems of high screening difficulty and low screening accuracy caused by the fact that the pathogenesis and clinical symptoms of paratyphoid and typhoid are basically similar.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of disease screening, in particular, to a typhoid and paratyphoid screening method, device, storage medium and electronic device. BACKGROUND

[0002] The inventors have found that typhoid and paratyphoid are a group of infectious diseases with similar clinical manifestations, which are caused by Salmonella typhi and Salmonella paratyphi invading the human body through the digestive tract. Since the pathogenesis and clinical symptoms of paratyphoid and typhoid are basically similar, it is difficult to screen them, especially to distinguish them.

[0003] In view of the problem that the pathogenesis and clinical symptoms of paratyphoid and typhoid are basically similar, causing difficulty in screening, and low screening accuracy, an effective solution has not yet been proposed. SUMMARY

[0004] The main purpose of the present application is to provide a typhoid and paratyphoid screening method, device, storage medium and electronic device to solve the problem that the pathogenesis and clinical symptoms of paratyphoid and typhoid are basically similar, causing difficulty in screening, and low screening accuracy.

[0005] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a typhoid and paratyphoid screening method is provided.

[0006] The typhoid and paratyphoid screening method according to the present application comprises: obtaining case information of typhoid and paratyphoid; wherein the case information at least includes pathogenic characteristics, stages of disease and clinical types of typhoid and paratyphoid; classifying the stages of disease and clinical types according to the pathogenic characteristics by using a kNN algorithm; training an association rule model of typhoid and paratyphoid based on the classification results using an Apriori algorithm; inputting the case information to be identified into the association rule model to obtain a typhoid or paratyphoid report.

[0007] Further, after obtaining the case information of typhoid and paratyphoid, before classifying the stages of disease and clinical types according to the pathogenic characteristics by using the kNN algorithm, it further comprises: performing a predetermined normalization processing operation on the case information to obtain normalized pathogenic characteristics.

[0008] Further, classifying the stages of disease and clinical types according to the pathogenic characteristics by using the kNN algorithm comprises: calculating the distance between the stages of disease, clinical types and pathogenic characteristics of each sample by using a Euclidean distance model or a Manhattan distance model; if the majority of the k most similar samples of a sample in the feature space belong to a certain pathogenic characteristic, then the sample also belongs to this pathogenic characteristic.

[0009] Further, the Apriori algorithm is used to train the association rule model of typhoid and paratyphoid based on the classification results, including: scanning the classification results as candidate frequent 1-item sets, calculating the support 1 of the candidate frequent 1-item sets, pruning the candidate frequent 1-item sets lower than the support 1 to obtain frequent 1-item sets; connecting the frequent 1-item sets to obtain candidate frequent 2-item sets, calculating the support 2 of the candidate frequent 2-item sets, screening out the candidate frequent 2-item sets lower than the support 2 to obtain frequent 2-item sets; and the like, iteratively until no frequent k+1-item set can be found, and the set of the corresponding frequent k-item sets is the output result.

[0010] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a screening device for typhoid and paratyphoid is provided.

[0011] The screening device for typhoid and paratyphoid according to the present application comprises: a data acquisition module for acquiring case information of typhoid and paratyphoid; wherein the case information at least includes pathogenic characteristics, disease stages and clinical types of typhoid and paratyphoid; a feature classification module for classifying the disease stages and clinical types according to the pathogenic characteristics by using a kNN algorithm; a model training module for training an association rule model of typhoid and paratyphoid based on the classification results by using an Apriori algorithm; and an association rule module for inputting to-be-identified case information into the association rule model to obtain a typhoid or paratyphoid report.

[0012] Further, it further comprises: a normalization module for performing a preset normalization processing operation on the case information to obtain normalized pathogenic characteristics.

[0013] Further, the feature classification module comprises: calculating the distance between the disease stages, clinical types and pathogenic characteristics of each sample by using a Euclidean distance model or a Manhattan distance model; and if most of the k most similar samples of a sample in the feature space belong to a certain pathogenic characteristic, the sample also belongs to the pathogenic characteristic.

[0014] Further, the association rule module comprises: scanning the classification results as candidate frequent 1-item sets, calculating the support 1 of the candidate frequent 1-item sets, pruning the candidate frequent 1-item sets lower than the support 1 to obtain frequent 1-item sets; connecting the frequent 1-item sets to obtain candidate frequent 2-item sets, calculating the support 2 of the candidate frequent 2-item sets, screening out the candidate frequent 2-item sets lower than the support 2 to obtain frequent 2-item sets; and the like, iteratively until no frequent k+1-item set can be found, and the set of the corresponding frequent k-item sets is the output result.

[0015] In order to achieve the above-mentioned purpose, according to another aspect of the present application, a computer readable storage medium is provided.

[0016] The computer readable storage medium according to the present application stores a computer program, wherein the computer program is configured to execute the screening method for typhoid fever and paratyphoid fever when running.

[0017] In order to achieve the above-mentioned purpose, according to another aspect of the present application, an electronic device is provided.

[0018] The electronic device according to the present application comprises a memory and a processor, wherein the memory stores a computer program, and the processor is configured to execute the computer program to execute the screening method for typhoid fever and paratyphoid fever.

[0019] In the embodiment of the present application, the typhoid fever and paratyphoid fever are screened by obtaining case information of typhoid fever and paratyphoid fever, wherein the case information at least includes pathogenic characteristics, disease stage and clinical type of typhoid fever and paratyphoid fever; the disease stage and clinical type are classified according to the pathogenic characteristics by using the kNN algorithm; the association rule model of typhoid fever and paratyphoid fever is trained based on the classification results by using the Apriori algorithm; the case information to be identified is input into the association rule model to obtain a typhoid fever or paratyphoid fever report; the purpose of accurately screening typhoid fever or paratyphoid fever is achieved, thereby realizing the technical effects of reducing the screening difficulty and improving the screening accuracy, and further solving the technical problems of relatively large screening difficulty and low screening accuracy caused by the fact that the pathogenesis and clinical symptoms of paratyphoid fever and typhoid fever are basically similar. BRIEF DESCRIPTION OF DRAWINGS

[0020] The accompanying drawings, which form a part of this application, are intended to provide further understanding of the application and are incorporated herein for a purpose of explanations and are not intended as an undue limitation on the application. In the drawings:

[0021] Figure 1 is the flow of the screening method for typhoid fever and paratyphoid fever according to the embodiment of the present application;

[0022] Figure 2 is a structural schematic diagram of the screening device for typhoid fever and paratyphoid fever according to the embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to make the person skilled in the art better understand the present application, the technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings of the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work should belong to the scope of protection of the present application.

[0024] It should be noted that the terms "first", "second", and the like in the description and claims of the application and the above drawings are used to distinguish between similar objects and are not necessarily used to describe a particular sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein are, for example, capable of orderly or chronological execution or performance even though the terms can not be expressly used in that context. Furthermore, the terms "comprise", "comprising", "include", "including", and the like are used herein to indicate the presence of stated features, steps or elements but do not preclude the presence or addition of one or more other features, steps, elements or groups thereof. In other words, these terms are used in their open-ended, non-limiting sense.

[0025] In the present application, the terms "upper", "lower", "left", "right", "front", "back", "top", "bottom", "inner", "outer", "middle", "vertical", "horizontal", "lateral", "longitudinal", and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings. These terms are mainly used to better describe the present application and its embodiments, and are not intended to limit the indicated devices, elements or components to have a specific orientation, or to be constructed and operated in a specific orientation.

[0026] In addition, in addition to indicating the orientation or positional relationship, the above-mentioned partial terms can also be used to indicate other meanings, for example, the term "upper" can also be used to indicate a certain dependent relationship or connection relationship in some cases. For those skilled in the art, the specific meaning of these terms in the present application can be understood according to the specific circumstances.

[0027] In addition, the terms "mount", "set", "provided with", "connected", "connected", "sleeved" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or a monolithic structure; it can be a mechanical connection, or an electrical connection; it can be directly connected, or indirectly connected through an intermediate medium, or internal communication between two devices, elements or components. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0028] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0029] According to the embodiments of the present application, a screening method for typhoid and paratyphoid is provided, as shown in the figure, the method comprises the following steps S101 to S104: Figure 1

[0030] ​In step S101, case information of typhoid and paratyphoid is acquired; wherein the case information at least includes pathogenic characteristics, disease stage and clinical type of typhoid and paratyphoid.

[0031] A hospital or a third-party medical platform can be accessed to acquire data and information of a certain number of typhoid and paratyphoid cases, including epidemiological history, clinical manifestations, laboratory detection and other data and information of typhoid and paratyphoid cases.

[0032] For example, the epidemiological history of typhoid and paratyphoid cases: having been to or lived in a typhoid and paratyphoid epidemic area within 30 days before the disease; having a close contact history with typhoid and paratyphoid patients and carriers; having bad health habits such as drinking raw water, etc.; the clinical manifestations of typhoid and paratyphoid cases: unexplained persistent fever; special toxic appearance (expression dull, dull), relatively slow pulse, skin roseola, hepatosplenomegaly, etc.; laboratory detection of typhoid and paratyphoid cases: reduction or disappearance of eosinophils, normal or low total white blood cells; "O" antibody agglutination titer of Weichardt reaction ≥1:80, "H" antibody agglutination titer ≥1:160; the specific antibody titer in convalescent serum is increased by more than 4 times compared with the specific antibody titer in acute phase serum; isolation of Salmonella typhi or Salmonella paratyphi from any one of blood, bone marrow, feces, bile, etc.

[0033] According to the embodiment of the application, preferably, after acquiring the case information of typhoid and paratyphoid, the disease stage and the clinical type are classified according to the pathogenic characteristics by using the kNN algorithm, and the method further comprises the following steps:

[0034] Performing a preset normalization processing operation on the case information to obtain normalized pathogenic characteristics.

[0035] The multi-source data and information of the acquired typhoid and paratyphoid cases are normalized to obtain a feature information set of typhoid and paratyphoid cases, respectively.

[0036] For example, the common feature information of typhoid and paratyphoid includes: persistent fever, nervous system poisoning symptoms and digestive symptoms, relatively slow pulse, roseola, hepatosplenomegaly, leukopenia, eosinopenia or disappearance, etc. Among them, the clinical symptoms of typhoid include: the incubation period is generally 7-14 days. Most of them have a slow onset, the body temperature gradually rises to 39-40℃, and lasts for 5-7 days. In addition to fever, the child is listless, has poor appetite, expressionless, headache, general malaise, abdominal distension, constipation or diarrhea, hepatosplenomegaly, and relatively slow pulse, that is, the pulse beats are not in proportion to the rise in body temperature. Reddish skin rash, called roseola, can be seen on the abdomen and chest, but it is rare. About half of the children have cough, pharyngeal congestion and other respiratory tract infection symptoms; the general clinical symptoms of paratyphoid include: there may be an incubation period of 8-10 days before the onset, and a few may be 3-16 days. After the onset, the main manifestations are persistent high fever, loss of appetite, abdominal distension, diarrhea, chest tightness, expressionless, abdominal tenderness, and rash.

[0037] In addition, the main symptoms of paratyphoid are: (1) persistent high fever, initial body temperature gradually rises to above 39℃, and fever persists; (2) loss of appetite, abdominal distension, constipation and diarrhea; (3) relatively slow pulse, but with myocardial damage, often with palpitations, chest tightness, arrhythmia; (4) expressionless, slow reaction, delirium coma in severe cases; (5) signs: body temperature 39℃-40℃, spirit trance, right lower quadrant tenderness, hepatomegaly, slow pulse, skin roseola, and severe cases may have complications such as intestinal bleeding and intestinal perforation.

[0038] In this way, data security can be provided for the classification of pathogenic characteristics.

[0039] Step S102, classifying the disease stage and clinical type according to the pathogenic characteristics by using the kNN algorithm;

[0040] Specifically, classifying the disease stage and clinical type according to the pathogenic characteristics by using the kNN algorithm includes:

[0041] Using the Euclidean distance model or Manhattan distance model to calculate the distance between the disease stage, clinical type and pathogenic characteristics of each sample;

[0042] If the majority of the k most similar samples of a sample in the feature space belong to a certain pathogenic characteristic, then the sample also belongs to this pathogenic characteristic.

[0043] KNN is to classify by measuring the distance between different feature values. The idea is: if a sample in the k most similar (i.e. the nearest neighbor in the feature space) of the sample belongs to a certain class, the majority of the sample belongs to this class. K is usually an integer not greater than 20. In the KNN algorithm, the selected neighbors are all correctly classified objects. This method only depends on the class of the nearest neighbor or several samples in the decision-making process to determine the class of the sample to be classified. In KNN, the distance between objects is used as a measure of dissimilarity between objects, avoiding the matching problem between objects. Here, the distance is usually the Euclidean distance or Manhattan distance:

[0044] Euclidean distance: Manhattan distance:

[0045] For example, typhoid fever can be classified into 7 types by classification: (1) mild: moderate fever, mild systemic symptoms, short course (1-3 weeks), lack of typical manifestations of the disease; (2) fulminant (severe): acute onset, severe toxemia, rapid disease progression. High fever or super high fever, chills, abdominal pain, diarrhea, can be complicated by shock, toxic disease (encephalopathy, myocarditis, hepatitis, intestinal paralysis) or DIC; (3) delayed: initial onset of typical clinical manifestations, but fever persists, fever type is remittent or intermittent fever, fever for several months, liver and spleen are significantly enlarged, mostly seen in patients with typhoid fever with chronic blood schistosomiasis; (4) free type: mild symptoms, normal work and life, not easy to detect, can suddenly develop intestinal bleeding and intestinal perforation; (5) abrupt type: acute onset, typical manifestations, fever for 1 week, symptoms rapidly relieved; (6) pediatric typhoid fever: atypical manifestations, but acute onset, severe disease. Infants and young children often have digestive symptoms, irregular high fever, convulsions, roseola is rare, and white blood cell count is increased, often complicated by bronchitis or pneumonia. School-age children are mostly mild or abrupt type, with mild symptoms, short course, and relatively slow pulse, white blood cell count is not decreased, and complications are rare; (7) elderly typhoid fever: atypical symptoms, weakness, persistent gastrointestinal dysfunction, memory loss, easy to develop bronchial pneumonia and heart failure. Long course, high mortality and other;

[0046] In addition, the symptoms caused by different types of paratyphoid salmonella are different. Therefore, paratyphoid can be divided into paratyphoid A, paratyphoid B and paratyphoid C. Their symptoms are also different. Among them, 1) paratyphoid A and paratyphoid B symptoms include: (1) body temperature: fluctuation is large, heat course is short, 2-3 weeks; (2) digestive system symptoms: obvious loss of appetite, thick and greasy tongue fur, abdominal discomfort, abdominal distension, constipation or diarrhea, light pressure pain in lower abdomen; (3) cardiovascular system symptoms: relative slow pulse and heavy pulse; (4) nervous system symptoms: expressionless, slow reaction, hearing loss, severe patients can have delirium, coma or meningeal irritation (virtual meningitis); (5) hepatosplenomegaly: most patients have splenomegaly, soft and tender. Part of the liver enlargement, complicated with toxic hepatitis, can appear liver function abnormalities or jaundice; (6) roseola: rash appears early. Complications such as intestinal bleeding and intestinal perforation are rare, and the mortality rate is low; 2) paratyphoid C has more complex clinical manifestations. Can be divided into: (1) paratyphoid type: similar to typhoid fever and paratyphoid A, paratyphoid B; (2) acute gastroenteritis type: mainly gastroenteritis symptoms, short course; (3) sepsis type: often seen in weak children. The onset is acute, with chills and high fever, more than half of the patients can have migratory suppurative complications. Intestinal bleeding, intestinal perforation is rare, etc.

[0047] The intelligent classification of the disease stage and clinical type according to the pathogenic characteristics is realized, manual labeling is not needed, and data support is provided for the training of the model.

[0048] In step S103, an association rule model of typhoid and paratyphoid is trained based on the classification result using the Apriori algorithm.

[0049] The association rule model of typhoid and paratyphoid cases is established by using the Apriori algorithm, and based on the above-mentioned comprehensive epidemiological data, clinical data and laboratory examination result data, small sample cognitive classification training is carried out to obtain the association rule model.

[0050] Specifically, the association rule model of typhoid and paratyphoid is trained based on the classification result using the Apriori algorithm, which includes:

[0051] Scan the classification result as a candidate frequent 1-item set, calculate the support 1 of the candidate frequent 1-item set, prune the candidate frequent 1-item set lower than the support 1, and obtain the frequent 1-item set;

[0052] Connect the frequent 1-item set to obtain a candidate frequent 2-item set, calculate the support 2 of the candidate frequent 2-item set, filter out the candidate frequent 2-item set lower than the support 2, and obtain the frequent 2-item set;

[0053] Iterate down in this way until no frequent k+1-item set can be found, and the set of corresponding frequent k-item sets is the output result.

[0054] Apriori algorithm is a commonly used algorithm for mining data association rules, it is used to find the frequently occurring data sets in the data values, find the pattern of these sets to help us make some decisions.

[0055] Specifically, the Apriori algorithm uses an iterative method, first search out candidate 1 set and the corresponding support, pruning to remove the 1 set below the support, get frequent 1 set. Then the remaining frequent 1 set is connected to get the candidate frequent 2 set, screening out the candidate frequent 2 set below the support, get the real frequent two sets, and so on, iteration, until no frequent k+1 set can be found, the corresponding frequent k set is the output result of the algorithm.

[0056] Apriori algorithm formula is:

[0057]

[0058]

[0059] Here, the commonly used evaluation criteria of frequent item set are support, confidence and lift.

[0060] (1) support is the number of several associated data in the data set, or the probability of the occurrence of several data association. If we have two data X and Y to analyze the association, the corresponding support is:

[0061]

[0062] Generally speaking, the data with high support does not necessarily constitute a frequent item set, but the data with too low support certainly does not constitute a frequent item set.

[0063] (2) confidence reflects the probability of the occurrence of another data after the occurrence of one data, or the conditional probability of the data. If we have two data X and Y to analyze the association, the confidence of X to Y is

[0064] (3) lift represents the probability of containing X under the condition of containing Y, and the ratio of the probability of the occurrence of X to the total probability of X, that is:

[0065] Lift reflects the association between X and Y, and lift greater than 1 is effective strong association rule, and lift less than or equal to 1 is invalid strong association rule. A special case, if X and Y are independent, then because P(X|Y) = P(X) at this time.

[0066] Generally, to select a frequent data set in a data set, a self-defined evaluation criterion is needed. The most commonly used evaluation criterion is a self-defined support, or a combination of a self-defined support and a confidence.

[0067] In a specific embodiment, the Apriori algorithm procedure is as follows:

[0068] Scan the entire data set to obtain all the data that appears as a candidate frequent 1-item set. k = 1, and the frequent 0-item set is an empty set.

[0069] Mine the frequent k-item set.

[0070] a) Scan the data to calculate the support of the candidate frequent k-item set.

[0071] b) Remove the data set with a support lower than the threshold value in the candidate frequent k-item set to obtain the frequent k-item set. If the obtained frequent k-item set is empty, directly return the set of frequent k-1-item sets as the algorithm result, and the algorithm ends. If the obtained frequent k-item set has only one item, directly return the set of frequent k-item sets as the algorithm result, and the algorithm ends.

[0072] c) Based on the frequent k-item set, connect to generate a candidate frequent k+1-item set.

[0073] Let k = k + 1, and go back to step 2.

[0074] Step S104, input the case information to be identified into the association rule model to obtain a typhoid or paratyphoid report.

[0075] According to the Apriori algorithm and the model, the typhoid and paratyphoid case information and data provided by the target patient are associated and classified to generate a cognitive report of the corresponding typhoid or paratyphoid case, and a suspected diagnosis, or a clinical diagnosis, or a definite diagnosis conclusion is made therein.

[0076] For example: 1) Suspected case: a carrier without any clinical manifestations, a typhoid or paratyphoid Salmonella isolated from feces, unexplained persistent fever, or a history of visiting or living in a typhoid or paratyphoid epidemic area within 30 days before the disease; or a close contact history with a typhoid or paratyphoid patient or carrier; or a bad health habit such as drinking raw water. If there is a typhoid Salmonella, it is a suspected typhoid case; if there is a paratyphoid Salmonella, it is a suspected paratyphoid case;

[0077] 2) Clinical diagnosis case: the carrier has no any clinical manifestation, and the Salmonella typhi or Salmonella paratyphi is isolated from the feces, or the fever of unknown origin lasts for a long time; or the special toxic face (expression is indifferent, dull), relative slow pulse, skin rose rash, hepatosplenomegaly; and the eosinophilic granulocyte is reduced or disappears, and the total white blood cell count is normal or low. If the Salmonella typhi exists, it is a clinical diagnosis case of typhoid fever; if the Salmonella paratyphi exists, it is a clinical diagnosis case of paratyphoid fever;

[0078] 3) Confirmed case: the carrier has no any clinical manifestation, and the Salmonella typhi or Salmonella paratyphi is isolated from the feces, the fever of unknown origin lasts for a long time, or the specific antibody titer in the convalescent serum is increased by more than 4 times compared with the specific antibody titer in the acute phase serum; or the Salmonella typhi or Salmonella paratyphi is isolated from any one of the blood, bone marrow, feces and bile. If the Salmonella typhi exists, it is a confirmed case of typhoid fever; if the Salmonella paratyphi exists, it is a confirmed case of paratyphoid fever.

[0079] From the above description, it can be seen that the present application achieves the following technical effects:

[0080] In the embodiment of the present application, the typhoid fever and paratyphoid fever are screened, and the case information of the typhoid fever and paratyphoid fever is acquired; wherein the case information at least includes the pathogenic characteristics, the disease stage and the clinical type of the typhoid fever and paratyphoid fever; the disease stage and the clinical type are classified according to the pathogenic characteristics by using the kNN algorithm; the association rule model of the typhoid fever and paratyphoid fever is trained based on the classification result by using the Apriori algorithm; the case information to be identified is input into the association rule model, and the typhoid fever or paratyphoid fever report is obtained; the purpose of accurately screening the typhoid fever or paratyphoid fever is achieved, thereby realizing the technical effects of reducing the screening difficulty and improving the screening accuracy, and further solving the technical problems of relatively large screening difficulty and low screening accuracy caused by the similar pathogenesis and clinical symptoms of the paratyphoid fever and typhoid fever.

[0081] It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a group of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in an order different from that shown herein.

[0082] According to the embodiment of the present application, a device for implementing the screening method of the typhoid fever and paratyphoid fever is also provided, as shown in the figure, the device comprises: Figure 2

[0083] The data acquisition module 10 is configured to acquire the case information of the typhoid fever and paratyphoid fever; wherein the case information at least includes the pathogenic characteristics, the disease stage and the clinical type of the typhoid fever and paratyphoid fever.

[0084] ​The hospital or third-party medical platform can be accessed to obtain a certain amount of data and information of typhoid and paratyphoid cases, including epidemiological history, clinical manifestations, laboratory tests, and other data and information of typhoid and paratyphoid cases.

[0085] For example, the epidemiological history of typhoid and paratyphoid cases: having been to or lived in a typhoid and paratyphoid epidemic area within 30 days before the disease; having a close contact history with typhoid and paratyphoid patients and carriers; having bad health habits such as drinking raw water, etc.; the clinical manifestations of typhoid and paratyphoid cases: unexplained persistent fever; special toxic appearance (expressionless, dull), relatively slow pulse, skin roseola, hepatosplenomegaly, etc.; laboratory tests of typhoid and paratyphoid cases: reduction or disappearance of eosinophils, normal or low total white blood cell count; "O" agglutination titer of Weichardt reaction ≥1:80, "H" agglutination titer ≥1:160; the specific antibody titer in convalescent serum is increased by more than 4 times compared with the specific antibody titer in acute phase serum; isolation of Salmonella typhi or Salmonella paratyphi from any one of blood, bone marrow, feces, and bile.

[0086] It also includes a normalization module for performing a predetermined normalization processing operation on the case information to obtain normalized pathogenic characteristics.

[0087] The normalized processing is performed on the obtained multi-source data and information of typhoid and paratyphoid cases to obtain characteristic information sets of typhoid and paratyphoid cases, respectively.

[0088] For example, the common characteristic information of typhoid and paratyphoid includes: persistent fever, nervous system poisoning symptoms and digestive symptoms, relatively slow pulse, roseola, hepatosplenomegaly, leukopenia, and reduction or disappearance of eosinophils, etc. Among them, the clinical symptoms of typhoid include: the incubation period is generally 7-14 days. Most of them have a slow onset, the body temperature gradually rises to 39-40℃, and lasts without decline. In addition to fever, the child is listless, has poor appetite, expressionless, headache, general malaise, abdominal distension, constipation or diarrhea, hepatosplenomegaly, and relatively slow pulse, i.e. the pulse beats are not in proportion to the rise in body temperature. Reddish skin rash, called roseola, can be seen on the abdomen and chest, but it is rare. About half of the children have cough, pharyngeal congestion, and other respiratory infection symptoms; the general clinical symptoms of paratyphoid include: an incubation period of 8-10 days before the onset, and a few of 3-16 days. After the onset, the main manifestations are persistent high fever, loss of appetite, abdominal distension, diarrhea, chest tightness, expressionless, abdominal tenderness, and skin rash.

[0089] In addition, the main symptoms of paratyphoid fever are: (1) persistent high fever, initial body temperature gradually rises to 39℃ or above, fever lingers and does not subside; (2) loss of appetite, abdominal distension, constipation and diarrhea; (3) relatively slow pulse, but often have palpitations, chest tightness, arrhythmia when myocardial damage; (4) expression is dull, reaction is slow, and the heavy one is delirious coma; (5) signs: body temperature 39℃-40℃, spirit is in a trance, right lower quadrant tenderness, hepatomegaly, slow pulse, skin roseola, and the heavy one can have complications of intestinal bleeding, intestinal perforation.

[0090] In this way, data security can be provided for the classification of pathogenic characteristics.

[0091] The feature classification module 20 is configured to classify the disease stages and clinical types according to pathogenic characteristics by using a kNN algorithm.

[0092] Specifically, the feature classification module 20 includes:

[0093] The Euclidean distance model or Manhattan distance model is used to calculate the distance between each sample, the disease stage, the clinical type and the pathogenic characteristic.

[0094] If the majority of the k most similar samples of a sample in the feature space belong to a certain pathogenic characteristic, then the sample also belongs to the pathogenic characteristic.

[0095] KNN is a classification method by measuring the distance between different feature values. The idea is: if the majority of the k most similar (i.e. the most adjacent in the feature space) samples of a sample in the feature space belong to a certain class, then the sample also belongs to this class. K is usually an integer not greater than 20. In the KNN algorithm, the selected neighbors are all correctly classified objects. This method only relies on the class of the nearest neighbor or several samples to determine the class to which the sample belongs. In KNN, the distance between objects is used as a non-similarity index between objects, avoiding the matching problem between objects. Here, the distance generally uses Euclidean distance or Manhattan distance:

[0096] Euclidean distance: Manhattan distance:

[0097] For example, the clinical types of typhoid fever can be classified into 7 types: (1) mild type: moderate fever, mild systemic symptoms, short course (1-3 weeks), lack of typical manifestations of the disease; (2) fulminant type (severe type): acute onset, severe toxemia, rapid disease progression. High fever or super high fever, aversion to cold, abdominal pain, diarrhea, and concurrent shock, toxic diseases (encephalopathy, myocarditis, hepatitis, intestinal paralysis) or DIC; (3) prolonged type: initial onset of typical clinical manifestations, but fever persists, fever type is remittent or intermittent fever, fever lasts for several months, liver and spleen are significantly enlarged, and is more common in patients with chronic blood fluke disease; (4) free type: mild symptoms, normal work and life, not easy to detect, sudden intestinal bleeding and intestinal perforation; (5) abrupt type: acute onset, typical symptoms, symptoms rapidly relieved after 1 week of fever; (6) pediatric typhoid fever: atypical manifestations, but acute onset and severe disease. Infants and young children often have digestive symptoms, irregular high fever, convulsions, and roseola are rare, and the white blood cell count is increased, often complicated by bronchitis or pneumonia. School-age children are mostly mild or abrupt types, with mild symptoms, short course, and relatively few tachycardia, white blood cell count is not decreased, and complications are rare; (7) elderly typhoid fever: atypical symptoms, weakness, persistent gastrointestinal dysfunction, memory loss, and easy to develop bronchitis and heart failure. Long course, high mortality and other symptoms;

[0098] In addition, the symptoms caused by different types of paratyphoid salmonella are different. Therefore, paratyphoid fever can be divided into paratyphoid fever A, paratyphoid fever B, and paratyphoid fever C. Their symptoms are also different. Among them, 1) paratyphoid fever A and paratyphoid fever B symptoms include: (1) body temperature: large fluctuations, short fever course, 2-3 weeks; (2) digestive system symptoms: obvious loss of appetite, thick and greasy tongue fur, abdominal discomfort, abdominal distension, constipation or diarrhea, and light abdominal tenderness; (3) cardiovascular system symptoms: relative bradycardia and tachycardia; (4) nervous system symptoms: may appear expressionless, slow reaction, hearing loss, severe patients may have delirium, coma or meningeal irritation (virtual meningitis); (5) hepatosplenomegaly: most patients have splenomegaly, soft and tender. Part of the liver is enlarged, and when toxic hepatitis is complicated, liver function abnormalities or jaundice may occur; (6) roseola: rash appears early. Complications such as intestinal bleeding and intestinal perforation are rare, and the mortality rate is low; 2) paratyphoid fever C has more complex clinical manifestations. Can be divided into: (1) paratyphoid type: similar to typhoid fever and paratyphoid fever A, paratyphoid fever B; (2) acute gastroenteritis type: mainly gastroenteritis symptoms, short course; (3) sepsis type: often seen in weak children. Acute onset, chills, high fever, more than half of the patients may have migratory suppurative complications. Intestinal bleeding, intestinal perforation is rare, etc.

[0099] Intelligent classification of disease stages and clinical types according to pathogenic characteristics is realized, manual labeling is not required, and data support is provided for model training.

[0100] The model training module 30 is configured to train an association rule model of typhoid and paratyphoid based on the classification result by using the Apriori algorithm.

[0101] The association rule model of typhoid and paratyphoid cases is established by using the Apriori algorithm, and based on the above-mentioned comprehensive epidemiological data, clinical data and laboratory examination result data, small sample cognitive classification training is performed to obtain the association rule model.

[0102] Specifically, the association rule module 30 comprises:

[0103] The scanning classification result is taken as a candidate frequent 1-item set, the support degree 1 of the candidate frequent 1-item set is calculated, the candidate frequent 1-item set lower than the support degree 1 is pruned to obtain a frequent 1-item set;

[0104] The frequent 1-item set is connected to obtain a candidate frequent 2-item set, the support degree 2 of the candidate frequent 2-item set is calculated, and the candidate frequent 2-item set lower than the support degree 2 is screened to obtain a frequent 2-item set;

[0105] In this way, iteration is performed until a frequent k+1-item set cannot be found, and a set of corresponding frequent k-item sets is the output result.

[0106] The Apriori algorithm is a commonly used algorithm for mining data association rules, which is used to find frequently occurring data sets in data values. Finding the patterns of these sets helps us make decisions.

[0107] Specifically, the Apriori algorithm uses an iterative method to search for candidate 1-item sets and corresponding support degrees, prune 1-item sets lower than the support degrees to obtain frequent 1-item sets. Then, the remaining frequent 1-item sets are connected to obtain candidate frequent 2-item sets, and the candidate frequent 2-item sets lower than the support degrees are screened to obtain true frequent 2-item sets. In this way, iteration is performed until a frequent k+1-item set cannot be found, and a set of corresponding frequent k-item sets is the output result of the algorithm.

[0108] The Apriori algorithm formula is:

[0109]

[0110]

[0111] Here, the commonly used evaluation criteria for frequent item sets are support, confidence and lift.

[0112] (1) The support is the proportion of the number of several associated data in the data set to the total data set. Or the probability of the occurrence of several data association. If we have two data X and Y that we want to analyze the association, the corresponding support is:

[0113]

[0114] Generally speaking, data with high support does not necessarily constitute a frequent itemset, but data with very low support certainly does not constitute a frequent itemset.

[0115] (2) Confidence level reflects the probability that one data point will appear after the other data point has appeared, or the conditional probability of the data. If we have two data points X and Y that we want to analyze the correlation between, and the confidence level of X with respect to Y is...

[0116] (3) Lift represents the ratio of the probability of having Y and X simultaneously to the overall probability of X occurring, i.e.:

[0117] Lift reflects the relationship between X and Y; a lift greater than 1 indicates a positive correlation. It is a valid strong association rule, and its lift is less than or equal to 1. This is an invalid strong association rule. A special case is when X and Y are independent, then... Because at this point P(X|Y) = P(X).

[0118] Generally, to select a frequent dataset from a collection, a custom evaluation criterion is needed. The most common evaluation criterion is a custom support score, or a combination of a custom support score and a confidence score.

[0119] In one specific implementation, the April algorithm flow is as follows:

[0120] Scan the entire dataset to obtain all occurrences of the data, which will be used as candidate frequent 1-itemsets. k=1, and the frequent 0-itemsets are empty.

[0121] Mining frequent k-itemsets;

[0122] a) Calculate the support of candidate frequent k-itemsets from the scanned data;

[0123] b) Remove datasets with support below a threshold from the candidate frequent k-itemsets to obtain frequent k-itemsets. If the obtained frequent k-itemsets are empty, return the set of frequent k-1 itemsets as the algorithm result, and the algorithm terminates. If the obtained frequent k-itemsets contain only one item, return the set of frequent k-itemsets as the algorithm result, and the algorithm terminates.

[0124] c) Generate candidate frequent k+1 itemsets by connecting frequent k-itemsets.

[0125] Let k = k + 1, then proceed to step 2.

[0126] The association rule module 40 is used for inputting the case information to be identified into an association rule model to obtain a typhoid or paratyphoid report.

[0127] According to the Apriori algorithm and the model, the typhoid and paratyphoid case information and data provided by the target patient are associated and classified, a cognitive report of the corresponding typhoid or paratyphoid case is generated, and a suspected diagnosis, or a clinical diagnosis, or a definite diagnosis is made in the report.

[0128] For example: 1) suspected case: a carrier without any clinical manifestations, Salmonella typhi or Salmonella paratyphi is isolated from feces, unexplained persistent fever occurs, or the patient has been to or lived in a typhoid or paratyphoid epidemic area within 30 days before the disease; or has a close contact history with a typhoid or paratyphoid patient or carrier; or has bad health habits such as drinking raw water. If there is Salmonella typhi, it is a suspected typhoid case; if there is Salmonella paratyphi, it is a suspected paratyphoid case;

[0129] 2) clinical diagnosis case: a carrier without any clinical manifestations, Salmonella typhi or Salmonella paratyphi is isolated from feces, unexplained persistent fever occurs; or special toxic appearance (expression is dull, dull), relatively slow pulse, skin roseola, hepatosplenomegaly; and hypereosinophilia or disappearance of eosinophils, normal or low total white blood cell count. If there is Salmonella typhi, it is a typhoid clinical diagnosis case; if there is Salmonella paratyphi, it is a paratyphoid clinical diagnosis case;

[0130] 3) confirmed case: a carrier without any clinical manifestations, Salmonella typhi or Salmonella paratyphi is isolated from feces, unexplained persistent fever occurs, or the specific antibody titer in the convalescent serum is increased by more than 4 times compared with the specific antibody titer in the acute phase serum; or Salmonella typhi or Salmonella paratyphi is isolated from any one of blood, bone marrow, feces, and bile. If there is Salmonella typhi, it is a typhoid confirmed case; if there is Salmonella paratyphi, it is a paratyphoid confirmed case.

[0131] From the above description, it can be seen that the present application achieves the following technical effects:

[0132] In the embodiment of the present application, typhoid and paratyphoid are screened in a manner that case information of typhoid and paratyphoid is acquired; wherein the case information at least includes pathogenic characteristics, disease stage and clinical type of typhoid and paratyphoid; the disease stage and clinical type are classified according to the pathogenic characteristics by using a kNN algorithm; an association rule model of typhoid and paratyphoid is trained based on the classification result by using an Apriori algorithm; the case information to be identified is input into the association rule model to obtain a typhoid or paratyphoid report; the purpose of accurately screening typhoid or paratyphoid is achieved, thereby realizing the technical effects of reducing screening difficulty and improving screening accuracy, and further solving the technical problems of great screening difficulty and low screening accuracy caused by the fact that the pathogenesis and clinical symptoms of paratyphoid and typhoid are basically similar.

[0133] Obviously, those skilled in the art should understand that the modules or steps of the present application described above can be realized by general computing devices, which can be concentrated on a single computing device or distributed on a network composed of multiple computing devices, and alternatively, they can be realized by program codes executable by computing devices, so that they can be stored in storage devices and executed by computing devices, or they can be respectively manufactured into individual integrated circuit modules, or multiple modules or steps among them can be manufactured into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

[0134] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A screening method for typhoid and paratyphoid fever, characterized in that, include: Obtain case information on typhoid and paratyphoid fever; wherein the case information includes at least: the pathogenic characteristics, stage of disease, and clinical type of typhoid and paratyphoid fever; A preset normalization process is performed on the case information to obtain normalized pathogenic characteristics; The kNN algorithm is used to classify disease stages and clinical types according to pathogenic features, including: using Euclidean distance model or Manhattan distance model to calculate the distance between the disease stage, clinical type and pathogenic feature of each sample; if most of the k most similar samples of a sample in the feature space belong to a certain pathogenic feature, then the sample also belongs to that pathogenic feature. The Apriori algorithm is used to train association rule models for typhoid and paratyphoid fever based on classification results. The process includes: scanning the classification results as candidate frequent 1-itemsets, calculating the support of 1 for each candidate frequent 1-itemets, pruning and removing candidate frequent 1-itemsets with support below 1 to obtain frequent 1-itemsets; joining the frequent 1-itemsets to obtain candidate frequent 2-itemsets, calculating the support of 2 for each candidate frequent 2-itemsets, filtering and removing candidate frequent 2-itemsets with support below 2 to obtain frequent 2-itemsets; and so on, iterating until no more frequent k+1-itemsets can be found. The set of the corresponding frequent k-itemsets is the output result. The information of the cases to be identified is input into the association rule model to obtain a typhoid or paratyphoid report. The screening report includes one of the following: suspected diagnosis, clinical diagnosis, or confirmed diagnosis.

2. A screening device for typhoid and paratyphoid fever, characterized in that, include: The data acquisition module is used to acquire case information of typhoid fever and paratyphoid fever; wherein the case information includes at least: the pathogenic characteristics, disease stage and clinical type of typhoid fever and paratyphoid fever; The normalization module is used to perform a preset normalization process on the case information to obtain normalized pathogenic characteristics; The feature classification module is used to classify disease stage and clinical type according to pathogenic features using the kNN algorithm. This includes: calculating the distance between the disease stage, clinical type and pathogenic feature of each sample using the Euclidean distance model or the Manhattan distance model; if most of the k most similar samples of a sample in the feature space belong to a certain pathogenic feature, then the sample also belongs to that pathogenic feature. The model training module is used to train association rule models for typhoid and paratyphoid fever based on classification results using the Apriori algorithm. This includes: scanning the classification results as candidate frequent 1-itemsets, calculating the support of 1 for each candidate frequent 1-itemets, pruning and removing candidate frequent 1-itemsets with support below 1 to obtain frequent 1-itemsets; connecting the frequent 1-itemsets to obtain candidate frequent 2-itemsets, calculating the support of 2 for each candidate frequent 2-itemsets, filtering and removing candidate frequent 2-itemsets with support below 2 to obtain frequent 2-itemsets; and so on, iterating until no more frequent k+1-itemsets can be found. The set of the corresponding frequent k-itemsets is the output result. The association rule module is used to input the case information to be identified into the association rule model to obtain a typhoid or paratyphoid report. The screening report includes one of the following: suspected diagnosis, clinical diagnosis, or confirmed diagnosis.

3. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the screening method for typhoid and paratyphoid fever as described in claim 1 when it is run.

4. An electronic device comprising: The memory and processor are characterized in that the memory stores a computer program, wherein the processor is configured to run the computer program to perform the screening method for typhoid and paratyphoid fever as described in claim 1.

Images