Method and system for detecting helicobacter pylori through tartar

Abstract

The invention discloses a method and system for detecting helicobacter pylori through tartar, and relates to the technical field of tartar detection.The system comprises a sample analysis module, a detection and judgment module, a data construction module and an evaluation and adjustment module; according to the technical scheme, the oral health state score, the urease activity prediction value and the ureamycin concentration are comprehensively considered, the oral health state score can visually reflect the oral health state of a corresponding patient, and data guidance can be provided for calculation of the urease activity prediction value; a comprehensive calculation formula reflecting the HP infection degree is built, and objective quantitative evaluation on the HP infection degree is achieved; in addition, according to the HP infection degree of the patient, personalized reexamination frequency can be set, and the situation of unnecessary frequent reexamination or insufficient reexamination is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of dental plaque detection, and in particular to a method and a system for detecting gastric Helicobacter pylori through dental plaque. Background Art

[0002] Tartar testing is a method of assessing oral health or detecting specific pathogens (such as Helicobacter pylori) by examining the plaque (tartar) on the surface of teeth; the following is a detailed explanation of tartar testing: Tartar, that is, the dirt on the surface of teeth, is mainly composed of food residues in the mouth, soft plaque, and bacteria attached to the surface of teeth; when teeth are unhealthy or not cleaned properly, tartar will gradually accumulate and may cause periodontal disease.

[0003] At present, the commonly used techniques for detecting gastric Helicobacter pylori through dental plaque are:

[0004] Tartar test paper can be used as a reagent for screening Helicobacter pylori infection. Judging from the color change reaction of the tartar test paper, it may use the urease detection method; urease is a metabolite produced by Helicobacter pylori, which can decompose urea and then produce ammonia and carbon dioxide; ammonia is an alkaline substance and reacts with the color developer of the test paper; the redder the color, the stronger the alkalinity of the sample, indicating that the more urease products there are, the more Helicobacter pylori are contained in the sample. This is the principle of the Helicobacter pylori urease experiment;

[0005] As for the means or technology of analyzing dental plaque, directly using test paper to test dental plaque can only determine whether the corresponding patient has Helicobacter pylori, and the accuracy of the judgment result is not enough; in addition, the commonly used methods for detecting Helicobacter pylori include: carbon-13, carbon-14 breath test, urease assay, and detection through fecal antigen, but none of the above methods are detected through dental plaque. Summary of the invention

[0006] 1. Technical issues to be resolved

[0007] In view of the deficiencies in the prior art, the present invention provides a method and system for detecting gastric Helicobacter pylori through dental plaque. By comprehensively considering various factors, a comprehensive calculation formula reflecting the degree of HP infection is established, thereby achieving an objective quantitative evaluation of the degree of HP infection. According to the patient's HP infection degree, a personalized review frequency can also be set, thereby improving the accuracy and effectiveness of the comprehensive index and solving the problems raised in the background technology.

[0008] (II) Technical solution

[0009] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0010] A system for detecting gastric Helicobacter pylori through dental plaque, comprising:

[0011] The sample analysis module collects dental plaque samples from M patients diagnosed with HP infection in any medical institution, obtains the uromycin concentration of each sample using mass spectrometry analysis technology, forms a uromycin concentration data set, and obtains the uromycin concentration range of HP infected patients after analysis and processing based on the data set, and provides verification result data;

[0012] The detection and judgment module collects dental plaque samples from the corresponding patients who go to the medical institution, and compares the uromycin concentration obtained by analysis with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within the range, an early warning signal is issued; otherwise, no response is given;

[0013] The data construction module, under the condition of receiving the early warning signal, builds an oral health status assessment model based on random forest, obtains the corresponding patient's oral parameter set as input features, and outputs the corresponding patient's oral health status score; simultaneously builds a urease activity prediction model based on linear regression, obtains the corresponding patient's oral health status score and environmental parameter set as input features, and outputs the corresponding patient's urease activity prediction value;

[0014] The evaluation and adjustment module corrects the uromycin concentration based on the standardized oral health status score, urease activity prediction value and uromycin concentration, combined with the verification result data, and builds a comprehensive calculation formula to generate a comprehensive index. The comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is built to set the review frequency of the corresponding patient according to the comparison results, and provide guidance to the patient according to the set review frequency.

[0015] Furthermore, the process of obtaining the uromycin concentration using mass spectrometry is as follows:

[0016] Sample pretreatment: The collected dental plaque samples were pretreated, including extraction and purification;

[0017] Mass spectrometry analysis: The sample is injected into a mass spectrometer for analysis. The mass spectrometer ionizes the sample molecules and separates and detects them according to their mass-to-charge ratio. The mass spectra are compared with the known database to identify the uromycin molecules in the sample and measure their concentration.

[0018] Data processing: Use mass spectrometry analysis software to process the mass spectrum, extract the peak area of ​​uromycin, and convert the peak area into uromycin concentration value according to the standard curve.

[0019] Furthermore, the process of analyzing and processing the data set of uromycin concentration is as follows:

[0020] Data cleaning: Check whether there are outliers or missing values ​​in the data set and perform data cleaning;

[0021] Statistical analysis: Calculate the mean and standard deviation of uromycin concentration and draw a distribution diagram of uromycin concentration;

[0022] Determine the threshold: Based on the statistical analysis results, select the mean value plus or minus 2 times the standard deviation as the interval;

[0023] Validation Evaluation: Using an independent validation dataset, the validation results are evaluated and calculated.

[0024] Furthermore, the independent validation dataset contained patient samples with known HP infection status, a portion of which were confirmed to be HP infected and another portion of which were uninfected, i.e., the ratio of infected patient samples to uninfected patient samples was 1:1.

[0025] Furthermore, the verification result data is calculated: after the evaluation index is calculated, the evaluation index includes sensitivity and specificity. After weighted calculation of the sensitivity and specificity, the comprehensive evaluation degree obtained is the verification result data.

[0026] Furthermore, the oral health status score, urease activity prediction value and uromycin concentration were standardized, and the standardization included converting the oral health status score, urease activity prediction value and uromycin concentration into values ​​between 0 and 1 using z-score standardization and minimum-maximum standardization methods.

[0027] Furthermore, the comprehensive calculation formula is as follows:

[0028] R=1 / (OH*(1+UEA)^(-1)*(1+UC*XZ)^(-1));

[0029] Wherein, R represents the comprehensive index, OH represents the oral health status score, UEA represents the predicted value of urease activity, UC represents the uromycin concentration, and XZ represents the verification result data.

[0030] Furthermore, the process of setting the reexamination frequency of the corresponding patient according to the comparison results is as follows:

[0031] When R < k, the patient is set to be reviewed every N years;

[0032] When k≤R<2k, the patient is set to be reviewed once every (N / 2) years;

[0033] When R ≥ 2k, the patient is set to be reviewed once every (N / 4) years;

[0034] The value of N is one year, and N is a positive integer.

[0035] A method for detecting gastric Helicobacter pylori through dental plaque comprises the following steps:

[0036] S1. In any medical institution, collect dental plaque samples from M patients who have been diagnosed with HP infection, use mass spectrometry to obtain the uromycin concentration of each sample, form a data set of uromycin concentration, analyze and process the data set to obtain the uromycin concentration range of HP infected patients, and provide verification result data;

[0037] S2. Collect dental plaque samples from the corresponding patients who go to the medical institution, and compare the uromycin concentration obtained by analysis with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within the range, an early warning signal is issued; otherwise, no response is given;

[0038] S3. Under the condition of receiving the early warning signal, build an oral health status assessment model based on random forest, use the corresponding patient's oral parameter set as input features, and output the corresponding patient's oral health status score; simultaneously build a urease activity prediction model based on linear regression, use the corresponding patient's oral health status score and environmental parameter set as input features, and output the corresponding patient's urease activity prediction value;

[0039] S4. Based on the standardized oral health status score, urease activity prediction value and uromycin concentration, the uromycin concentration is corrected in combination with the verification result data, and a comprehensive calculation formula is constructed to generate a comprehensive index. The comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is constructed. The reexamination frequency of the corresponding patient is set according to the comparison results, and guidance is provided to the patient according to the set reexamination frequency.

[0040] (III) Beneficial effects

[0041] The present invention provides a method and system for detecting gastric Helicobacter pylori through dental plaque, which has the following beneficial effects:

[0042] (1) The concentration of uromycin in dental plaque samples was detected by mass spectrometry and the diagnostic interval was determined by statistical analysis, which can more accurately identify HP-infected patients. The detailed sample collection, processing and analysis steps ensured the reliability and consistency of the data, providing a basis for subsequent research and application.

[0043] (2) This scheme comprehensively considers the oral health status score, urease activity prediction value and uromycin concentration. The oral health status score can not only intuitively reflect the oral health status of the corresponding patient, but also provide data guidance for the calculation of the urease activity prediction value, build a comprehensive calculation formula reflecting the degree of HP infection, and achieve an objective quantitative assessment of the degree of HP infection. According to the patient's HP infection degree, this scheme can also set a personalized re-examination frequency to avoid unnecessary frequent re-examination or insufficient re-examination;

[0044] (3) When conducting preliminary sample analysis, the verification result data is given while obtaining the uromycin concentration range. This can not only assist in determining the credibility of whether the corresponding patient is infected with HP, but also provide correction guidance for the subsequent calculation of the comprehensive index reflecting the degree of HP infection, further improving the accuracy and effectiveness of the comprehensive index. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A modular schematic diagram of a system for detecting gastric Helicobacter pylori through dental plaque according to the present invention;

[0046] Figure 2 The present invention is a schematic diagram of the overall process of a method for detecting gastric Helicobacter pylori through dental plaque. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in 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] Research and development concept:

[0049] At present, the commonly used techniques for detecting gastric Helicobacter pylori through dental plaque are:

[0050] Tartar test strips can be used as a reagent for screening Helicobacter pylori infection. Judging from the color change reaction of the tartar test strips, it may use the urease detection method; urease is a metabolite produced by Helicobacter pylori, which can decompose urea and then produce ammonia and carbon dioxide; ammonia is an alkaline substance, which reacts with the color developer of the test strip; the redder the color, the stronger the alkalinity of the sample, indicating that the more urease products there are, the more Helicobacter pylori are contained in the sample, which is the principle of the Helicobacter pylori urease experiment; but directly using the test strips to detect or apply tartar is not accurate enough;

[0051] Helicobacter pylori cannot colonize in oral epithelial cells, it can only colonize in gastric epithelial cells, and oral epithelial cells and gastric epithelial cells are different cells. A positive dental plaque test paper can only indicate that people's mouths contain bacteria that produce urease, but this bacterium is most likely not Helicobacter pylori. Therefore, the method of using dental plaque to detect whether a patient has Helicobacter pylori is questionable.

[0052] In order to improve its accuracy and rationality, the researchers sampled data from local people according to different regions, and conducted a series of biochemical analysis methods to draw conclusions from dental plaque about which part of the population is more likely to be infected with Helicobacter pylori. They also combined actual conditions and comprehensive indicators from all aspects to determine the severity of Helicobacter pylori infection in patients, and proposed recommended re-examination frequencies based on the severity of the infection. This not only improves the accuracy of detecting whether Helicobacter pylori is present in patients from dental plaque, but also can provide follow-up services to patients, providing further protection for their health.

[0053] Embodiment 1:

[0054] See also Figure 1 This embodiment provides a system for detecting gastric Helicobacter pylori through dental plaque, the system comprising a sample analysis module, a detection and determination module, a data construction module and an evaluation and adjustment module, each module operates in sequence, and can detect and determine whether a patient is infected with gastric Helicobacter pylori and the degree of infection through dental plaque;

[0055] Sample analysis module:

[0056] In any medical institution, dental plaque samples are collected from M patients who have been diagnosed with HP infection, and the uromycin concentration of each sample is obtained by mass spectrometry analysis technology to form a data set of uromycin concentration. After analysis and processing based on the data set, the uromycin concentration range of HP infected patients is obtained, and the verification result data is given;

[0057] When the system is used to detect dental plaque of patients in the future, it is necessary to ensure that the patient is within the corresponding range, which is: with the corresponding medical institution where the dental plaque sample is collected as the center, a circular area with a radius of S kilometers is determined as the corresponding range. The value of S is usually 30, which can be set according to specific actual needs;

[0058] The purpose of ensuring that the patient's dental plaque is within the corresponding range when testing is to:

[0059] Taking into account the regional issues, the physical constitution, climate and other factors of patients in the same area are roughly the same, so patients or people in the same area can be used as samples. The results obtained after analyzing the samples can be applied to the people in the same area, and its accuracy is guaranteed to a certain extent. Therefore, subsequent patients need to be in the same area as the corresponding medical institutions; by limiting the scope of detection and considering regional factors, the application of this program in different regions is more targeted and accurate;

[0060] It should be noted that the medical institutions referred to in this embodiment generally refer to: hospitals;

[0061] Collect dental plaque samples from M patients diagnosed with HP infection, where M ranges from 100 to 1000; Collecting dental plaque samples from patients diagnosed with HP infection is an important step in analyzing uromycin concentration;

[0062] Therefore, the unified process for collecting dental plaque samples is as follows:

[0063] S101. Ensure that the subjects have been diagnosed with HP infection; explain the purpose, process, possible risks and benefits of the experiment in detail to the subjects, and obtain their informed consent; ask the subjects about their oral health status, and exclude factors that may affect the quality of the samples, such as severe oral diseases, recent use of antibiotics or immunosuppressants;

[0064] S102. Before sampling, the subject should avoid eating, brushing teeth or using mouthwash to reduce interference from food residues and impurities in the oral cavity; prepare sterile collection tools, such as dental scrapers or samplers, and sterile EP tubes or storage tubes;

[0065] S103. Ask the subject to open his mouth to expose the tooth surface; use a sterile collection tool to gently scrape the dental plaque on the buccal and lingual sides of the teeth; be careful not to damage the gums; put the collected dental plaque sample into a sterile EP tube or storage tube, cover it to prevent contamination, and mark the basic information of the corresponding subject, including name, age and storage time;

[0066] S104. Place the sample on ice or in a refrigerated environment to maintain the freshness and stability of the sample; and send the sample to the laboratory for analysis to avoid sample degradation caused by long-term storage; if it is not possible to send it to the laboratory immediately, use an appropriate preservation solution for preservation and transport it according to the requirements of the preservation solution;

[0067] The process of obtaining uromycin concentration using mass spectrometry is as follows:

[0068] Sample pretreatment: The collected dental plaque samples are properly pretreated, such as extraction and purification steps, to remove interfering substances and enrich the target analyte (uromycin); the specific method of pretreatment may vary depending on laboratory conditions and sample characteristics, but usually includes solvent extraction, solid phase extraction, and derivatization steps;

[0069] Mass spectrometry: The processed sample is injected into a mass spectrometer for analysis; the mass spectrometer ionizes the sample molecules and separates and detects them according to their mass-to-charge ratio (m / z); by comparing the mass spectra of standards or known databases, the uromycin molecules in the sample can be identified and their concentrations can be determined;

[0070] Data processing: Use mass spectrometry analysis software to process the mass spectrum and extract the peak area or peak height information of uromycin; according to the standard curve or internal standard method, convert the peak area or peak height information into the concentration value of uromycin;

[0071] For example:

[0072] Assume that a laboratory collects dental plaque samples from a patient who has been diagnosed with HP infection, and analyzes them using mass spectrometry after pretreatment. The mass spectrum shows a peak that matches the uromycin standard. By comparing the standard curve and peak area information, the concentration of uromycin in the sample is calculated to be Xμg / mL. The laboratory records this result in a data set and associates it with other clinical information of the patient for subsequent analysis and research. The specific values ​​in the above process (such as Xμg / mL) are only examples and should be determined according to laboratory conditions and sample characteristics in actual analysis.

[0073] The process of analyzing and processing the data set is as follows:

[0074] Data cleaning: Check whether there are outliers or missing values ​​in the data set and perform data cleaning;

[0075] Statistical analysis: Calculate the mean and standard deviation of uromycin concentration, and draw a distribution diagram of uromycin concentration, such as a histogram or box plot, to observe its distribution characteristics;

[0076] Determine the threshold: Based on the statistical analysis results, select the mean plus or minus 2 times the standard deviation as the interval, so that when the uromycin concentration is detected within this interval, it can highly indicate that the patient is infected with HP;

[0077] Validation evaluation: Use an independent validation dataset (containing patient samples with known HP infection status) to evaluate the accuracy of this interval; calculate and derive validation result data;

[0078] For example:

[0079] Suppose that a data set of uromycin concentrations was obtained from 100 patients diagnosed with HP infection, and necessary cleaning and sorting were performed; the mean uromycin concentration was calculated to be 50 μg / mL, and the standard deviation was 10 μg / mL; a histogram was drawn to show that the uromycin concentration was normally distributed; the mean plus or minus 2 times the standard deviation was selected as the interval, that is, 30 μg / mL to 70 μg / mL; this interval means that when uromycin concentrations are detected within this range, there is a high probability that the patient is infected with HP; an independent validation data set (containing patient samples with known HP infection status) was used to evaluate the accuracy of this interval; the sensitivity was calculated to be 90% and the specificity was 85%, indicating that this interval has good performance in diagnosing HP infection; you can explain that when a patient's uromycin concentration is detected between 30 μg / mL and 70 μg / mL, there is a high probability that the patient is infected with HP.

[0080] Further explanation of the validation assessment is as follows:

[0081] After determining the diagnostic interval of uromycin concentration (e.g., 30 μg / mL to 70 μg / mL), in order to evaluate the accuracy of this interval, we need to use an independent validation data set for testing; this validation data set should contain patient samples with known HP infection status, that is, one part of the samples is confirmed to be HP infected, and the other part of the samples is uninfected, that is, the ratio of infected to uninfected is 1:1;

[0082] Applying diagnostic intervals: For each sample in the validation dataset, we measured its uromycin concentration and determined whether the concentration fell within the diagnostic interval; if the concentration was within the interval, we determined it to be HP infected; if the concentration was not within the interval, we determined it to be uninfected.

[0083] Calculation of evaluation indicators: Sensitivity: the proportion of patients who are truly infected with HP and are correctly judged as infected; Specificity: the proportion of patients who are truly not infected with HP and are correctly judged as not infected.

[0084] Example explanation:

[0085] Assume that there are 100 samples with known HP infection status in the validation data set, of which 50 are truly infected and 50 are uninfected; after applying our diagnostic interval, 45 truly infected samples are correctly judged as infected (true positives), and 5 are wrongly judged as uninfected (false negatives); at the same time, 42 uninfected samples are correctly judged as uninfected (true negatives), and 8 are wrongly judged as infected (false positives);

[0086] Therefore, sensitivity = true positive / (true positive + false negative) = 45 / (45+5) = 90%;

[0087] Specificity = true negative / (true negative + false positive) = 42 / (42 + 8) = 84% (Note: the numbers are adjusted here to make the example more specific, but the principle remains the same);

[0088] Interpretation of results: A sensitivity of 90% means that within our diagnostic interval, 90% of patients who are truly infected with HP are correctly identified; a specificity of 84% means that outside our diagnostic interval, 84% of patients who are truly not infected with HP are correctly identified;

[0089] Calculate and obtain the verification result data:

[0090] After the evaluation index is calculated, the evaluation index includes sensitivity and specificity. After weighted calculation of sensitivity and specificity, the comprehensive evaluation index obtained is the verification result data.

[0091] Detection and judgment module:

[0092] Dental plaque samples are collected from the corresponding patients who go to the medical institution, and the uromycin concentration obtained by analysis is compared with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within this range, an early warning signal is issued, indicating that the patient is HP infected; if the uromycin concentration of the corresponding patient is not within this range, no response is given, indicating that the patient is not HP infected.

[0093] Combining the above technical solutions, the concentration of uromycin in dental plaque samples is detected by mass spectrometry technology, and the diagnostic range is determined by combining statistical analysis methods, which can more accurately identify HP-infected patients; the detailed sample collection, processing and analysis steps ensure the reliability and consistency of the data, providing a basis for subsequent research and application.

[0094] The data construction module, under the condition of receiving the early warning signal, builds an oral health status assessment model based on random forest, obtains the corresponding patient's oral parameter set as input features, and outputs the corresponding patient's oral health status score; simultaneously builds a urease activity prediction model based on linear regression, obtains the corresponding patient's oral health status score and environmental parameter set as input features, and outputs the corresponding patient's urease activity prediction value;

[0095] The process of building the model is as follows:

[0096] Data collection: Collect data on oral examination records, microbial culture results, biochemical indicators (such as saliva pH), and patient lifestyle habits (such as smoking, drinking, and oral hygiene habits);

[0097] Data preprocessing: Clean and preprocess the collected data, including removing outliers, filling missing values, and standardizing data to ensure data quality and consistency;

[0098] Feature selection and engineering: Extract features related to oral health status and urease activity from the preprocessed data; these features include: microbial species and quantity, saliva pH, inflammatory factor levels, dietary habits, lifestyle habits, and oral examination indicators (such as gingival bleeding index and dental plaque index);

[0099] Build a model: Select an appropriate machine learning or statistical model to build an oral health status assessment model. Commonly used models include logistic regression, random forest, and support vector machine. The goal of the model is to predict the patient's oral health status score based on the input features. At the same time, another model can be built to predict urease activity. This model can consider the oral health status score as one of the inputs and combine it with other features related to urease activity (such as environmental pH, nickel ion concentration, etc.) for prediction.

[0100] Model training and validation: Use historical data to train the model and evaluate the model performance through cross-validation. Adjust model parameters or select a more appropriate model based on the evaluation results.

[0101] Practical application and adjustment: Apply the trained model to the oral health assessment and urease activity prediction of actual patients; based on the results output by the model, doctors can diagnose the patient's oral health status and make corresponding adjustments to the urease activity (such as by changing eating habits, using oral care products, etc.);

[0102] Specifically, an example of an oral health status assessment model is as follows:

[0103] Input features: Microbial species and quantity: Taking the quantity of three major oral microorganisms as an example, they are bacteria A (quantity: 1000 CFU / mL), bacteria B (quantity: 500 CFU / mL), and bacteria C (quantity: 200 CFU / mL); saliva pH value: 6.5; Inflammatory factor level: Taking interleukin-6 (IL-6) as an example, assume that its concentration is 5 pg / mL;

[0104] Model processing: These input features are substituted into the oral health status assessment model based on random forest. The model will weight and combine these features internally and assess the oral health status according to the patterns learned during training;

[0105] Output: Oral health status score: Assume that the model outputs a score of 75 (the range may be 0-100, with higher scores indicating better oral health status);

[0106] The detailed description of the oral health status assessment model is as follows:

[0107] When we substitute the input features (such as microbial species and quantity, saliva pH, and inflammatory factor levels) into the oral health status assessment model based on random forests, the model will process according to the following steps to obtain the oral health status score:

[0108] Feature processing: The model first preprocesses each input feature, such as normalization or standardization, to ensure that the numerical ranges of different features are consistent and to prevent certain features from occupying too large a weight during model training.

[0109] Decision tree construction: The random forest model consists of multiple decision trees. During the training phase, each decision tree is constructed using a portion of the training data (obtained through bootstrap sampling) and a portion of the features (obtained through random selection). Each decision tree learns the decision rule based on the relationship between the features in the training data and the target variable (i.e., oral health status score).

[0110] Feature weighting and combination: When new patient data is input, each decision tree will classify or regress the data based on the learned decision rules; in random forests, the prediction results of each decision tree are weighted, and these weighted results are combined to get the final prediction value;

[0111] Scoring output: For the oral health status assessment model, the final prediction value is an oral health status score; assuming we have 100 decision trees, each tree gives a prediction score, then the final oral health status score may be the average of these 100 prediction scores;

[0112] It should be noted that in the random forest model, each decision tree will learn a set of decision rules based on the training data; for new input data, each decision tree will give a prediction score based on these rules; finally, the random forest model will average these prediction scores to get the final oral health status score;

[0113] Now, suppose we have only one decision tree (actually a random forest contains multiple trees, but for simplicity we will only consider one); this decision tree may contain the following decision rules:

[0114] If the number of bacteria A is greater than 800 CFU / mL, 20 points will be added;

[0115] If the number of B bacteria is greater than 300 CFU / mL, 15 points will be added;

[0116] If the number of C bacteria is greater than 100 CFU / mL, 10 points will be added;

[0117] If the salivary pH is less than 7.0, 5 points are added;

[0118] If the IL-6 concentration was greater than 3 pg / mL, 15 points were added;

[0119] The basic score is 20 points;

[0120] According to these rules, we can calculate the score of the input features:

[0121] Bacteria count: 1000 CFU / mL (add 20 points if condition 1 is met);

[0122] Bacteria count: 500 CFU / mL (add 15 points if condition 2 is met);

[0123] C bacteria count: 200 CFU / mL (10 points if condition 3 is met);

[0124] Saliva pH: 6.5 (satisfies condition 4, plus 5 points);

[0125] IL-6 concentration: 5 pg / mL (15 points if condition 5 is met);

[0126] The basic score is 20 points;

[0127] Add up these scores: 20 (basic score) + 20 (A bacteria) + 15 (B bacteria) + 10 (C bacteria) + 5 (pH value) + 15 (IL-6) = 85 points; however, please note that this is just a simplified example; in a real random forest model, the decision rules of each decision tree will be more complex, and there will be more trees giving different prediction scores; in the end, these prediction scores will be averaged to obtain an oral health status score between 0 and 100, which is 74 points in the above example;

[0128] Specifically, an example of a urease activity prediction model is as follows:

[0129] Input features: oral health status score: 75 points; environmental pH value: assumed to be 7.0; nickel ion concentration: assumed to be 0.1mmol / L;

[0130] Model processing: These input features are substituted into the urease activity prediction model based on linear regression. The model will predict the urease activity level based on these features and the coefficients learned during training;

[0131] Output: Prediction value of urease activity: Assume that the result predicted by the model is 25U / mL (U represents the unit of urease activity, and the specific value is determined according to the experiment); It should be noted that for the urease activity prediction model, its principle is similar to that of the cavity health status assessment model, so it will not be elaborated here.

[0132] Assessment and Adjustment Module:

[0133] Under the condition that the patient is determined to have HP, the uromycin concentration is corrected based on the standardized oral health status score, urease activity prediction value and uromycin concentration, combined with the verification result data, and a comprehensive calculation formula is built to generate a comprehensive index reflecting the degree of HP infection of the corresponding patient, and the comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is built to set the reexamination frequency of the corresponding patient according to the comparison results, and provide guidance to the patient according to the set reexamination frequency; the above-mentioned thresholds are set according to actual conditions or historical data, and will not be elaborated here;

[0134] Among them, the oral health status score, urease activity prediction value and uromycin concentration were standardized for comparison and calculation in the formula, and the standardization included the use of z-score standardization and minimum-maximum standardization methods to convert the indicator values ​​into values ​​between 0 and 1;

[0135] The comprehensive calculation formula is as follows:

[0136] R=1 / (OH*(1+UEA)^(-1)*(1+UC*XZ)^(-1));

[0137] In the formula, R represents the comprehensive index, OH represents the oral health status score, UEA represents the predicted value of urease activity, UC represents the uromycin concentration, and XZ represents the verification result data, which is used as the correction value in this formula;

[0138] The formula is as follows:

[0139] The larger the OH, the higher the oral health. In the formula, OH is inversely proportional to the HP infection level, that is, the larger the OH, the lower the HP infection level. The larger the UEA, the higher the urease activity, which may indicate a more severe HP infection level. In the formula, we use (1+UEA)^(-1) to ensure that UEA is proportional to the HP infection level, that is, the larger the UEA, the smaller the (1+UEA)^(-1), so that the result of the entire formula (HP infection level) is greater. The larger the UC, the higher the uromycin concentration, which may indicate a more severe HP infection level. In the formula, we use (1+UC)^(-1) to ensure that UC is proportional to the HP infection level, that is, the larger the UC, the smaller the (1+UC)^(-1), so that the result of the entire formula (HP infection level) is greater.

[0140] Example calculation:

[0141] Assume that the standardized indicator data of a patient is:

[0142] Oral health status score OH = 0.8 (after standardization, the higher the score, the healthier the oral health);

[0143] Urease activity prediction value UEA = 0.6 (after standardization, the higher the value, the stronger the activity);

[0144] Uromycin concentration UC = 0.8 (after normalization, higher values ​​indicate higher concentrations);

[0145] Verification result data XZ = (0.5*90+0.5*84)% = 87%

[0146] Then R, which reflects the degree of HP infection in patients, is:

[0147] R=1 / (0.8*(1+0.6)^(-1)*(1+0.8*0.87)^(-1))≈3.4;

[0148] The process of setting the reexamination frequency for the corresponding patient according to the comparison results is as follows:

[0149] When R < k, the patient is set to be reviewed every N years;

[0150] When k≤R<2k, the patient is set to be reviewed once every (N / 2) years;

[0151] When R ≥ 2k, the patient is set to be reviewed once every (N / 4) years;

[0152] The value of N is usually one year; the specific value is set according to actual needs, and N is a positive integer;

[0153] In the above example, the calculated HP infection level is 3.4; based on this value, we can set different thresholds to determine the severity of HP infection and recommend the frequency of reexamination for patients accordingly; for example:

[0154] R<2: Review once a year;

[0155] 2≤R<4: Review every six months;

[0156] R≥4: Review once every quarter;

[0157] Since the patient's HP infection level is 3.4, it is recommended that he undergo a reexamination every six months.

[0158] This program comprehensively considers the oral health status score, urease activity prediction value and uromycin concentration. The oral health status score can not only intuitively reflect the oral health status of the corresponding patient, but also provide data guidance for the calculation of the urease activity prediction value, build a comprehensive calculation formula reflecting the HP infection level, and achieve an objective quantitative evaluation of the HP infection level. According to the patient's HP infection level, this program can also set a personalized re-examination frequency to avoid unnecessary frequent re-examination or insufficient re-examination.

[0159] When conducting preliminary sample analysis, the verification result data is given while obtaining the uromycin concentration range. This can not only assist in determining the credibility of whether the corresponding patient is infected with HP, but also provide correction guidance for the subsequent calculation of the comprehensive index reflecting the degree of HP infection, further improving the accuracy and effectiveness of the comprehensive index.

[0160] Embodiment 2:

[0161] See also Figure 2 Based on Example 1, this example also provides a method for detecting gastric Helicobacter pylori through dental plaque, comprising the following specific steps:

[0162] S1. In any medical institution, collect dental plaque samples from M patients who have been diagnosed with HP infection, use mass spectrometry to obtain the uromycin concentration of each sample, form a data set of uromycin concentration, analyze and process the data set to obtain the uromycin concentration range of HP infected patients, and provide verification result data;

[0163] S2. Collect dental plaque samples from the corresponding patients who go to the medical institution, and compare the uromycin concentration obtained by analysis with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within the range, an early warning signal is issued; otherwise, no response is given;

[0164] S3. Under the condition of receiving the early warning signal, build an oral health status assessment model based on random forest, use the corresponding patient's oral parameter set as input features, and output the corresponding patient's oral health status score; simultaneously build a urease activity prediction model based on linear regression, use the corresponding patient's oral health status score and environmental parameter set as input features, and output the corresponding patient's urease activity prediction value;

[0165] S4. Based on the standardized oral health status score, urease activity prediction value and uromycin concentration, the uromycin concentration is corrected in combination with the verification result data, and a comprehensive calculation formula is constructed to generate a comprehensive index. The comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is constructed. The reexamination frequency of the corresponding patient is set according to the comparison results, and guidance is provided to the patient according to the set reexamination frequency.

[0166] In the application, the several formulas involved are all calculated by taking their numerical values ​​after removing the dimensions, and the formulas are established by collecting a large amount of data and performing software simulation to obtain a formula for the most recent actual situation. Some coefficients or weights in the formulas are set by technical personnel in this field according to actual conditions, so they will not be elaborated here.

[0167] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in electronic hardware, or in a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.

[0168] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0169] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A system for detecting gastric Helicobacter pylori through dental plaque, characterized in that: The system includes: The sample analysis module collects dental plaque samples from M patients diagnosed with HP infection in any medical institution, obtains the uromycin concentration of each sample using mass spectrometry analysis technology, forms a uromycin concentration data set, and obtains the uromycin concentration range of HP infected patients after analysis and processing based on the data set, and provides verification result data; The detection and judgment module collects dental plaque samples from the corresponding patients who go to the medical institution, and compares the uromycin concentration obtained by analysis with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within the range, an early warning signal is issued; otherwise, no response is given; The data construction module, under the condition of receiving the early warning signal, builds an oral health status assessment model based on random forest, obtains the corresponding patient's oral parameter set as input features, and outputs the corresponding patient's oral health status score; simultaneously builds a urease activity prediction model based on linear regression, obtains the corresponding patient's oral health status score and environmental parameter set as input features, and outputs the corresponding patient's urease activity prediction value; The evaluation and adjustment module corrects the uromycin concentration based on the standardized oral health status score, urease activity prediction value and uromycin concentration, combined with the verification result data, and builds a comprehensive calculation formula to generate a comprehensive index. The comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is built to set the review frequency of the corresponding patient according to the comparison results, and provide guidance to the patient according to the set review frequency.

2. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 1, characterized in that: The process of obtaining uromycin concentration using mass spectrometry is as follows: Sample pretreatment: The collected dental plaque samples were pretreated, including extraction and purification; Mass spectrometry analysis: The sample is injected into a mass spectrometer for analysis. The mass spectrometer ionizes the sample molecules and separates and detects them according to their mass-to-charge ratio. The mass spectra are compared with the known database to identify the uromycin molecules in the sample and measure their concentration. Data processing: Use mass spectrometry analysis software to process the mass spectrum, extract the peak area of ​​uromycin, and convert the peak area into uromycin concentration value according to the standard curve.

3. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 2, characterized in that: The process of analyzing and processing the data set based on uromycin concentration is as follows: Data cleaning: Check whether there are outliers or missing values ​​in the data set and perform data cleaning; Statistical analysis: Calculate the mean and standard deviation of uromycin concentration and draw a distribution diagram of uromycin concentration; Determine the threshold: Based on the statistical analysis results, select the mean value plus or minus 2 times the standard deviation as the interval; Validation Evaluation: Using an independent validation dataset, the validation results are evaluated and calculated.

4. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 3, characterized in that: The independent validation dataset contains patient samples with known HP infection status, one part of which is confirmed to be HP infected and the other part is uninfected, that is, the ratio of infected patient samples to uninfected patient samples is 1:

1.

5. The system for detecting gastric Helicobacter pylori through dental plaque according to claim 3, characterized in that: Calculate the verification result data: After the evaluation index is calculated, the evaluation index includes sensitivity and specificity. After weighted calculation of sensitivity and specificity, the comprehensive evaluation degree obtained is the verification result data.

6. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 5, characterized in that: The oral health status score, urease activity prediction value and uromycin concentration were standardized, and the standardization included converting the oral health status score, urease activity prediction value and uromycin concentration into values ​​between 0 and 1 using the z-score standardization and minimum-maximum standardization methods.

7. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 6, characterized in that: The comprehensive calculation formula is as follows: R=1 / (OH*(1+UEA)^(-1)*(1+UC*XZ)^(-1)); Wherein, R represents the comprehensive index, OH represents the oral health status score, UEA represents the predicted value of urease activity, UC represents the uromycin concentration, and XZ represents the verification result data.

8. A system for detecting gastric Helicobacter pylori through dental plaque according to claim 7, characterized in that: The process of setting the reexamination frequency for the corresponding patient according to the comparison results is as follows: When R < k, the patient is set to be reviewed every N years; When k≤R<2k, the patient is set to be reviewed once every (N / 2) years; When R ≥ 2k, the patient is set to be reviewed once every (N / 4) years; The value of N is one year, and N is a positive integer.

9. A method for detecting gastric Helicobacter pylori through dental plaque, using the system according to any one of claims 1 to 8, characterized in that: The steps include: S1. In any medical institution, collect dental plaque samples from M patients who have been diagnosed with HP infection, use mass spectrometry to obtain the uromycin concentration of each sample, form a data set of uromycin concentration, analyze and process the data set to obtain the uromycin concentration range of HP infected patients, and provide verification result data; S2. Collect dental plaque samples from the corresponding patients who go to the medical institution, and compare the uromycin concentration obtained by analysis with the uromycin concentration range. If the uromycin concentration of the corresponding patient is within the range, an early warning signal is issued; Otherwise, no response will be made; S3. Under the condition of receiving the early warning signal, build an oral health status assessment model based on random forest, use the corresponding patient's oral parameter set as input features, and output the corresponding patient's oral health status score; simultaneously build a urease activity prediction model based on linear regression, use the corresponding patient's oral health status score and environmental parameter set as input features, and output the corresponding patient's urease activity prediction value; S4. Based on the standardized oral health status score, urease activity prediction value and uromycin concentration, the uromycin concentration is corrected in combination with the verification result data, and a comprehensive calculation formula is constructed to generate a comprehensive index. The comprehensive index is compared with the set threshold group {k, 2k}, and a rule engine is constructed. The reexamination frequency of the corresponding patient is set according to the comparison results, and guidance is provided to the patient according to the set reexamination frequency.

Images