Prediction model-combined vaginal flora index-based pregnancy position prediction method and device, equipment and medium
By combining prediction models and vaginal microbial indicators in pregnancy position prediction, key index data are extracted and mathematical model prediction is carried out, the problem of inaccurate pregnancy position prediction in the prior art is solved, and the accuracy of prediction and the convenience of clinical application are improved.
Patent Information
- Application Number
- CN202411992210.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-31
- Publication Date
- 2025-06-06
AI Technical Summary
The prior art cannot accurately predict the position of pregnancy and cannot provide an accurate basis for judging the position of pregnancy.
Using a method based on a prediction model combined with vaginal microbial indicators, the detection report and medical history data of the target object are obtained, key index data, including bacterial-related information and key medical history information, and the mathematical model is used to predict, and the pregnancy position prediction results are obtained.
It improves the accuracy of pregnancy position prediction, provides convenience for clinical practice, optimizes treatment plans, and solves the problem that the existing technology cannot accurately predict pregnancy position.
Smart Images
Figure CN120108685A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of data mining and analysis, and in particular to a method, device, equipment and medium for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators. Background Art
[0002] At present, identifying the location of early pregnancy is a bottleneck in clinical diagnosis and treatment. There are many methods for monitoring the population of unknown pregnancy (PUL). The most commonly used clinical examination methods are transvaginal ultrasound (TVS) and serological marker detection. In serological tests, progesterone (P) and human chorionic gonadotropin (HCG) are often used for judgment, but they only confirm whether pregnancy is present and lack accuracy in determining the location of pregnancy. With the continuous development of ultrasound technology, TVS has improved the accuracy and specificity of determining the location of pregnancy in women. However, studies have shown that when women diagnosed with EP at the first TVS were compared with women diagnosed with EP after being classified as PUL, it was found that the smaller gestational age and gestational sac, which are difficult to be found under B-ultrasound, are the reasons why the location of pregnancy cannot be determined during the initial color Doppler ultrasound examination.
[0003] Related studies have shown that vaginal flora is related to adverse pregnancy, and the focus has also been shifted to the vaginal microecology in early pregnancy. However, existing solutions have failed to accurately predict the location of pregnancy based on vaginal flora, and cannot provide an accurate basis for determining the location of pregnancy. Summary of the invention
[0004] The present application provides a method, device, equipment and medium for predicting the position of pregnancy based on a prediction model combined with vaginal flora indicators. Vaginal flora indicators are added on the basis of key indicators, and the indicators are predicted using mathematical models. The pregnancy position of PUL can be accurately evaluated to provide convenience for clinical practice, promote communication between doctors and patients, optimize treatment plans, and solve the problem that existing plans cannot accurately predict the position of pregnancy.
[0005] In a first aspect, the present application provides a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators, comprising:
[0006] Obtaining a test report and medical history data of a target subject, wherein the test report at least includes vaginal flora test information and symptom-related information of the target subject;
[0007] According to preset key analysis items, data analysis and extraction are performed based on the test report and the medical history data to obtain key indicator data, wherein the key indicator data includes flora-related information and key medical history information related to the pregnancy position;
[0008] By using a preset prediction model, feature analysis is performed based on the key indicator data to obtain a multi-indicator prediction feature, wherein the multi-indicator prediction feature includes clinical information features and flora analysis features related to pregnancy position prediction;
[0009] Predicting the position of pregnancy based on the multi-index prediction feature to obtain a prediction result of the position of pregnancy;
[0010] The pregnancy position prediction result is used to optimize the treatment plan for the target object.
[0011] Optionally, obtain the test report and medical history data of the target subject, including:
[0012] Collect medical history data of the target subject through the information collection page;
[0013] Scan the documents provided by the target object through the recognition scanning technology to obtain the test report, or obtain the relevant data input by the target object through the information input interface to obtain the test report;
[0014] The vaginal flora detection information in the detection report is detection information obtained by detecting and analyzing the flora sample of the target object.
[0015] Optionally, according to preset key analysis items, data analysis and extraction are performed based on the test report and the medical history data to obtain key indicator data, including:
[0016] Extracting the analysis test items corresponding to the test report and the medical history analysis items corresponding to the medical history data from the preset key indicator items;
[0017] According to the analysis and detection items, extract relevant detection data to be analyzed from the detection report;
[0018] Performing analysis and processing based on the detection data to be analyzed to obtain flora-related information;
[0019] According to the medical history analysis items, the medical history data is preprocessed and analyzed to obtain key medical history information.
[0020] Optionally, a preset prediction model is used to perform feature analysis based on the key indicator data to obtain multi-indicator prediction features, including:
[0021] Analyzing and extracting the key medical history information through a preset prediction model to obtain clinical information features related to pregnancy position prediction;
[0022] By using high-throughput sequencing technology, the flora-related information is used to perform flora composition and diversity analysis and extraction, and obtain flora analysis features related to pregnancy position prediction;
[0023] The microbial community analysis characteristics include at least two of abundance characteristics, diversity index characteristics, structural change characteristics and metabolic characteristics.
[0024] Optionally, performing pregnancy position prediction based on the multi-index prediction feature to obtain a pregnancy position prediction result includes:
[0025] Analyze and evaluate the multi-indicator prediction features to obtain the score information corresponding to the key analysis items of the target object;
[0026] The pregnancy position is predicted according to the score information to obtain a pregnancy position prediction result.
[0027] Optionally, performing analysis and evaluation based on the multi-indicator prediction features to obtain score information corresponding to the target object in key analysis items includes:
[0028] Get the nomogram construction function;
[0029] Through the nomogram construction function, the multi-index prediction characteristics are used for analysis to obtain score information, and a nomogram is constructed according to the score information for display.
[0030] Optionally, after obtaining the score information, the step further includes:
[0031] High-risk screening is performed based on the score information to obtain high-risk prediction information, and the high-risk prediction information is used to optimize the treatment plan for the target object in combination with the pregnancy position prediction result.
[0032] In a second aspect, the present application provides a pregnancy position prediction device based on a prediction model combined with vaginal flora indicators, comprising:
[0033] A module for acquiring data to be analyzed, used to acquire a test report and medical history data of a target subject, wherein the test report at least includes vaginal flora test information and symptom-related information of the target subject;
[0034] A data analysis and extraction module, used to perform data analysis and extraction based on the test report and the medical history data according to preset key analysis items to obtain key indicator data, wherein the key indicator data includes flora-related information and key medical history information related to the pregnancy position;
[0035] A feature analysis module, used to perform feature analysis based on the key indicator data and the medical history data through a preset prediction model to obtain multi-indicator prediction features, wherein the multi-indicator prediction features include clinical information features and flora analysis features related to pregnancy position prediction;
[0036] A prediction module is used to predict the position of pregnancy based on the multi-indicator prediction feature to obtain a pregnancy position prediction result; wherein the pregnancy position prediction result is used to optimize the treatment plan for the target object.
[0037] In a third aspect, the present application provides an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus;
[0038] Memory, used to store computer programs;
[0039] The processor is used to implement the steps of the method for predicting the location of pregnancy based on the prediction model combined with vaginal flora indicators as described in any embodiment of the first aspect when executing the program stored in the memory.
[0040] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators as described in any embodiment of the first aspect.
[0041] In summary, the embodiment of the present application extracts key indicator data such as flora-related information and key medical history information related to the position of pregnancy from the acquired test reports and medical history data according to key analysis items, and then extracts multi-indicator prediction features related to the prediction of the position of pregnancy from the key indicator data through mathematical model prediction, and then uses the multi-indicator prediction features to predict the position of pregnancy to obtain a pregnancy position prediction result. This embodiment uses a mathematical model to predict the subsequent pregnancy position of the PUL population based on the vaginal flora and clinical indicators, thereby improving the accuracy of pregnancy position prediction, providing convenience for clinical practice, optimizing treatment plans, and solving the problem that the prior art cannot accurately predict the position of pregnancy based on the vaginal flora and cannot provide an accurate basis for the judgment of the position of pregnancy. BRIEF DESCRIPTION OF THE DRAWINGS
[0042] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.
[0043] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.
[0044] Figure 1 A schematic diagram of a flow chart of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators provided in an embodiment of the present application;
[0045] Figure 2 This is a schematic diagram of the steps of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators provided by an optional embodiment of the present application;
[0046] Figure 3 It is a flowchart of a method of combining vaginal flora indicators to predict the location of pregnancy provided by an optional example of the present application;
[0047] Figure 4 It is a nomogram provided by an optional example of this application;
[0048] Figure 5 A structural block diagram of a pregnancy position prediction device based on a prediction model combined with vaginal flora indicators provided in an embodiment of the present application;
[0049] Figure 6 It is a structural schematic diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION
[0050] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0051] To facilitate the understanding of the embodiments of the present application, further explanation will be given below in conjunction with the drawings and specific embodiments. The embodiments do not constitute a limitation on the embodiments of the present application.
[0052] Figure 1 This is a flow chart of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators provided in an embodiment of the present application. Figure 1 As shown, the method for predicting the position of pregnancy based on the prediction model combined with vaginal flora indicators provided in the embodiment of the present application may specifically include the following steps:
[0053] Step 110, obtaining the test report and medical history data of the target object.
[0054] The detection report at least includes vaginal flora detection information and symptom-related information of the target object.
[0055] In this embodiment, the target object may refer to a patient, and the test report may include, in addition to a series of test information related to the vaginal flora information of the target object, symptom-related information, such as but not limited to: vaginal bleeding, abdominal pain, human chorionic gonadotropin, progesterone, LACTOBACILLU, GARDNERELLA, etc. The medical history data can be understood as the past medical history of the target object, which may include but not limited to: history of ectopic pregnancy, history of pelvic infection, etc.
[0056] Step 120, according to preset key analysis items, data analysis and extraction are performed based on the test report and the medical history data to obtain key indicator data.
[0057] Among them, the key indicator data includes flora-related information and key medical history information related to the pregnancy position.
[0058] In this embodiment, the key analysis items are preset analysis items related to the prediction of pregnancy location, which may include but are not limited to: vaginal bleeding, abdominal pain, history of ectopic pregnancy, history of pelvic infection, human chorionic gonadotropin, progesterone, LACTOBACILLU, GARDNERELLA.
[0059] In a specific implementation, both the test report and the medical history data may contain irrelevant data, redundant data, etc. To prevent these irrelevant data from affecting the accuracy of the pregnancy position prediction, the present embodiment can preset key analysis items according to demand. The key analysis items usually include the analysis items required for the subsequent pregnancy position prediction. Then, based on the acquisition of the test report and the medical history data, the data corresponding to the key analysis items are extracted from the test report and the medical history data according to the key analysis items to obtain the key indicator data. The key indicator data can be used as the basis for the subsequent pregnancy position prediction.
[0060] Step 130, performing feature analysis based on the key indicator data using a preset prediction model to obtain multi-indicator prediction features.
[0061] Among them, the multi-index prediction features include clinical information features and microbiome analysis features related to pregnancy position prediction.
[0062] In the related art, the prediction of pregnancy position is a complex medical problem involving multiple factors. The prior art usually has problems such as low accuracy in predicting pregnancy position. To solve the problems of the prior art, the type of prediction model and the algorithm used in the prediction model in this embodiment can be selected according to the specific characteristics of the data and the complexity of the problem. For example, it can include but is not limited to: support vector machine (SVM) model, random forest model and logistic regression, etc.
[0063] In this embodiment, the key indicator data extracted from the test report and medical history data can be input into a pre-trained prediction model, and the prediction model performs feature analysis and extraction based on the input key indicator data, and can extract flora analysis features related to the prediction of pregnancy position from flora-related information, and extract clinical information features related to the prediction of pregnancy position from key medical history information. Thus, this embodiment realizes the extraction of prediction features of multiple indicators from test reports and medical history data.
[0064] Step 140: Predict the pregnancy position based on the multi-index prediction feature to obtain a pregnancy position prediction result.
[0065] The pregnancy position prediction result is used to optimize the treatment plan for the target object.
[0066] In this embodiment, the flora analysis feature in the multi-indicator prediction feature is obtained based on the analysis of the test report related to the vaginal flora. Therefore, by using the multi-indicator prediction feature to predict the position of pregnancy, the position of pregnancy based on the vaginal flora can be predicted. The multi-indicator prediction feature also includes clinical information features and other features extracted from the target object's medical history. This embodiment combines the two features to simultaneously predict the position of pregnancy, thereby improving the accuracy of pregnancy position prediction.
[0067] It can be seen that the embodiment of the present application extracts key indicator data such as flora-related information and key medical history information related to the position of pregnancy from the acquired test reports and medical history data according to key analysis items, and then extracts multi-indicator prediction features related to the prediction of the position of pregnancy from the key indicator data through mathematical model prediction, and then uses the multi-indicator prediction features to predict the position of pregnancy to obtain a pregnancy position prediction result. This embodiment uses a mathematical model to predict the subsequent pregnancy position of the PUL population based on the vaginal flora and clinical indicators, thereby improving the accuracy of pregnancy position prediction, providing convenience for clinical practice, optimizing treatment plans, and solving the problem that the existing technology cannot accurately predict the position of pregnancy based on the vaginal flora.
[0068] Reference Figure 2 , shows a schematic flow chart of the steps of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators provided by an optional embodiment of the present application. The method may specifically include the following steps:
[0069] Step 210, obtaining the test report and medical history data of the target object.
[0070] The detection report at least includes vaginal flora detection information and symptom-related information of the target object.
[0071] In a specific implementation, the test report is usually a relevant test report obtained by inspecting the target object. Figure 3 Taking the prediction process shown as an example, this embodiment can obtain the test report of the target object by means of data collection, or by setting an information entry page to obtain relevant information input by the target object through the information entry interface as a test report, or, it can also be through setting an information collection device, using OCR (optical character recognition), etc., by detecting the scanned copy of the test report provided by the target object, etc., to identify and obtain the test report; for medical history data, this embodiment can set an information entry interface, and obtain the relevant medical history information by the target object inputting the relevant medical history information through the information entry interface, and of course, it can also be obtained by identifying the scanned copy of the medical history information provided by the target object through the information collection device. This embodiment does not limit the method of obtaining the test report and medical history data.
[0072] In an optional embodiment, the embodiment of the present application obtains the test report and medical history data of the target object, which may include: recording the clinical information of the target object to obtain medical history data; scanning and processing the test file of the target object through recognition scanning technology to obtain a test report, or obtaining relevant data input by the target object through an information entry interface to obtain a test report; wherein the vaginal flora detection information in the test report is detection information obtained by detecting and analyzing the flora sample of the target object.
[0073] In actual implementation, predicting the location of pregnancy based on vaginal flora is a complex process, which involves an in-depth understanding of the vaginal microbiome and advanced bioinformatics and statistical methods. For the flora detection information in the test report, this embodiment can be obtained by detecting and analyzing the flora samples of the target object. For example, the vaginal flora samples in the flora samples can be detected by high-throughput sequencing technology (such as 16S rRNA sequencing) to obtain detailed flora composition information as the flora detection information, such as Figure 3 The process shown is taken as an example. Subsequently, data analysis can be performed using bioinformatics and statistical methods to identify specific bacterial communities or bacterial community combinations associated with different gestational positions.
[0074] Step 220, extracting the analysis test items corresponding to the test report and the medical history analysis items corresponding to the medical history data from the preset key indicator items.
[0075] Step 230: extract relevant test data to be analyzed from the test report according to the analysis test items.
[0076] Step 240: Analyze and process the detection data to be analyzed to obtain flora-related information.
[0077] Step 250, preprocessing and analyzing the medical history data according to the medical history analysis items to obtain key medical history information.
[0078] Steps 220 to 250 are described uniformly:
[0079] In this embodiment, the key analysis items may include analysis and detection items and medical history analysis items. The analysis and detection items refer to the analysis items of vaginal flora, and the medical history analysis items refer to the analysis items of past medical history. The analysis items of vaginal flora may include but are not limited to: flora species, flora abundance, flora diversity, flora structure, and metabolites; the analysis items of past medical history may include but are not limited to: the target subject's age, reproductive history, and past medical history.
[0080] In the specific implementation, this embodiment first extracts analysis test items and medical history analysis items from the key analysis items. For the test report, this embodiment extracts relevant test data to be analyzed from the test report according to the analysis test items, and then, on this basis, performs analysis and processing based on the test data to be analyzed to obtain flora-related information; for medical history data, this embodiment extracts corresponding key medical history information from the medical history data according to the medical history analysis items.
[0081] In actual processing, both the test report and the medical history data may contain data irrelevant to the pregnancy position prediction, such as personal information and other irrelevant data, and in particular, there may be a lot of irrelevant data in the medical history data of the target object. In order to prevent these irrelevant data from affecting the accuracy of the pregnancy position prediction, this embodiment can perform pre-processing such as data filtering on the test report and the medical history data, remove irrelevant data, and then use key analysis items to extract data, thereby ensuring that the extracted data is relevant to the pregnancy position prediction and can effectively reduce the amount of calculation of the model processing data.
[0082] Step 260: Analyze and extract the key medical history information through a preset prediction model to obtain clinical information features related to pregnancy position prediction.
[0083] Step 270, using high-throughput sequencing technology, the flora-related information is used to perform flora composition and diversity analysis and extraction to obtain flora analysis features related to pregnancy position prediction.
[0084] The microbial community analysis characteristics include at least two of abundance characteristics, diversity index characteristics, structural change characteristics and metabolic characteristics.
[0085] A unified description of steps 260 to 270 is given as follows:
[0086] In the specific implementation, in the prediction model based on vaginal flora for predicting the location of pregnancy, the specific features involved mainly include the composition and diversity of the vaginal microbiome. This embodiment uses the trained prediction model to analyze and extract key medical history information and flora-related information, thereby obtaining clinical information features corresponding to key medical history information and flora analysis features corresponding to flora-related information. These multi-index prediction features are closely related to the prediction of pregnancy location. Among them, the flora analysis features can also be called vaginal flora features, which is not limited in this embodiment.
[0087] In this embodiment, since the relationship between vaginal flora and pregnancy position may be affected by multiple factors, it is necessary to comprehensively consider multiple features when extracting relevant features of flora using the prediction model. In order to improve the accuracy of pregnancy position prediction, flora analysis features in this embodiment may include abundance features, diversity index features, structural change features, and metabolic features, and at least two flora analysis features may be used for subsequent pregnancy position prediction according to actual needs. It should be noted that in this embodiment, the more types of features included in the flora analysis features, the higher the accuracy of subsequent pregnancy position prediction can be.
[0088] Among them, the abundance feature refers to the abundance of a specific bacterial community. Specifically, the abundance of certain bacterial species or bacterial communities may be related to the location of pregnancy. For example, certain Lactobacillus species may dominate in intrauterine pregnancy, but may be less in ectopic pregnancy. Therefore, the abundance of these specific bacterial communities can be used as one of the important input features of the prediction model.
[0089] The diversity index feature refers to the flora diversity index. Specifically, the diversity of the vaginal microbiome can be measured by a diversity index, such as the Shannon index or the Simpson index. These indices reflect the richness and uniformity of the microbial species in the vagina. Different gestational positions may correspond to different flora diversity levels, so the diversity index is also an important feature in the prediction model.
[0090] Structural change characteristics refer to changes in the structure of the microbiota. Specifically, the structure of the vaginal microbiota may differ in different pregnancy positions. For example, ectopic pregnancy may be associated with the proliferation of certain pathogenic bacteria and the reduction of probiotics. Therefore, by analyzing the changes in the microbiota structure, features related to the pregnancy position can be extracted.
[0091] Metabolic signatures usually refer to the characteristics of metabolites. Specifically, the metabolites of the vaginal microbiota may also be related to the position of pregnancy. These metabolites can be detected and analyzed using metabolomics techniques. The levels of certain metabolites may be associated with specific gestational positions and therefore can also be used as one of the features of the predictive model.
[0092] In actual processing, in addition to vaginal flora characteristics, the prediction model can also combine clinical information on the basis of vaginal flora characteristics to improve prediction accuracy. For example, clinical information such as the age, reproductive history, and previous medical history of pregnant women may be related to the position of pregnancy and can therefore be used as auxiliary input features of the prediction model.
[0093] Furthermore, in actual implementation, before the prediction model is used for feature extraction in this embodiment, the prediction model can also be trained. Specifically, an initial prediction model is constructed using machine learning or statistical methods, and then relevant data is collected, including the test results of flora samples of different target objects, medical history data, and relevant pregnancy position conditions, etc. The collected relevant data are divided into a training set and a test set. The prediction model is then trained using the training set, and the trained prediction model is evaluated using the test set to check its prediction accuracy on unknown data. According to the evaluation results, the prediction model is optimized to improve its prediction performance. Among them, the model optimization process includes but is not limited to: adjusting model parameters, changing model algorithms, or adding new feature variables, etc. As a result, this embodiment obtains an optimal model that meets the preset accuracy, which is used as a trained prediction model for accurate vaginal flora-based pregnancy position prediction.
[0094] Step 280: Analyze and evaluate the multi-index prediction features to obtain score information corresponding to the key analysis items of the target object.
[0095] In actual implementation, this embodiment can set corresponding scoring indicators for each prediction feature in the multi-indicator prediction feature. After obtaining the multi-indicator prediction feature, each prediction feature can be analyzed and evaluated according to the preset indicators to obtain the corresponding score.
[0096] In a specific implementation, this embodiment can combine the scores corresponding to each indicator to determine the final score of the target object, and the score can reflect the risk situation of the target object so that the treatment plan can be optimized.
[0097] Exemplarily, each patient is scored based on the following medical history, symptoms and laboratory indicators: vaginal bleeding, abdominal pain, history of ectopic pregnancy, history of pelvic infection (if the patient has not had an infection, this item is scored as 0, if it has, it is scored as 1, i.e. 0 = no, 1 = yes), human chorionic gonadotropin, progesterone, LACTOBACILLU, GARDNERELLA.
[0098] In an optional embodiment, the embodiment of the present application performs analysis and evaluation based on the multi-indicator prediction characteristics to obtain score information corresponding to the target object in the key analysis items, including: obtaining a nomogram construction function; using the nomogram construction function to analyze using the multi-indicator prediction characteristics to obtain score information, and constructing a nomogram according to the score information for display.
[0099] In a specific implementation, the nomogram construction function mainly includes a scoring function for assigning points to each prediction feature, and a graph construction function for constructing a nomogram. In this embodiment, the nomogram construction function can be used to first assign points to each prediction feature to obtain a corresponding score, and then a nomogram can be constructed to display the score of each item through the nomogram.
[0100] For example, refer to Figure 4 The nomogram shown. Each patient is assigned a score based on the patient's medical history, symptoms and laboratory indicators: vaginal bleeding, abdominal pain, history of ectopic pregnancy, history of pelvic infection (0 = no, 1 = yes), human chorionic gonadotropin, progesterone, LACTOBACILLU, GARDNERELLA. For example, a patient with a history of ectopic pregnancy can score 40 points based on this feature analysis.
[0101] Optionally, after obtaining the score information, the embodiment of the present application may further include: performing high-risk screening processing according to the score information to obtain high-risk prediction information, and the high-risk prediction information is used to optimize the treatment plan for the target object in combination with the pregnancy position prediction result.
[0102] In a specific implementation, this embodiment performs high-risk screening on the target object based on the score information, determines whether the target object has a high risk, and optimizes the treatment plan for the target object in combination with the pregnancy position prediction result. Therefore, this embodiment further performs reasonable high-risk screening based on the score based on the prediction feature analysis score, thereby helping medical personnel to propose a reasonable treatment plan.
[0103] For example, for patients with a history of ectopic pregnancy, according to this feature analysis, the score read in the nomogram may be 40 points, and high-risk screening is performed based on the score. Specifically, if the score is less than 130, the possibility of intrauterine pregnancy is high. For example, if there is vaginal bleeding at this time, it is considered that there is a high possibility of threatened abortion, and symptomatic tocolysis can be treated. If the score is greater than 130 points, the possibility of ectopic pregnancy is high, and close clinical observation is required. If the score is greater than 150 points, the possibility of pregnancy is more than 0.8, or surgical intervention treatment needs to be considered.
[0104] Exemplarily, in order to reasonably calculate the score of each prediction item in the nomogram, this embodiment proposes the following formula to achieve reasonable calculation: logit(PREGNCYLOCATION)=0.60255-1.64836*LACTOBACILLUS+0.81124*(PREVIOUSPELVICINFECTION=1)+2.66320*(PREVIOUSECTOPICPREGNCY=1)-0.13437*PROGESTIN+0.18322*(UTERINEBLEEDING=1)+1.77060*(ABDOMILPAIN=1)+0.46866*NEWHCG.LOG1+0.27553*GARDNERELLA.LOG1.
[0105] Step 290: Predict the pregnancy position according to the score information to obtain a pregnancy position prediction result.
[0106] The pregnancy position prediction result is used to optimize the treatment plan for the target object.
[0107] In this embodiment, a prediction model for the pregnancy location of the population with unknown pregnancy location is established by combining vaginal flora with clinical indicators, which can provide doctors with prediction ideas, so as to make timely corrections and maintain the balance of vaginal flora when preventing pregnancy outcomes.
[0108] In the relevant technology, previous statistical results show that the number of deaths caused by tubal pregnancy rupture accounts for 2.7% of all maternal deaths. This embodiment uses digital model prediction to provide convenience for clinical practice, increase the decision-making ability of medical workers, and optimize follow-up and treatment plans. The prediction model of this embodiment adds vaginal flora indicators to evaluate the pregnancy location of PUL, including vaginal bleeding, abdominal pain, history of ectopic pregnancy, history of pelvic infection, human chorionic gonadotropin, progesterone, LACTOBACILLU, GARDNERELLA, etc. In terms of presentation, it also changes the usual way of presenting the prediction model with mathematical formulas, and uses nomograms to intuitively present the evaluation process and results to clinical staff, patients and their families, promoting more information visibility in communication between doctors and patients. At the same time, it also has a certain reference significance for improving the vaginal microecology of patients and conducting microbiological diagnosis, and provides new intervention targets at the microecological level for women with potential EP risks.
[0109] In summary, the embodiment of the present application extracts the test data from the acquired search report according to the analysis test items, performs analysis and processing, obtains flora-related information, and performs data preprocessing and analysis on the acquired medical history data according to the medical history analysis items to obtain key medical history information, and then uses the prediction model, combined with high-throughput sequencing technology, to perform multiple corresponding analysis and extraction on the key medical history information and flora-related information, respectively, to obtain flora analysis features and clinical information features related to the prediction of pregnancy position, and then analyzes and evaluates according to the acquired prediction features, uses the obtained score information to predict the pregnancy position, obtains the pregnancy position prediction result, and thus optimizes the treatment plan for the target object. As a result, the embodiment of the present application achieves the improvement of the accuracy of pregnancy position prediction, solves the problem that the prior art cannot accurately predict the pregnancy position based on the vaginal flora, and cannot provide an accurate basis for the judgment of the pregnancy position.
[0110] Furthermore, the pregnancy location prediction method based on the prediction model combined with vaginal flora indicators provided in the embodiment of the present application also has social and economic benefits. This embodiment uses digital model prediction to provide convenience for clinical practice, increase the decision-making ability of medical workers, and optimize follow-up and treatment plans. In terms of treatment, whether it is ectopic pregnancy embryocidal treatment or threatened abortion tocolysis treatment, traditional Chinese medicine treatment has significant advantages. For patients with ectopic pregnancy who wish to have children, conservative embryocidal treatment with traditional Chinese medicine can preserve the patient's fertility compared to salpingectomy; in terms of economic benefits, conservative embryocidal treatment with traditional Chinese medicine can significantly reduce the economic burden of the disease, thereby avoiding the situation where the patient can only wait and observe when the location of the pregnancy is unclear, which leads to multiple risks, including: Risk one is that the patient risks the life of a possible rupture of an ectopic pregnancy, and risk two is missing the best time for conservative treatment with traditional Chinese medicine.
[0111] It should be noted that, for the purpose of simple description, the method embodiments are expressed as a series of action combinations, but those skilled in the art should be aware that the embodiments of the present application are not limited to the described order of actions, because according to the embodiments of the present application, certain steps may be performed in other orders or simultaneously.
[0112] like Figure 5 As shown, the embodiment of the present application also provides a pregnancy position prediction device 500 based on a prediction model combined with vaginal flora indicators, comprising:
[0113] The data acquisition module 510 to be analyzed is used to acquire the test report and medical history data of the target object, wherein the test report at least includes the vaginal flora test information and symptom-related information of the target object;
[0114] A data analysis and extraction module 520 is used to perform data analysis and extraction based on the test report and the medical history data according to preset key analysis items to obtain key indicator data, wherein the key indicator data includes flora-related information and key medical history information related to the pregnancy position;
[0115] A feature analysis module 530 is used to perform feature analysis based on the key indicator data and the medical history data through a preset prediction model to obtain a multi-indicator prediction feature, wherein the multi-indicator prediction feature includes clinical information features and flora analysis features related to pregnancy position prediction;
[0116] The prediction module 540 is used to predict the pregnancy position based on the multi-indicator prediction feature to obtain a pregnancy position prediction result; wherein the pregnancy position prediction result is used to optimize the treatment plan for the target object.
[0117] Optionally, the to-be-analyzed data acquisition module 510 includes:
[0118] A medical history data acquisition submodule, used to collect the medical history data of the target object through an information collection page;
[0119] The test report acquisition submodule is used to scan and process the file provided by the target object through the identification scanning technology to obtain the test report, or to obtain the relevant data input by the target object through the information entry interface to obtain the test report; wherein the vaginal flora detection information in the test report is the detection information obtained by detecting and analyzing the flora sample of the target object.
[0120] Optionally, the data analysis and extraction module 520 includes:
[0121] The analysis item extraction submodule is used to extract the analysis test items corresponding to the test report and the medical history analysis items corresponding to the medical history data from the preset key indicator items.
[0122] A test data extraction submodule, used to extract relevant test data to be analyzed from the test report according to the analysis test items;
[0123] An analysis and processing submodule, used to perform analysis and processing based on the detection data to be analyzed to obtain flora-related information;
[0124] The preprocessing and analysis submodule is used to preprocess and analyze the medical history data according to the medical history analysis items to obtain key medical history information.
[0125] Optionally, the feature analysis module 530 includes:
[0126] A first analysis and extraction submodule is used to analyze and extract the key medical history information through a preset prediction model to obtain clinical information features related to pregnancy position prediction;
[0127] The second analysis and extraction submodule is used to use the microbial community-related information to perform microbial community composition and diversity analysis and extraction through high-throughput sequencing technology to obtain microbial community analysis features related to pregnancy position prediction; wherein the microbial community analysis features include at least two of abundance features, diversity index features, structural change features, and metabolic features.
[0128] Optionally, the prediction module 540 includes:
[0129] An analysis and evaluation submodule, used to perform analysis and evaluation based on the multi-indicator prediction characteristics to obtain score information corresponding to the key analysis items;
[0130] The prediction submodule is used to predict the pregnancy position according to the score information to obtain a pregnancy position prediction result.
[0131] Optionally, the pregnancy position prediction device 500 further includes:
[0132] A screening and processing module is used to perform high-risk screening processing according to the score information to obtain high-risk prediction information, and the high-risk prediction information is used to optimize the treatment plan for the target object in combination with the pregnancy position prediction result.
[0133] Optionally, the analysis and evaluation submodule includes:
[0134] A construction function acquisition module is used to obtain a nomogram construction function;
[0135] A score determination submodule is used to construct a function through the nomogram, analyze using multi-indicator prediction features, and obtain score information;
[0136] The nomogram construction and display module is used to construct a nomogram for display according to the score information.
[0137] It should be noted that the pregnancy location prediction device based on the prediction model combined with vaginal flora indicators provided in the embodiment of the present application can execute the pregnancy location prediction method based on the prediction model combined with vaginal flora indicators provided in any embodiment of the present application, and has the corresponding functions and beneficial effects of the execution method.
[0138] In a specific implementation, the above-mentioned pregnancy position prediction device based on the prediction model combined with the vaginal flora index can be integrated in a device, so that the device can predict the pregnancy position of the target object by using the vaginal flora information through the data model, as an electronic device, to improve the accuracy of pregnancy position prediction, and then achieve the purpose of optimizing the treatment plan. The electronic device can be composed of two or more physical entities, or it can be composed of one physical entity, such as the electronic device can be a personal computer (PC), a computer, a server, etc., and the embodiment of the present application does not make specific restrictions on this.
[0139] like Figure 6 As shown, an embodiment of the present application provides an electronic device, including a processor 111, a communication interface 112, a memory 113 and a communication bus 114, wherein the processor 111, the communication interface 112, and the memory 113 communicate with each other through the communication bus 114; the memory 113 is used to store computer programs; the processor 111 is used to execute the program stored in the memory 113, and implement the steps of the pregnancy location prediction method based on the prediction model combined with the vaginal flora index provided in any of the aforementioned method embodiments. Exemplarily, the steps of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators may include the following steps: obtaining a test report and medical history data of a target subject, the test report at least containing vaginal flora test information and symptom-related information of the target subject; performing data analysis and extraction based on the test report and the medical history data according to preset key analysis items to obtain key indicator data, the key indicator data containing flora-related information and key medical history information related to the location of pregnancy; performing feature analysis based on the key indicator data through a preset prediction model to obtain multi-indicator prediction features, the multi-indicator prediction features including clinical information features and flora analysis features related to the prediction of the location of pregnancy; performing pregnancy location prediction based on the multi-indicator prediction features to obtain a pregnancy location prediction result; wherein the pregnancy location prediction result is used to optimize the treatment plan for the target subject.
[0140] An embodiment of the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the steps of a method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators as provided in any of the aforementioned method embodiments are implemented.
[0141] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.
[0142] The above description is only a specific implementation of the present application, so that those skilled in the art can understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest range consistent with the principles and novel features applied for herein.
Claims
1. A method for predicting pregnancy position based on a prediction model combined with vaginal flora indicators, characterized in that: include: Obtaining a test report and medical history data of a target subject, wherein the test report at least includes vaginal flora test information and symptom-related information of the target subject; According to preset key analysis items, data analysis and extraction are performed based on the test report and the medical history data to obtain key indicator data, wherein the key indicator data includes flora-related information and key medical history information related to the pregnancy position; By using a preset prediction model, feature analysis is performed based on the key indicator data to obtain a multi-indicator prediction feature, wherein the multi-indicator prediction feature includes clinical information features and flora analysis features related to pregnancy position prediction; Predicting the position of pregnancy based on the multi-index prediction feature to obtain a prediction result of the position of pregnancy; The pregnancy position prediction result is used to optimize the treatment plan for the target object.
2. The method according to claim 1, characterized in that Obtain the test report and medical history data of the target subject, including: Collect medical history data of the target subject through the information collection page; Scan the documents provided by the target object through the recognition scanning technology to obtain the test report, or obtain the relevant data input by the target object through the information input interface to obtain the test report; The vaginal flora detection information in the detection report is detection information obtained by detecting and analyzing the flora sample of the target object.
3. The method according to claim 1, characterized in that According to the preset key analysis items, data analysis and extraction are performed based on the test report and the medical history data to obtain key indicator data, including: Extracting the analysis test items corresponding to the test report and the medical history analysis items corresponding to the medical history data from the preset key indicator items; According to the analysis and detection items, extract relevant detection data to be analyzed from the detection report; Performing analysis and processing based on the detection data to be analyzed to obtain flora-related information; According to the medical history analysis items, the medical history data is preprocessed and analyzed to obtain key medical history information.
4. The method according to claim 3, characterized in that: Through the preset prediction model, feature analysis is performed according to the key indicator data to obtain multi-indicator prediction features, including: Analyzing and extracting the key medical history information through a preset prediction model to obtain clinical information features related to pregnancy position prediction; By using high-throughput sequencing technology, the flora-related information is used to perform flora composition and diversity analysis and extraction, and obtain flora analysis features related to pregnancy position prediction; The microbial community analysis characteristics include at least two of abundance characteristics, diversity index characteristics, structural change characteristics and metabolic characteristics.
5. The method according to claim 1, characterized in that Predicting the position of pregnancy based on the multi-index prediction feature to obtain a pregnancy position prediction result includes: Analyze and evaluate the multi-indicator prediction features to obtain the score information corresponding to the key analysis items of the target object; The pregnancy position is predicted according to the score information to obtain a pregnancy position prediction result.
6. The method according to claim 5, characterized in that Analyze and evaluate the multi-indicator prediction features to obtain the score information corresponding to the key analysis items of the target object, including: Get the nomogram construction function; Through the nomogram construction function, the multi-index prediction characteristics are used for analysis to obtain score information, and a nomogram is constructed according to the score information for display.
7. The method according to any one of claims 1 to 6, characterized in that: After obtaining the score information, the method further includes: High-risk screening is performed based on the score information to obtain high-risk prediction information, and the high-risk prediction information is used to optimize the treatment plan for the target object in combination with the pregnancy position prediction result.
8. A pregnancy position prediction device based on a prediction model combined with vaginal flora indicators, characterized in that: include: A module for acquiring data to be analyzed, used to acquire a test report and medical history data of a target subject, wherein the test report at least includes vaginal flora test information and symptom-related information of the target subject; A data analysis and extraction module, used to perform data analysis and extraction based on the test report and the medical history data according to preset key analysis items to obtain key indicator data, wherein the key indicator data includes flora-related information and key medical history information related to the pregnancy position; A feature analysis module, used to perform feature analysis based on the key indicator data and the medical history data through a preset prediction model to obtain multi-indicator prediction features, wherein the multi-indicator prediction features include clinical information features and flora analysis features related to pregnancy position prediction; A prediction module is used to predict the position of pregnancy based on the multi-indicator prediction feature to obtain a pregnancy position prediction result; wherein the pregnancy position prediction result is used to optimize the treatment plan for the target object.
9. An electronic device, characterized in that: It includes a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other through the communication bus; Memory, used to store computer programs; The processor is used to implement the steps of the method for predicting the location of pregnancy based on the prediction model combined with vaginal flora indicators as described in any one of claims 1 to 7 when executing the program stored in the memory.
10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for predicting the location of pregnancy based on a prediction model combined with vaginal flora indicators as described in any one of claims 1 to 7 are implemented.