Risk event detection method and system based on antenatal care identity recognition

Abstract

The invention discloses a risk event detection method and system based on antenatal care identity recognition, and the method comprises the steps: extracting a historical treatment track of a patient from a preset medical information database according to identity information; mapping the historical doctor-seeing track into a preset geographic grid matrix, and generating a doctor-seeing activity thermal distribution diagram; based on the treatment activity thermodynamic distribution diagram, extracting associated crowds appearing in the same unit grid as the patient, and constructing a space-time association chain; performing forward comparison on the second biological characteristic information and pre-stored information; meanwhile, the first biological characteristic information is reversely compared with pre-stored information of the associated crowd; when a deviation value between the forward comparison result and the reverse comparison result exceeds a preset threshold value, calculating an aggregation degree feature of the space-time association chain; and judging whether the patient triggers risk early warning or not according to the aggregation degree characteristics. According to the invention, the real-time performance and accuracy of risk event detection are improved.

Description

Technical Field

[0001] The present application relates to the field of prenatal examination risk detection, and in particular to a risk event detection method and system based on prenatal examination identity recognition. Background Art

[0002] With the continuous development of medical technology, prenatal examinations have become an important part of pregnancy care. However, in actual applications, there are violations such as identity impersonation and information falsification. These behaviors not only disrupt the normal medical order, but may also trigger a series of potential legal and ethical risk events, causing adverse effects on social stability.

[0003] In the relevant technology, an identity authentication system based on biometric recognition can be used to solve the above problems. This method collects the patient's biometric information such as fingerprints and faces, and compares it with the pre-entered identity information, thereby ensuring the authenticity of the identity during the prenatal examination process and reducing the incidence of identity impersonation and information falsification.

[0004] However, when multiple people are involved in collaboration, it is difficult to detect highly concealed violations through a single biometric comparison, which reduces the real-time and accuracy of risk event detection. Summary of the invention

[0005] The present application provides a risk event detection method and system based on prenatal examination identity recognition, which are used to improve the real-time and accuracy of risk event detection.

[0006] In a first aspect, the present application provides a risk event detection method based on prenatal examination identity recognition, which obtains the first biometric information and identity information of the patient, wherein the first biometric information includes fingerprint information and facial feature information; According to the identity information, the patient's historical medical history is extracted from the preset medical information database, and the historical medical history includes the medical time sequence, the medical location sequence and the corresponding medical treatment item combination; Mapping the historical medical treatment trajectories into a preset geographic grid matrix to generate a heat distribution map of medical treatment activities. The preset geographic grid matrix divides the area where the medical institution is located into several unit grids; Based on the heat distribution map of medical activities, the associated population that appears in the same unit grid as the patients is extracted, and a spatiotemporal association chain is constructed. The spatiotemporal association chain records the temporal relationship and spatial migration pattern between the associated populations. Obtain the second biometric information of the associated group, and perform a positive comparison between the second biometric information and the pre-stored information; and perform a reverse comparison between the first biometric information and the pre-stored information of the associated group; When the deviation value between the forward comparison result and the reverse comparison result exceeds the preset threshold, calculate the aggregation degree feature of the spatio-temporal association chain, and the aggregation degree feature characterizes the convergence law of the associated population at different time points and different locations; Determine whether the patient triggers a risk warning according to the aggregation degree feature.

[0007] By adopting the above technical solution, by obtaining the biometric information and identity information of the patient, and combining the historical medical treatment trajectory to generate a heat map of medical treatment activities, the spatio-temporal activity law of the patient can be intuitively displayed. Based on this heat map, associated populations are extracted and a spatio-temporal association chain is constructed, which can reveal potential associated behavior patterns. The first biometric information is compared bidirectionally with the pre-stored information of the associated population, and the abnormal degree of the identity characteristics is judged by the deviation value of the comparison result. When an abnormality is found, by calculating the aggregation degree feature of the spatio-temporal association chain, the group behavior characteristics of the associated population can be quantitatively measured. This multi-dimensional analysis method combines biometric recognition, spatio-temporal trajectory analysis and quantification of group behavior characteristics. On the basis of finding an abnormality in a single biometric, the risk event is further verified through the behavior pattern of the associated population, improving the real-time performance and accuracy of risk identification.

[0008] Combined with some embodiments of the first aspect, in some embodiments, when the deviation value between the forward comparison result and the reverse comparison result exceeds the preset threshold, calculate the aggregation degree feature of the spatio-temporal association chain, specifically including: Extract the time position mark and geographical position mark of each associated population from the spatio-temporal association chain; Calculate the spatio-temporal distribution density of the associated population in each unit grid based on the time position mark and geographical position mark; Calculate the aggregation degree feature according to the spatio-temporal distribution density, and the aggregation degree feature includes the degree of population aggregation in the unit grid per unit time.

[0009] By adopting the above technical solution, by extracting the time position mark and geographical position mark of each associated population in the spatio-temporal association chain, calculating the spatio-temporal distribution density in each unit grid, and then obtaining the aggregation degree feature. By analyzing the degree of population aggregation in the unit grid per unit time, the spatial distribution law and time evolution characteristics of the associated population can be accurately described. Through precise mathematical modeling and quantitative analysis, the scientificity and reliability of risk event detection are improved, and the probability of misjudgment and missed judgment is reduced.

[0010] Combined with some embodiments of the first aspect, in some embodiments, determine whether the patient triggers a risk warning according to the aggregation degree feature, specifically including: Obtain the preset threshold of normal medical treatment behavior characteristics, and the preset threshold of normal medical treatment behavior characteristics includes the maximum aggregation degree threshold within the preset time period; Compare the aggregation degree feature with the preset threshold of normal medical visit behavior characteristics; When the aggregation degree feature exceeds the threshold of normal medical visit behavior characteristics, it is determined that the medical visitor triggers a risk warning; When the aggregation degree feature does not exceed the threshold of normal medical visit behavior characteristics, it is determined that the medical visitor does not trigger a risk warning.

[0011] By adopting the above technical solution, a preset threshold of normal medical visit behavior characteristics is introduced as the judgment benchmark. By comparing the actually observed aggregation degree feature with the threshold, the preset threshold includes the maximum aggregation degree threshold within a preset time period. This design takes into account the differences in the distribution of the medical visit population at different time periods. By comparing with the threshold, clear warning trigger conditions are obtained, making the judgment of risk events have a unified evaluation standard. This technical solution establishes a quantitative risk assessment system, improving the standardization and accuracy of risk warning.

[0012] Combined with some embodiments of the first aspect, in some embodiments, obtain the second biometric information of the associated population, and perform a positive comparison between the second biometric information and the pre-stored information, specifically including: Collect multiple groups of second biometric information of the associated population at different time points, and each group of second biometric information includes fingerprint information and facial feature information; Extract the pre-stored biometric information of the associated population from the preset biometric database, and the pre-stored biometric information includes fingerprint information and facial feature information collected historically; Calculate the similarity scores between multiple groups of second biometric information and the pre-stored biometric information respectively, and generate a positive comparison result.

[0013] By adopting the above technical solution, multiple groups of biometric information of the associated population are collected at different time points, and the similarity scores are calculated with the pre-stored biometric information, realizing dynamic identity feature comparison. The feature collection at multiple time points increases the time dimension of biometric data and can capture the subtle changes of biometric features over time. By calculating the similarity scores between each group of feature information and the pre-stored information, a quantitative evaluation index is established. This technical solution based on multiple groups of feature comparison improves the reliability of biometric recognition and reduces the occurrence probability of deception behavior using forged biometric features. Through dynamic feature collection and multi-dimensional similarity calculation, the accuracy and anti-counterfeiting ability of identity recognition are enhanced, providing more reliable biometric verification support for risk event detection.

[0014] Combined with some embodiments of the first aspect, in some embodiments, perform a reverse comparison between the first biometric information and the pre-stored information of the associated population, specifically including: Extract the set of feature points in the first biometric information, and the set of feature points includes fingerprint feature points and facial key points; Match the set of feature points with the pre-stored biometric information of the associated population to obtain the matching correspondence of the feature points; Calculate the Euclidean distance between the corresponding feature points based on the matching correspondence, and perform a weighted sum of the Euclidean distances of all feature points to obtain the reverse comparison result.

[0015] By adopting the above technical solution, by extracting the set of feature points in the first biometric information, performing feature matching with the pre-stored biometric information of the associated population to obtain the matching correspondence, and then calculating the Euclidean distance between the corresponding feature points based on the matching correspondence and performing a weighted sum to obtain the reverse comparison result, it is possible to accurately measure biometric features at the micro-feature level. The set of feature points includes fingerprint feature points and facial key points, and can perform comparisons from multiple biometric dimensions simultaneously, avoiding the problem that a single feature is easily forged. By calculating the Euclidean distance between feature points and performing a weighted sum, the importance of different feature points can be quantified, improving the accuracy of the comparison. This reverse comparison method based on feature points can identify behaviors of disguising by means such as makeup and wearing ornaments, reducing the risk of medical fraud. At the same time, this method can resist attacks of forged biometric features such as photos and videos, enhancing the security of identity authentication.

[0016] Combined with some embodiments of the first aspect, in some embodiments, after determining whether the patient triggers a risk warning according to the aggregation degree feature, the method further includes: When it is determined that the risk warning is triggered, obtain the current item information to be treated of the patient; Construct a project correlation network based on the current item information to be treated, and the project correlation network represents the degree of association between different treatment items; Classify and hierarchically control the current item to be treated according to the project correlation network.

[0017] By adopting the above technical solution, when the risk warning is triggered, obtain the current item information to be treated of the patient and construct a project correlation network. By using the network to represent the degree of association between different treatment items, it is possible to deeply analyze the project association pattern behind the suspicious treatment behavior. The project correlation network can reveal the internal connection between different treatment items and reflect whether there are abnormalities in certain project combinations. By performing classification and hierarchical control based on the project correlation network, it is possible to accurately identify high-risk treatment item combinations and avoid unreasonable treatment item collocations.

[0018] Combined with some embodiments of the first aspect, in some embodiments, classifying and hierarchically controlling the current item to be treated according to the project correlation network specifically includes: Calculate the key degree index of the current item to be treated in the project correlation network, and the key degree index includes the degree centrality and betweenness centrality of the project node; Divide the current items to be treated into core items and non-core items based on the criticality index; Set a mandatory waiting period for core items, and core items are prohibited from being executed during the mandatory waiting period; Perform grading processing on non-core items and set different verification mechanisms according to different levels.

[0019] By adopting the above technical solution, by calculating the criticality index of the current items to be treated in the project relevance network, including the degree centrality and betweenness centrality of project nodes, the importance and influence of each treatment item can be accurately evaluated. Based on the criticality index, the items are divided into core items and non-core items, and a mandatory waiting period is set for core items and a differential verification mechanism is implemented for non-core items, which can achieve precise hierarchical control. The degree centrality reflects the direct correlation degree of project nodes, and the betweenness centrality reflects the mediating role of project nodes. The combined use of these two indicators can comprehensively evaluate the critical degree of the project. By implementing the mandatory waiting period management for core items, the implementation rhythm of high-risk project combinations can be interrupted, and the risk of dangerous events can be reduced. The differential verification mechanism can take corresponding control measures according to the risk levels of non-core items, ensuring the control effect and avoiding excessive interference with normal medical treatment.

[0020] In a second aspect, an embodiment of the present application provides a risk event detection system based on antenatal care identity recognition. The risk event detection system based on antenatal care identity recognition includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code. The computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer-readable storage medium, including instructions, which when running on the system, enable the system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] In a fourth aspect, an embodiment of the present application provides a computer program product, which when running on the system, enables the system to execute the method described in any possible implementation manner in the first aspect.

[0023] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. The present application provides a risk event detection method based on antenatal care identity recognition. By obtaining the biometric information and identity information of the patient, and generating a heat map of medical visit activities in combination with the historical medical visit trajectory, it can intuitively display the spatio-temporal activity patterns of the patient. Based on this heat map, associated populations are extracted and a spatio-temporal association chain is constructed, which can reveal potential associated behavior patterns. The first biometric information is compared bidirectionally with the pre-stored information of the associated populations, and the degree of abnormality of the identity characteristics is judged through the deviation value of the comparison result. When an abnormality is found, the aggregation degree characteristics of the spatio-temporal association chain are calculated to quantitatively measure the group behavior characteristics of the associated populations. This multi-dimensional analysis method combines biometric recognition, spatio-temporal trajectory analysis, and quantification of group behavior characteristics. On the basis of discovering individual biometric abnormalities, risk events are further verified through the behavior patterns of the associated populations, improving the real-time performance and accuracy of risk identification.

[0024] 2. The present application provides a risk event detection method based on antenatal care identity recognition. Multiple groups of biometric information of the associated populations are collected at different time points, and the similarity scores are calculated by comparing with the pre-stored biometric information, realizing dynamic comparison of identity characteristics. The feature collection at multiple time points adds the time dimension to the biometric data, and can capture the subtle changes of biometric characteristics over time. By calculating the similarity scores of each group of feature information and the pre-stored information, a quantitative evaluation index is established. This technical solution based on multi-group feature comparison improves the reliability of biometric recognition and reduces the occurrence probability of deception using forged biometric characteristics. Through dynamic feature collection and multi-dimensional similarity calculation, the accuracy and anti-counterfeiting ability of identity recognition are enhanced, providing more reliable biometric verification support for risk event detection.

[0025] 3. The present application provides a risk event detection method based on antenatal care identity recognition. When a risk warning is triggered, the current to-be-visited item information of the patient is obtained and a project correlation network is constructed. By representing the association degree between different medical visit items through the network, the project association pattern behind the suspicious medical visit behavior can be deeply analyzed. The project correlation network can reveal the internal connections between different medical visit items and reflect whether there are abnormalities in certain item combinations. Based on the project correlation network, classified and hierarchical control is carried out, and high-risk medical visit item combinations can be accurately identified to avoid unreasonable medical visit item collocations. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is a flowchart of a risk event detection method based on antenatal care identity recognition in an embodiment of the present application.

[0027] Figure 2 is a flowchart of a hierarchical control method based on project correlation in an embodiment of the present application.

[0028] Figure 3 It is a schematic structural diagram of an entity device of a risk event detection system based on prenatal examination identity recognition provided by an embodiment of the present application. Detailed implementation manners

[0029] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. As used in the specification and appended claims of the present application, the singular forms "a", "an", "the", "above-mentioned", "said", and "this" are also intended to include the plural forms, unless clearly indicated to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations including one or more of the listed items.

[0030] Hereinafter, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0031] Next, a risk event detection method based on prenatal examination identity recognition in an embodiment of the present application will be described by using an example and in combination with Figure 1 , as follows: Please refer to Figure 1 , which is a flowchart of a risk event detection method based on prenatal examination identity recognition in an embodiment of the present application.

[0032] S101. Obtain the first biometric information and identity information of the patient; The system obtains the first biometric information and identity information of the patient. The first biometric information includes fingerprint information and facial feature information. In this step, the system first needs to obtain the first biometric information and identity information of the patient. The first biometric information may include, but is not limited to, fingerprint information, facial feature information, iris information, palmprint information, etc. The identity information may include personal information such as the patient's name, gender, age, ID number, medical insurance card number, etc. The system can obtain the above information through various methods, such as directly collecting through biometric collection devices, reading from hospital information systems or government identity information systems, etc.

[0033] Specifically, the system can set up fingerprint collectors and cameras at the registration desk of the hospital. When a patient registers for medical treatment, their fingerprint information and facial images are collected. At the same time, the system reads the patient's identity information from the identity documents provided by the patient (such as ID cards, medical insurance cards, etc.). The collected fingerprint images can be processed through image enhancement, feature extraction, etc. to generate fingerprint feature information; the facial images can be processed through face detection, key point localization, feature extraction, etc. to generate facial feature information. These biometric feature information and identity information can be associated and stored for convenient subsequent use.

[0034] S102. Extract the historical medical treatment trajectory of the patient from the preset medical information database according to the identity information; The system extracts the historical medical treatment trajectory of the patient from the preset medical information database according to the identity information. The historical medical treatment trajectory includes a time series of medical treatments, a sequence of medical treatment locations, and the corresponding combination of medical treatment items. In this step, the system needs to extract the historical medical treatment trajectory of the patient from the preset medical information database according to the patient's identity information. The historical medical treatment trajectory can include information such as a time series of medical treatments, a sequence of medical treatment locations, and the corresponding combination of medical treatment items. The preset medical information database can be an electronic medical record system built by the hospital, or a regional or national medical information sharing platform. The database should store the medical treatment records of the patient in each medical institution. These records can be sorted in chronological order to form a time series. The medical treatment location information can identify the medical institution where the medical treatment occurred, and the combination of medical treatment items records the specific medical behaviors such as examinations, treatments, medications, etc. involved in each medical treatment.

[0035] When specifically implemented, the system can use the unique identifiers such as the patient's ID number and medical insurance card number to retrieve the patient's historical medical treatment records in the preset medical information database. The retrieved medical treatment records can be sorted in chronological order to form a time series of medical treatments. The medical institution information included in each medical treatment record can be extracted to form a sequence of medical treatment locations corresponding to the time series. At the same time, the system can also extract item information such as examinations, treatments, medications, etc. from the medical treatment details of the medical treatment records to form a combination of medical treatment items corresponding to each medical treatment. The extracted time series of medical treatments, sequence of medical treatment locations, and the corresponding combination of medical treatment items together constitute the historical medical treatment trajectory of the patient.

[0036] In this step, you may face the problem of missing medical records due to incomplete records in the medical information database. To address this problem, the system can search multiple data sources when extracting medical records. For example, in addition to the hospital's self-built electronic medical record system, you can also search the medical insurance settlement system, regional medical information sharing platform, etc. Through the complementarity of multiple data sources, the patient's historical medical records can be restored as completely as possible. In addition, when certain fields of the medical record are missing (such as the name of the medical institution is missing), the system can fill in the missing fields based on experience. For example, the name of the medical institution is inferred based on the address of the medical institution. Through the above method, the system can obtain a relatively complete and coherent historical medical record, providing data support for subsequent analysis.

[0037] S103, mapping the historical medical consultation trajectory to a preset geographic grid matrix to generate a medical consultation activity heat distribution map; The system maps the historical medical treatment trajectory to the preset geographic grid matrix to generate a heat distribution map of medical treatment activities. The preset geographic grid matrix divides the area where the medical institution is located into several unit grids. In this step, the system needs to map the extracted historical medical treatment trajectory to the preset geographic grid matrix to generate a heat distribution map of medical treatment activities. The preset geographic grid matrix divides the area where the medical institution is located into several unit grids, and each unit grid represents a geographical area. The heat distribution map of medical treatment activities shows the level of medical treatment activity of patients in various geographical areas in an intuitive form. Through this step, the medical treatment trajectory can be combined with the geographic location information to explore the spatial pattern of medical treatment behavior.

[0038] In specific implementation, the system first needs to build a preset geographic grid matrix. The area can be gridded according to the appropriate granularity (such as 1km×1km) based on the geographical scope of the area where the medical institution is located. Each unit grid can be uniquely identified by its geographical coordinates or regional name. Then, the system maps the sequence of medical treatment locations in the historical medical treatment trajectory to the geographic grid matrix one by one. The unit grid where the medical institution is located can be determined by the address information or longitude and latitude coordinates of the medical institution, and the medical treatment record can be associated with the corresponding grid. A unit grid may correspond to multiple medical treatment records. Next, the system can count the number of medical treatments in each unit grid and calculate the medical treatment heat according to the number of medical treatments. The more medical treatments, the more frequent the medical treatment activities in the geographical area, and the higher the medical treatment heat. Finally, the system draws a thermal distribution map based on the medical treatment heat of each unit grid. The color or brightness can represent the level of medical treatment heat, and intuitively present the distribution pattern of medical treatment activities in the geographical space.

[0039] S104. Based on the heat distribution map of medical treatment activities, extract the associated people who appear in the same unit grid as the patients, and construct a spatiotemporal association chain; Based on the heat map of medical visit activities, the system extracts the associated population that appears in the same unit grid as the patient, and constructs a spatio-temporal association chain. The spatio-temporal association chain records the temporal relationship and spatial migration pattern among the associated population. In this step, the system needs to extract the associated population that visits in the same unit grid as the patient based on the heat map of medical visit activities, and construct an association chain that reflects the spatio-temporal relationship of the population. The associated population refers to other patients who appear in the same geographical area as the patient within the same time period. By analyzing the temporal and spatial connections between the patient and the associated population, some hidden medical visit behavior patterns can be discovered. The spatio-temporal association chain represents the temporal relationship and spatial migration trajectory among the associated population in the form of a graph structure.

[0040] In specific implementation, the system can traverse each unit grid in the heat map of medical visit activities to find all medical visit records falling into this grid. For each medical visit record, extract the corresponding patient ID and medical visit time. Then, the system can compare the medical visit times of different patients within this unit grid. If the time interval is less than a preset threshold (such as 1 day), these patients are classified as an associated population. The preset time threshold can be determined according to factors such as the incubation period of infectious diseases and the distribution law of hospital visit times. Through this process, the system can extract the associated population in each unit grid. Next, the system can construct a spatio-temporal association chain. Starting from the patient, connect the associated population to which it belongs in chronological order to form a time chain. At the same time, record the sequence of medical visit locations of each population member. The time chain reflects the temporal correlation of the population members, and the spatial sequence depicts the transfer trajectory of the population in the geographical space. Combine the spatio-temporal association chains of all associated populations to form a global spatio-temporal association network, which comprehensively reflects the complex association pattern of the population in the spatio-temporal dimension.

[0041] S105. Obtain the second biometric information of the associated population, and perform a positive comparison between the second biometric information and the pre-stored information; at the same time, perform a reverse comparison between the first biometric information and the pre-stored information of the associated population; The system obtains the second biometric information of the associated population and performs a positive comparison between the second biometric information and the pre-stored information. Specifically: collect multiple groups of second biometric information of the associated population at different time points, and each group of second biometric information includes fingerprint information and facial feature information; extract the pre-stored biometric information of the associated population from the preset biometric database, and the pre-stored biometric information includes the fingerprint information and facial feature information collected historically; calculate the similarity scores between multiple groups of second biometric information and the pre-stored biometric information respectively to generate a positive comparison result.

[0042] Meanwhile, the first biometric information is reversely compared with the pre-stored information of the associated population, which specifically includes: extracting the set of feature points in the first biometric information, where the set of feature points includes fingerprint feature points and facial key points; performing feature matching between the set of feature points and the pre-stored biometric information of the associated population to obtain the matching correspondence of the feature points; calculating the Euclidean distance between the corresponding feature points based on the matching correspondence, and performing weighted summation of the Euclidean distances of all feature points to obtain the reverse comparison result. In this step, the system needs to obtain the second biometric information of the associated population and perform a forward comparison with the pre-stored information; meanwhile, it also needs to reversely compare the first biometric information of the patient with the pre-stored information of the associated population. Biometric information can be information for identity recognition such as fingerprints and faces. Through forward comparison, it can be confirmed whether the identity of the associated population is consistent with the system's pre-stored information; through reverse comparison, it can be determined whether there is a duplicate identity between the patient and the associated population. The biometric comparison results can be used to analyze the authenticity and relevance of population identities.

[0043] In specific implementation, the system first needs to obtain the second biometric information of the associated population. When the members of the associated population seek medical treatment, their biometric information such as fingerprints and faces can be obtained through on-site collection or by retrieving hospital files. Since a person's biometric features may change over time, the system can collect multiple sets of second biometric information to improve the reliability of the comparison. The collected second biometric information is converted into a feature template through feature extraction for subsequent comparison. Next, the system extracts the pre-stored biometric information of the associated population from the preset biometric database as a reference sample for comparison. Then, the system uses corresponding comparison algorithms (such as fingerprint feature matching, face similarity calculation, etc.) to compare multiple sets of second biometric information with the pre-stored information respectively to obtain the similarity score for each comparison. By synthesizing each component score, the forward comparison result can be obtained. At the same time, the system also reversely compares the first biometric information of the patient with the pre-stored information of the associated population. Specifically, the set of feature points (such as fingerprint feature points, facial key points, etc.) in the first biometric information can be extracted, and then the set of feature points is subjected to feature matching with the pre-stored information to calculate the spatial distance between the feature points. By performing weighted summation of the distances of all feature points, the reverse comparison result can be obtained.

[0044] S106. When the deviation value between the forward comparison result and the reverse comparison result exceeds the preset threshold, calculate the aggregation degree feature of the spatio-temporal association chain; When the deviation value between the forward comparison result and the reverse comparison result exceeds the preset threshold, calculate the aggregation degree feature of the spatio-temporal association chain. The aggregation degree feature characterizes the convergence law of the associated population at different time points and different locations. Specifically: Extract the time position marker and geographical position marker of each associated population from the spatio-temporal association chain; Calculate the spatio-temporal distribution density of the associated population in each unit grid based on the time position marker and geographical position marker; Calculate the aggregation degree feature according to the spatio-temporal distribution density. The aggregation degree feature includes the degree of population aggregation in the unit grid per unit time.

[0045] In this step, the system determines whether to further calculate the aggregation degree feature of the spatio-temporal association chain by comparing the size relationship between the deviation value of the forward comparison result and the reverse comparison result and the preset threshold. The purpose of this step is to screen out the patients who may be at risk for further in-depth analysis. In addition to using the method of comparing the deviation value with the threshold, the system can also adopt other screening conditions. For example, set the confidence range of the forward comparison result and the reverse comparison result. When the confidence levels of both meet or do not meet the range at the same time, the calculation of the aggregation degree feature is performed.

[0046] Specifically, the system can implement this step through the following technology: First, the system calculates the difference between the forward comparison result and the reverse comparison result to obtain the deviation value; Then, compare the deviation value with the preset threshold. When the deviation value exceeds the preset threshold, trigger the calculation of the aggregation degree feature; Finally, the system extracts the time position marker and geographical position marker of each associated population from the spatio-temporal association chain, and calculates the spatio-temporal distribution density of the associated population in each unit grid based on these markers, and then obtains the aggregation degree feature. Among them, the aggregation degree feature can characterize the convergence law of the associated population at different time points and different locations, such as the degree of population aggregation in the unit grid per unit time, etc.

[0047] S107. Determine whether the patient triggers a risk warning according to the aggregation degree feature.

[0048] The system determines whether the patient triggers a risk warning according to the aggregation degree feature, which specifically includes: Obtain the preset normal medical behavior feature threshold, and the preset normal medical behavior feature threshold includes the maximum aggregation degree threshold within the preset time period; Compare the aggregation degree feature with the preset normal medical behavior feature threshold; When the aggregation degree feature exceeds the normal medical behavior feature threshold, determine that the patient triggers a risk warning; When the aggregation degree feature does not exceed the normal medical behavior feature threshold, determine that the patient does not trigger a risk warning.

[0049] In this step, the system determines whether a risk warning is triggered for the patient based on the aggregation feature calculated in the previous step. The purpose of this step is to identify patients with abnormal behaviors so that corresponding measures can be taken in a timely manner. In addition to directly comparing the aggregation feature with a preset threshold, the system can also adopt other judgment methods, such as using the aggregation feature as input and applying a machine learning model for risk prediction, etc.

[0050] Specifically, the system can implement this step through the following techniques: First, the system obtains the preset threshold of normal patient behavior features, which can include the maximum aggregation threshold within a preset time period, etc.; then, the aggregation feature is compared with the preset threshold. When the aggregation feature exceeds the threshold, it is determined that the risk warning is triggered for the patient; when the aggregation feature does not exceed the threshold, it is determined that the risk warning is not triggered for the patient.

[0051] In the above embodiment, by obtaining the biometric information and identity information of the patient and generating a heat map of the patient's activity in combination with the historical medical treatment trajectory, the spatio-temporal activity pattern of the patient can be intuitively displayed. Based on this heat map, associated people are extracted and a spatio-temporal association chain is constructed, which can reveal potential associated behavior patterns. The first biometric information is compared bidirectionally with the pre-stored information of the associated people, and the degree of abnormality of the identity characteristics is judged through the deviation value of the comparison result. When an abnormality is found, by calculating the aggregation feature of the spatio-temporal association chain, the group behavior characteristics of the associated people can be quantitatively measured. This multi-dimensional analysis method combines biometric recognition, spatio-temporal trajectory analysis, and quantification of group behavior characteristics. On the basis of finding an abnormality in a single biometric feature, the risk event is further verified through the behavior pattern of the associated people, improving the real-time performance and accuracy of risk identification.

[0052] After completing the above detection process of risk events, in order to further strengthen the control of suspicious patient behaviors, this application also provides a hierarchical control method based on project relevance. This method constructs a refined control mechanism from the dimension of the medical treatment project on the basis of the system determining that a risk warning is triggered, and realizes differential management of projects with different risk levels. The following combines Figure 2 to describe a hierarchical control method based on project relevance in the embodiments of this application: Please refer to Figure 2 which is a schematic flowchart of a hierarchical control method based on project relevance in the embodiments of this application.

[0053] S201. When it is determined that a risk warning is triggered, obtain the current to-be-treated project information of the patient; In this step, after the system determines to trigger a risk warning, it further obtains the information of the medical treatment items that the patient is currently about to undergo. The purpose of this step is to provide a necessary data basis for subsequent hierarchical control. In addition to directly obtaining the current medical treatment item information to be seen, the system can also adopt other data collection methods, such as extracting medical treatment items similar to the current situation from the patient's historical medical treatment records, or inferring the possible medical treatment items based on the patient's personal information (such as age, gender, etc.).

[0054] Specifically, the system can implement this step through the following techniques: First, the system receives a trigger signal from the risk warning module; then, the system accesses the information management system of the medical institution to obtain the current medical treatment plan and appointment information of the patient; finally, the system extracts specific medical treatment items from the obtained information, such as physical examination items, treatment items, surgical items, etc., and stores them in a structured data format for subsequent analysis and processing.

[0055] S202. Construct a project correlation network based on the current medical treatment item information to be seen; The system constructs a project correlation network based on the current medical treatment item information to be seen, and the project correlation network represents the degree of association between different medical treatment items. In this step, the system constructs a project correlation network based on the current medical treatment item information obtained in the previous step. The purpose of this network is to depict the degree of association between different medical treatment items and provide a basis for subsequent hierarchical control. In addition to directly constructing the network based on the current medical treatment item to be seen, the system can also consider introducing more background information, such as the patient's historical medical treatment items, common medical treatment item combinations of other patients, etc., to construct a more comprehensive and accurate correlation network.

[0056] Specifically, the system can implement this step through the following techniques: First, the system abstracts each medical treatment item as a node in the network; then, the system calculates the correlation between different items as the edge connecting the nodes, and the correlation can be defined based on factors such as the co-occurrence frequency of the items, medical domain knowledge, etc.; finally, the system constructs a weighted undirected graph, that is, the project correlation network, according to the calculated correlation values.

[0057] S203. Calculate the key index of the current medical treatment item to be seen in the project correlation network; The system calculates the key - degree indicators of the current item to be treated in the item relevance network. The key - degree indicators include the degree centrality and betweenness centrality of the item nodes. In this step, based on the construction of the item relevance network, the system further calculates the key - degree indicators of the current item to be treated. The purpose of this indicator is to quantitatively evaluate the importance of each item in the entire network and provide a basis for subsequent item classification. In addition to using common centrality indicators (such as degree centrality, betweenness centrality, etc.), the system can also design other key - degree evaluation methods according to actual needs, such as considering factors like the medical importance of the item and the risk level of the item.

[0058] Specifically, the system can implement this step through the following techniques: First, for each node in the item relevance network (i.e., each item to be treated), the system calculates its degree centrality, which is the number of connections between this node and other nodes; then, the system calculates the betweenness centrality of each node, which is the frequency of this node appearing on the shortest paths between other nodes; finally, the system comprehensively evaluates the key - degree of each node based on the degree centrality and betweenness centrality. Nodes with high key - degree indicate that they play an important central role in the network.

[0059] S204. Classify the current item to be treated into core items and non - core items based on the key - degree indicators; In this step, the system classifies the current item to be treated into core items and non - core items according to the key - degree indicators calculated in the previous step. The purpose of this step is to set differentiated control strategies for items of different importance. In addition to using a simple threshold - based classification method, the system can also adopt other classification algorithms, such as clustering algorithms, decision trees, etc., to achieve a more refined and reasonable item classification.

[0060] Specifically, the system can implement this step through the following techniques: First, the system sets a key - degree threshold, which can be determined according to empirical values or the statistical distribution of historical data; then, for each item to be treated, the system compares its key - degree indicator with the threshold. If it is greater than the threshold, the item is classified as a core item; otherwise, it is classified as a non - core item; finally, the system outputs two sets of core items and non - core items as the objects for subsequent control.

[0061] S205. Set a mandatory waiting period for core items, and core items are prohibited from being executed during the mandatory waiting period; In this step, the system sets a mandatory waiting period for the identified core items, and the execution of core items is suspended during this period. The purpose of this step is to reserve a time window for subsequent risk verification and intervention by delaying the progress of core items. In addition to simply setting a waiting period of a fixed length, the system can also dynamically adjust the duration of the waiting period according to the specific situation of the item to balance the requirements of risk control and treatment efficiency.

[0062] Specifically, the system can implement this step through the following technologies: first, the system obtains a preset mandatory waiting period from the risk management knowledge base, such as 24 hours; then, the system sends instructions to the medical institution's information system to set an execution prohibition mark for all the divided core projects, and indicates the deadline for prohibiting execution; finally, the system monitors the status of the core projects, and once the end time of the mandatory waiting period is reached, the execution prohibition mark is automatically removed, allowing the core projects to re-enter the execution process.

[0063] S206. Non-core projects are graded and differentiated verification mechanisms are set up according to different levels.

[0064] In this step, the system further grades non-core projects and sets verification and control measures of different strengths according to the importance level of the projects. The purpose of this step is to minimize intervention in non-core projects and improve the overall efficiency of the medical process while ensuring that risks are controllable. In addition to adopting a centralized hierarchical verification solution, the system can also explore other control modes, such as delegating part of the verification work to various business links to achieve distributed multi-level verification.

[0065] Specifically, the system can implement this step through the following technologies: first, the system divides non-core projects into multiple levels according to the criticality indicators of the projects, such as level 2 non-core, level 3 non-core, etc.; then, the system sets corresponding verification rules and processes for each level. The higher the level of the project, the higher the verification intensity, which may require more review steps and stricter access conditions; finally, the system triggers the corresponding verification link according to the level of the project, and only allows the project to enter the execution stage after the verification is passed.

[0066] In the above embodiment, when the risk warning is triggered, the patient's current waiting medical project information is obtained and a project correlation network is constructed. By characterizing the degree of correlation between different medical projects through the network, the project correlation pattern behind the suspicious medical behavior can be deeply analyzed. The project correlation network can reveal the intrinsic connection between different medical projects and reflect whether there are abnormalities in certain project combinations. Classification and grading management based on the project correlation network can accurately identify high-risk medical project combinations and avoid unreasonable medical project combinations.

[0067] The following describes the system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the physical device structure of a risk event detection system based on prenatal examination identity recognition provided in an embodiment of the present application.

[0068] It should be noted that Figure 3The structure of the system shown is only an example and should not impose any limitation on the functions and scope of use of the embodiments of the present invention.

[0069] As Figure 3 shown, the system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 302 or the program loaded from the storage section 308 into the Random Access Memory (RAM) 303, such as executing the method in the above embodiments. In the RAM 303, various programs and data required for the system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. An Input / Output (I / O) interface 305 is also connected to the bus 304.

[0070] The following components are connected to the I / O interface 305: an input section 306 including a camera, an infrared sensor, etc.; an output section 307 including a Liquid Crystal Display (LCD) and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as required. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as required so that a computer program read from it can be installed into the storage section 308 as required.

[0071] Specifically, according to the embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments of the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the Central Processing Unit (CPU) 301, various functions defined in the present invention are executed.

[0072] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of a computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above.

[0073] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram can represent a module, a program segment, or a part of code, and the above module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram or flowchart, and the combination of blocks in the block diagram or flowchart, can be implemented by a dedicated hardware-based system for performing the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0074] As another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or may exist alone without being assembled into the system. The above storage medium carries one or more computer programs, and when the one or more computer programs are executed by a processor of a system, the system implements the method provided in the above embodiments.

[0075] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0076] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", or "after...", or "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" can be interpreted as "if determining...", or "in response to determining...", or "when detecting (the stated condition or event)", or "in response to detecting (the stated condition or event)".

[0077] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wire (such as coaxial cable, optical fiber, digital subscriber line) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access, or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid-state drive), etc.

[0078] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by relevant hardware instructed by a computer program. This program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The foregoing storage medium includes various media that can store program codes, such as ROM, random access memory (RAM), magnetic disks, or optical discs.

Claims

1. A risk event detection method based on prenatal examination identity recognition, characterized in that Including: Obtain the first biometric information and identity information of the patient, where the first biometric information includes fingerprint information and facial feature information; According to the identity information, extract the historical medical treatment trajectory of the patient from a preset medical information database, where the historical medical treatment trajectory includes a sequence of medical treatment times, a sequence of medical treatment locations, and their corresponding combinations of medical treatment items; Map the historical medical treatment trajectory into a preset geographical grid matrix to generate a heat distribution map of medical treatment activities, where the preset geographical grid matrix divides the area where the medical institution is located into several unit grids; Based on the heat distribution map of medical treatment activities, extract the associated population that appears in the same unit grid as the patient, and construct a spatio-temporal association chain, where the spatio-temporal association chain records the temporal relationship and spatial migration pattern between the associated populations; Obtain the second biometric information of the associated population, and perform a forward comparison between the second biometric information and the pre-stored information; at the same time, perform a reverse comparison between the first biometric information and the pre-stored information of the associated population; When the deviation value between the forward comparison result and the reverse comparison result exceeds a preset threshold, calculate the aggregation degree feature of the spatio-temporal association chain, where the aggregation degree feature characterizes the convergence law of the associated population at different time points and different locations; Determine whether the patient triggers a risk warning according to the aggregation degree feature.

2. The method according to claim 1, wherein The step of calculating the aggregation degree feature of the spatio-temporal association chain when the deviation value between the forward comparison result and the reverse comparison result exceeds a preset threshold specifically includes: Extract the time position mark and geographical position mark of each associated population from the spatio-temporal association chain; Calculate the spatio-temporal distribution density of the associated population in each unit grid based on the time position mark and the geographical position mark; Calculate the aggregation degree feature according to the spatio-temporal distribution density, where the aggregation degree feature includes the degree of population aggregation in the unit grid per unit time.

3. The method according to claim 1, wherein The step of determining whether the patient triggers a risk warning according to the aggregation degree feature specifically includes: Obtain a preset normal medical treatment behavior feature threshold, where the preset normal medical treatment behavior feature threshold includes a maximum aggregation degree threshold within a preset time period; Compare the aggregation degree feature with the preset normal medical treatment behavior feature threshold; When the aggregation degree feature exceeds the normal medical treatment behavior feature threshold, determine that the patient triggers a risk warning; When the aggregation degree feature does not exceed the normal medical treatment behavior feature threshold, determine that the patient does not trigger the risk warning.

4. The method according to claim 1, wherein The step of obtaining the second biometric information of the associated population and performing a forward comparison between the second biometric information and the pre-stored information specifically includes: Collect multiple groups of second biometric information of the associated population at different time points, where each group of second biometric information includes fingerprint information and facial feature information; Extract the pre-stored biometric information of the associated population from a preset biometric database, where the pre-stored biometric information includes historically collected fingerprint information and facial feature information; Calculate the similarity scores between the multiple groups of second biometric information and the pre-stored biometric information respectively, and generate a forward comparison result.

5. The method according to claim 1, wherein Performing reverse comparison between the first biometric information and the pre-stored information of the associated population specifically includes: Extracting a set of feature points from the first biometric information, where the set of feature points includes fingerprint feature points and facial key points; Performing feature matching between the set of feature points and the pre-stored biometric information of the associated population to obtain a matching correspondence of the feature points; Calculating the Euclidean distance between corresponding feature points based on the matching correspondence, and performing weighted summation of the Euclidean distances of all the feature points to obtain a reverse comparison result.

6. The method according to claim 1, wherein After determining whether the patient triggers a risk warning according to the aggregation degree feature, the method further includes: When it is determined that the risk warning is triggered, obtaining the current information of the items to be treated for the patient; Constructing a project correlation network based on the current information of the items to be treated, where the project correlation network represents the association degree between different treatment items; Performing classified and graded control on the current items to be treated according to the project correlation network.

7. The method according to claim 6, wherein Performing classified and graded control on the current items to be treated according to the project correlation network specifically includes: Calculating the key degree index of the current items to be treated in the project correlation network, where the key degree index includes the degree centrality and betweenness centrality of the project nodes; Dividing the current items to be treated into core items and non-core items based on the key degree index; Setting a mandatory waiting period for the core items, and prohibiting the execution of the core items during the mandatory waiting period; Performing grading processing on the non-core items, and setting a differentiated verification mechanism according to different levels.

8. A risk event detection system based on prenatal examination identity recognition, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the system to execute the method according to any one of claims 1-7.

9. A computer-readable storage medium, comprising instructions, characterized in that, When the instructions run on the system, enabling the system to execute the method according to any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product runs on the system, enabling the system to execute the method according to any one of claims 1-7.

Images