Dental pulp regeneration postoperative curative effect dynamic evaluation system and evaluation method

Abstract

The invention relates to the technical field of dental pulp treatment, and discloses a dental pulp regeneration postoperative curative effect dynamic evaluation system and method, and the system comprises a data standardization module, a correlation analysis module, a map construction module, a mode recognition module, a trend research and judgment module and a report generation module. The method comprises the following steps: performing feature scale normalization on multi-modal curative effect monitoring data of a patient to obtain standardized time series data, constructing an association relationship matrix through phase synchronization analysis, and constructing a dynamic data association map by taking a time point as a node and association strength as an edge weight; performing change pattern recognition on the topological clustering structure in the dynamic data association map to obtain biological pattern evolution characteristics; performing prognosis trend study and judgment on the dental pulp regeneration process of the patient based on biological model evolution characteristics to obtain a comprehensive curative effect evaluation conclusion; according to the comprehensive curative effect evaluation conclusion, generating a dental pulp regeneration curative effect evaluation report; the efficiency of dental pulp regeneration curative effect evaluation can be improved.

Description

Technical Field

[0001] This invention relates to the field of endodontic treatment technology, and in particular to a dynamic evaluation system and method for the efficacy of endodontic regeneration surgery. Background Technology

[0002] Postoperative efficacy assessment of pulp regeneration requires the integration of multimodal monitoring data. Existing technologies are often limited to isolated analysis of single-dimensional data or only conduct static time-point assessments, making it difficult to capture the dynamic correlation and evolution characteristics of various parameters during pulp repair. This results in a lack of systematic and forward-looking judgment of efficacy, which is prone to assessment bias and cannot fully reflect the patient's actual recovery status.

[0003] Traditional assessment methods rely on manual data processing and result analysis, which is cumbersome and highly subjective. This not only consumes a lot of manpower and time, but also makes it difficult to quickly extract core information from multimodal data, thus failing to provide accurate evidence for clinical diagnosis and treatment in a timely manner. Consequently, it affects the optimization and adjustment of subsequent treatment plans and restricts the overall quality of diagnosis and treatment services. Therefore, how to improve the efficiency of efficacy assessment after pulp regeneration has become an urgent problem to be solved. Summary of the Invention

[0004] This invention provides a dynamic evaluation system and method for the efficacy of pulp regeneration surgery to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides a dynamic evaluation system for the efficacy of pulp regeneration surgery, characterized in that the system includes a data standardization module, a correlation analysis module, an atlas construction module, a pattern recognition module, a trend analysis module, and a report generation module, wherein:

[0006] The data standardization module is used to normalize the feature scale of the patient's multimodal efficacy monitoring data to obtain the patient's standardized time-series data.

[0007] The correlation analysis module is used to perform phase synchronization analysis on the temporal correlation between parameters in the standardized time series data in order to construct the correlation matrix of the patient.

[0008] The graph construction module is used to construct a dynamic data association graph of the patient based on the association matrix, using time points in the standardized time series data as nodes and association strength in the association matrix as edge weights.

[0009] The pattern recognition module is used to perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient.

[0010] The trend analysis module is used to analyze the prognostic trend of the patient's pulp regeneration process based on the biological pattern evolution characteristics, and to obtain a comprehensive efficacy evaluation conclusion for the patient.

[0011] The report generation module is used to generate a pulp regeneration efficacy evaluation report for the patient based on the comprehensive efficacy evaluation conclusion.

[0012] In a preferred embodiment, when the data standardization module performs feature scale normalization on the patient's multimodal efficacy monitoring data to obtain the patient's standardized time-series data, it is specifically used for:

[0013] Time axis registration was performed on the multimodal efficacy monitoring data of the patients to obtain the registered data of the patients;

[0014] The registered data is smoothed for outliers to obtain the patient's clean time-series data.

[0015] Key feature parameters reflecting the pulp regeneration status are extracted from the clean time-series data to obtain the patient's initial feature set;

[0016] The initial feature set is parameterized dimensionlessly to obtain the standardized time-series data of the patient.

[0017] In a preferred embodiment, when the correlation analysis module performs phase synchronization analysis on the temporal correlations between parameters in the standardized time-series data to construct the correlation matrix for the patient, it is specifically used for:

[0018] Based on the time series length of the standardized time series data, a time window of the time series length is determined, and the window time series data within the time window is extracted;

[0019] The instantaneous phases between the monitoring parameters in the window time series data are analyzed, and the instantaneous phases are unwrapped to obtain the phase difference sequence of the instantaneous phases;

[0020] Within the time window, based on the stability of the phase difference sequence, the stability of the phase synchronization between the monitoring parameters is evaluated to obtain a quantitative index of the synchronization of the monitoring parameters.

[0021] The synchronicity quantification index is reconstructed into a matrix to obtain the correlation matrix of the patient.

[0022] In a preferred embodiment, when the atlas construction module constructs a dynamic data association atlas of the patient based on the association matrix, using time points in the standardized time-series data as nodes, and the association strength in the association matrix as edge weights, it is specifically used for:

[0023] Node mapping is performed on the time points in the standardized time series data to obtain the node set of the time points;

[0024] The elements representing the correlation strength between different time points in the correlation matrix are connected topologically to obtain the connection edges of the correlation matrix.

[0025] The node set and the connecting edges are integrated into a graph topology to obtain the initial weighted graph of the patient.

[0026] Based on the weight distribution characteristics of the connecting edges, the initial weighted graph is subjected to salient connection screening to obtain the patient's core topology.

[0027] The core topology is visualized and encoded to obtain a dynamic data association map of the patient.

[0028] In a preferred embodiment, when the graph construction module performs salient connection screening on the initial weighted graph based on the weight distribution characteristics of the connection edges to obtain the patient's core topology, it is specifically used for:

[0029] The weight distribution of the connecting edges is analyzed to obtain the distribution contour of the connecting edges;

[0030] The saliency threshold of the distribution profile is derived to obtain the screening threshold of the distribution profile;

[0031] Based on the filtering threshold, weak connections are filtered out from the connecting edges of the initial weighted graph to obtain a simplified graph of the initial weighted graph.

[0032] By removing isolated nodes from the simplified graph, the candidate core topology of the simplified graph is obtained;

[0033] Based on the screening threshold, the connectivity strength of the candidate core topology is verified to obtain the patient's core topology.

[0034] In a preferred embodiment, when the pattern recognition module performs change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient, it is specifically used for:

[0035] Extract the temporal slice structure from the dynamic data association graph to obtain the topological snapshot sequence of the dynamic data association graph;

[0036] Structural continuity is evaluated on the topological snapshot sequence to obtain the stable community evolution path of the topological snapshot sequence;

[0037] Steady-state structure identification is performed on the stable evolution path of the community to obtain the key steady-state topological patterns of the stable evolution path of the community;

[0038] By performing feature mapping on the key steady-state topological patterns, the biological pattern evolution characteristics of the patient can be obtained.

[0039] In a preferred embodiment, when the trend analysis module performs a prognostic trend analysis of the patient's pulp regeneration process based on the biological pattern evolution characteristics to obtain a comprehensive efficacy evaluation conclusion for the patient, it is specifically used for:

[0040] Event detection is performed on the topological pattern transition events in the biological pattern evolution features to obtain the evolutionary events of the biological pattern evolution features;

[0041] Analyze the temporal distribution patterns and intensity variation trends of the evolutionary events to obtain the evolutionary patterns of the events;

[0042] Based on the aforementioned evolutionary pattern, the healing process of the patient is deduced within the current observation window to obtain the patient's regenerative prognosis path;

[0043] The similarity index of the regenerative prognosis path is obtained by comparing it with the typical healing pattern library of the patient.

[0044] Based on the aforementioned consistency index, the efficacy level of the patient's pulp regeneration process was determined, and a comprehensive efficacy evaluation conclusion for the patient was obtained.

[0045] In a preferred embodiment, when the trend analysis module analyzes the temporal distribution pattern and intensity change trend of the evolutionary events to obtain the evolutionary pattern of the events, it is specifically used for:

[0046] Extract the associated time points and association strengths of events in the evolutionary events;

[0047] Based on the associated time points, a temporal clustering analysis is performed on the temporal distribution pattern of the evolutionary events to obtain the temporal distribution law of the evolutionary events;

[0048] Based on the correlation strength, the evolution direction of the intensity change pattern of the evolutionary event is identified to obtain the intensity change trend of the evolutionary event;

[0049] The temporal distribution pattern and the intensity change trend are integrated into the evolutionary pattern of the evolutionary event.

[0050] In a preferred embodiment, when the report generation module generates a pulp regeneration efficacy assessment report for the patient based on the comprehensive efficacy assessment conclusion, it is specifically used for:

[0051] The efficacy level and prognostic direction in the comprehensive efficacy assessment conclusion are taken as the core conclusions for the patient, and the key stages in the biological pattern evolution characteristics are taken as the supporting evidence for the patient.

[0052] The core conclusions and supporting evidence were compiled into a report, resulting in a draft report for the patient.

[0053] In the draft report, the dynamic data correlation map and the regeneration prognosis path are visualized and integrated to obtain the patient's pulp regeneration efficacy evaluation report.

[0054] To address the above problems, this invention also provides a method for dynamic evaluation of the efficacy of pulp regeneration surgery, the method comprising:

[0055] S1. Perform feature scale normalization on the multimodal efficacy monitoring data of the patients to obtain the standardized time-series data of the patients;

[0056] S2. Perform phase synchronization analysis on the temporal correlation between parameters in the standardized time series data to construct the correlation matrix of the patient;

[0057] S3. Based on the correlation matrix, using the time points in the standardized time series data as nodes and the correlation strength in the correlation matrix as edge weights, construct the dynamic data correlation graph of the patient;

[0058] S4. Perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient;

[0059] S5. Based on the evolutionary characteristics of the biological model, the prognostic trend of the patient's pulp regeneration process is analyzed to obtain the comprehensive efficacy evaluation conclusion of the patient;

[0060] S6. Based on the comprehensive efficacy evaluation conclusion, generate the pulp regeneration efficacy evaluation report for the patient.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] 1. This invention utilizes a data standardization module to perform time-axis registration, outlier smoothing, and dimensionless processing on multimodal therapeutic efficacy monitoring data, generating standardized time-series data that ensures data quality and comparability. Through phase synchronization analysis in the correlation analysis module, it accurately captures temporal correlations between parameters and constructs a correlation matrix, enabling in-depth data correlation mining. Furthermore, the graph construction module integrates and generates dynamic data correlation graphs using time points as nodes and correlation strength as weights, intuitively presenting the core topological structure and facilitating rapid understanding of the inherent correlation patterns within the data.

[0063] 2. This invention extracts topological evolution features through a pattern recognition module to accurately depict the evolutionary trajectory of biological patterns; it analyzes evolutionary event patterns and deduces regeneration prognostic paths based on a trend analysis module, and quantifies efficacy levels by combining consistency indicators to ensure scientific and accurate assessment conclusions; through a report generation module, it integrates core conclusions, supporting evidence, and visualization maps to generate a comprehensive and standardized assessment report, providing accurate reference for clinical diagnosis and treatment, and significantly improving the systematicness, accuracy, and efficiency of postoperative efficacy assessment after pulp regeneration. Attached Figure Description

[0064] Figure 1 This is a system architecture diagram of a dynamic evaluation system for postoperative efficacy of pulp regeneration provided in an embodiment of the present invention;

[0065] Figure 2 This is a flowchart illustrating a method for dynamic evaluation of the efficacy of pulp regeneration surgery according to an embodiment of the present invention.

[0066] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0067] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0069] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0070] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0071] In practice, the server-side equipment deployed in the dynamic evaluation system for post-pulp regeneration efficacy may consist of one or more devices. This system can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node to provide dynamic evaluation of post-pulp regeneration efficacy to various users. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage various users. Or, it can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more devices configured to provide dynamic evaluation of post-pulp regeneration efficacy to various users.

[0072] In terms of implementation, the dynamic evaluation system for post-pulp regeneration efficacy and the user terminal are mutually compatible. That is, if the dynamic evaluation system for post-pulp regeneration efficacy is implemented as an application installed on a cloud service platform, then the user terminal is a client that establishes a communication connection with the application; or if the dynamic evaluation system for post-pulp regeneration efficacy is implemented as a website, then the user terminal is implemented as a webpage; or if the dynamic evaluation system for post-pulp regeneration efficacy is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0073] like Figure 1 The figure shown is a system architecture diagram of a dynamic evaluation system for postoperative efficacy of pulp regeneration provided in an embodiment of the present invention.

[0074] The dynamic evaluation system 100 for postoperative efficacy of pulp regeneration described in this invention can be located on a cloud server. In terms of implementation, it can be one or more service devices, or an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the functions implemented, the dynamic evaluation system 100 for postoperative efficacy of pulp regeneration may include a data standardization module 101, a correlation analysis module 102, a graph construction module 103, a pattern recognition module 104, a trend analysis module 105, and a report generation module 106. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0075] In this embodiment of the invention, in the dynamic evaluation system for postoperative efficacy of pulp regeneration, each of the above modules can be implemented independently and can call other modules. Here, "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. In the dynamic evaluation system for postoperative efficacy of pulp regeneration provided by this embodiment of the invention, the applicable scope of the system architecture can be adjusted by adding modules and directly calling them without modifying the program code, achieving cluster-based horizontal expansion to quickly and flexibly expand the dynamic evaluation system for postoperative efficacy of pulp regeneration. In practical applications, the above modules can be set in the same device or different devices, or they can be set in a virtual device, such as a service instance in a cloud server.

[0076] The following describes the components and workflow of the dynamic evaluation system for post-pulp regeneration surgery efficacy, using specific examples:

[0077] The data standardization module 101 is used to normalize the feature scale of the patient's multimodal efficacy monitoring data to obtain the patient's standardized time-series data.

[0078] In this embodiment of the invention, when the data standardization module performs feature scale normalization on the multimodal efficacy monitoring data of the patient to obtain the standardized time-series data of the patient, it is specifically used for:

[0079] Time axis registration was performed on the multimodal efficacy monitoring data of the patients to obtain the registered data of the patients;

[0080] The registered data is smoothed for outliers to obtain the patient's clean time-series data.

[0081] Key feature parameters reflecting the pulp regeneration status are extracted from the clean time-series data to obtain the patient's initial feature set;

[0082] The initial feature set is parameterized dimensionlessly to obtain the standardized time-series data of the patient.

[0083] Multimodal efficacy monitoring data of patients were analyzed, encompassing various sources including imaging, physiological, and biochemical indicators. Using the time of the patient's first post-pulp regeneration surgery efficacy monitoring as a unified time reference, fixed time nodes were established every 12 hours. The time difference between the actual acquisition time of each monitoring data type and each fixed time node was calculated. The actual acquisition time was mapped to the fixed time node with the smallest time difference. If no corresponding data was acquired at a fixed time node, a null value was marked at that time node. This method ensured precise alignment of all multimodal data on the same timeline, ultimately forming the registered data.

[0084] Pre-operative baseline data of the same type collected before the patient's pulp regeneration surgery were retrieved. Combined with the clinically defined normal fluctuation range for pulp-related monitoring data (80% to 120% of the pre-operative baseline data for a given monitoring indicator), each data value in the registered data was compared and verified. Data values ​​below 80% or above 120% of the pre-operative baseline data were considered outliers. For each identified outlier, the most recent normal data value before and after the identified time point was found, and the arithmetic mean of these two normal data values ​​was calculated. This average was then used to replace the corresponding outlier. After processing all outliers, clean time-series data was generated.

[0085] Key feature parameter categories reflecting the pulp regeneration status are pre-defined, specifically including four categories: pulp tissue calcification-related data, periodontal ligament integrity-related data, pulp blood flow-related data, and inflammation-related biochemical indicators. Each data item in the clean time-series data is categorized, and all data belonging to these four pre-defined categories are selected. The data is then organized according to the separate classification of each feature parameter category. During the classification process, all time-series data corresponding to each parameter category are arranged sequentially according to time nodes to ensure the completeness and absence of time-series data for each parameter category, ultimately forming the initial feature set.

[0086] For each key feature parameter in the initial feature set, a clinically recognized effective value range was set for that parameter. Specifically, the effective range for pulp blood flow-related data was set to 0.5 to 5.0 mL / min, and the effective range for inflammation-related biochemical indicators was set to 0 to 20 ng / mL. Other key feature parameters also had corresponding effective value ranges set according to clinical standards. For each data value under each parameter category, a fixed transformation method was used: subtract the minimum value of the corresponding parameter's effective range from the data value, divide the result by the difference between the maximum and minimum values ​​of the parameter's effective range, and then multiply the result by 100. This transformation converted all key feature parameter data of different categories into values ​​between 0 and 100. After all parameters were transformed, they were integrated according to time nodes and parameter categories to form standardized time-series data.

[0087] The beneficial effects include: precise alignment of multimodal data along the timeline by clearly defining the starting point as the patient's first postoperative efficacy monitoring and setting fixed time nodes every 12 hours; defining outliers as the normal fluctuation range of 80% to 120% of the preoperative baseline data and replacing them with the arithmetic mean of adjacent normal data values, effectively ensuring data integrity; pre-setting four categories of key feature parameters to ensure the targeting and effectiveness of feature extraction; and setting the effective value range of each parameter according to clinically recognized standards and achieving dimensionless parameters through a fixed conversion method, making different categories of data comparable. The resulting standardized time-series data is characterized by time uniformity, data reliability, feature focus, and numerical comparability, providing high-quality, directly applicable data support for subsequent correlation matrix construction and the entire efficacy evaluation process, effectively ensuring the accuracy and repeatability of the evaluation results.

[0088] The correlation analysis module 102 is used to perform phase synchronization analysis on the temporal correlation relationship between parameters in the standardized time series data in order to construct the correlation matrix of the patient.

[0089] In this embodiment of the invention, when the correlation analysis module performs phase synchronization analysis on the temporal correlation relationships between parameters in the standardized time-series data to construct the correlation matrix of the patient, it is specifically used for:

[0090] Based on the time series length of the standardized time series data, a time window of the time series length is determined, and the window time series data within the time window is extracted;

[0091] The instantaneous phases between the monitoring parameters in the window time series data are analyzed, and the instantaneous phases are unwrapped to obtain the phase difference sequence of the instantaneous phases;

[0092] Within the time window, based on the stability of the phase difference sequence, the stability of the phase synchronization between the monitoring parameters is evaluated to obtain a quantitative index of the synchronization of the monitoring parameters.

[0093] The synchronicity quantification index is reconstructed into a matrix to obtain the correlation matrix of the patient.

[0094] The total number of data points in the standardized time-series data was used as the time series length. Combined with the routine clinical cycle for monitoring pulp regeneration efficacy, a fixed time window size of 20 consecutive data points was set, with a window sliding step of 5 data points. This setting effectively captures short-term dynamic changes in parameters while avoiding redundancy caused by repeated data analysis. Starting from the first data point in the standardized time-series data, 20 consecutive data points were sequentially extracted according to the sliding step, forming the first window of time-series data. Subsequent extractions were performed every 5 data points until the last data point was included within a window, completing the extraction of all window time-series data.

[0095] For each monitoring parameter in the time-series data of each window, the curve of its value changing over time is treated as a continuous fluctuation signal. Using the peak value of the signal as a reference, the fluctuation position corresponding to each time point is determined. The time difference between this position and the peak position is converted into an angle value between 0 and 360 degrees; this angle value is the instantaneous phase at that time point. When the instantaneous phase difference between two adjacent time points exceeds 180 degrees, 360 degrees is added to the instantaneous phase value of the latter time point. If the difference is less than -180 degrees, 360 degrees is subtracted, ensuring that the phase difference between adjacent time points remains between -180 and 180 degrees, thus completing phase untangling. Then, for any two monitoring parameters, the difference in their instantaneous phase at each time point is calculated, and these differences are arranged sequentially in chronological order to form a phase difference sequence.

[0096] In the phase difference sequence corresponding to the current time window, the average value of all differences is first calculated. The allowable deviation range is set to ±30 degrees of this average value, which is determined based on the clinically reasonable range of fluctuations in dental pulp physiological parameters. Each difference in the phase difference sequence is checked one by one, and the number of differences falling within the allowable deviation range is counted. This number is divided by the total number of data points in the phase difference sequence to obtain a percentage value. This percentage value is the quantitative indicator of the synchronicity between the two monitoring parameters, and its value ranges from 0 to 1.

[0097] Arrange all monitoring parameters according to their priority in clinical diagnosis and treatment, constructing an empty matrix with the same number of rows and columns as the total number of monitoring parameters. The rows and columns of this matrix correspond to the arranged monitoring parameters. Fill the intersection of the row and column of any two monitoring parameters with their corresponding synchronicity quantification index. For example, if parameter X is in row 3 and parameter Y is in column 5, fill the synchronicity quantification index of X and Y in row 3, column 5, and simultaneously fill the index in row 5, column 3, ensuring the symmetry of the matrix. After filling the matrix with the synchronicity quantification indexes for all pairs of monitoring parameters, the correlation matrix is ​​complete.

[0098] The beneficial effects are as follows: by clarifying the specific values ​​and setting basis of the time window size and sliding step size, the standardization and repeatability of window time series data extraction are ensured; the specific operation rules of phase unwrapping eliminate the analysis interference caused by instantaneous phase changes, making the phase difference sequence more reflective of the true correlation of parameters; the clear standard of stability assessment provides a unified basis for the calculation of synchronicity quantification index, making the results more convincing; the orderly process of matrix reconstruction ensures the integrity and symmetry of the correlation matrix, and the final correlation matrix can accurately and objectively reflect the temporal correlation strength between monitoring parameters, providing a high-quality and directly applicable data foundation for the subsequent construction of dynamic data correlation maps.

[0099] The graph construction module 103 is used to construct a dynamic data association graph of the patient based on the association matrix, using time points in the standardized time series data as nodes and the association strength in the association matrix as edge weights.

[0100] In this embodiment of the invention, when the atlas construction module constructs a dynamic data association atlas of the patient based on the association matrix, using time points in the standardized time-series data as nodes, and the association strength in the association matrix as edge weights, it is specifically used for:

[0101] Node mapping is performed on the time points in the standardized time series data to obtain the node set of the time points;

[0102] The elements representing the correlation strength between different time points in the correlation matrix are connected topologically to obtain the connection edges of the correlation matrix.

[0103] The node set and the connecting edges are integrated into a graph topology to obtain the initial weighted graph of the patient.

[0104] Based on the weight distribution characteristics of the connecting edges, the initial weighted graph is subjected to salient connection screening to obtain the patient's core topology.

[0105] The core topology is visualized and encoded to obtain a dynamic data association map of the patient.

[0106] When the graph construction module performs salient connection screening on the initial weighted graph based on the weight distribution characteristics of the connection edges to obtain the patient's core topology, it is specifically used for:

[0107] The weight distribution of the connecting edges is analyzed to obtain the distribution contour of the connecting edges;

[0108] The saliency threshold of the distribution profile is derived to obtain the screening threshold of the distribution profile;

[0109] Based on the filtering threshold, weak connections are filtered out from the connecting edges of the initial weighted graph to obtain a simplified graph of the initial weighted graph.

[0110] By removing isolated nodes from the simplified graph, the candidate core topology of the simplified graph is obtained;

[0111] Based on the screening threshold, the connectivity strength of the candidate core topology is verified to obtain the patient's core topology.

[0112] Each time point in the standardized time series data is assigned a unique identifier. The identifier is prefixed with T and followed by a series of consecutive Arabic numerals, starting from 0 and increasing sequentially in chronological order. T0 strictly corresponds to the time point of the patient's first postoperative efficacy monitoring after pulp regeneration. Each subsequent time point is numbered sequentially at fixed 12-hour intervals as T1, T2, T3, and so on. All time points with unique identifiers are arranged into an ordered set by ascending order of the Arabic numerals in the suffix, forming a node set of time points.

[0113] The rows and columns of the association matrix correspond one-to-one with the time point identifiers in the node set. The specific value corresponding to each element in the matrix is ​​the association strength between the corresponding row identifier time point and the corresponding column identifier time point. All non-zero elements are completely extracted from the association matrix. Each non-zero element corresponds to a separate connection edge. The two endpoints of the connection edge are determined as the time point node of the row and the time point node of the column where the element is located, respectively. The specific value of the non-zero element is directly used as the weight of this connection edge. All connection edges determined in this way are sorted and summarized in the order of the endpoint node identifier combination to obtain the connection edges of the association matrix.

[0114] All nodes in the node set are evenly distributed on the horizontal axis of the two-dimensional plane according to the order of the Arabic numerals following the identifier symbols. The center-to-center distance between two adjacent nodes is set to 10mm. The length of the horizontal axis is flexibly adjusted according to the total number of nodes, ensuring that the leftmost and rightmost nodes have a 5mm edge distance. Each node is represented by a solid circle icon with a diameter of 5mm. The corresponding identifier is marked in black Song typeface, size 4, at the center of the circle icon. Then, according to the correspondence of the endpoints of all connecting edges, a thin black solid line is drawn between the two endpoint nodes of each connecting edge. The thickness, color, and other attributes of the line segment do not distinguish the weight difference of the connecting edges for the time being. The only requirement is that all nodes and connecting edges are fully presented, and the node positions, edge connection relationships, and corresponding identifiers are completely accurate. Finally, an initial weighted graph containing all nodes and connecting edges is formed.

[0115] Collect the weight values ​​of all connecting edges and arrange them in ascending order, ensuring no values ​​are missed or duplicated. Then, divide the arranged weight values ​​into five equidistant weight intervals: 0 to 20, 21 to 40, 41 to 60, 61 to 80, and 81 to 100. The value range of each interval is continuous and non-overlapping. Next, classify each weight value into intervals one by one and count the specific number of connecting edges contained in each weight interval. Plot a planar curve with the weight interval as the horizontal axis and the number of connecting edges in the corresponding interval as the vertical axis. Set the scale length of each interval on the horizontal axis to 1cm and the scale length of each unit on the vertical axis to 0.5cm. The curve is drawn as a continuous solid line, which is the distribution outline of the connecting edges.

[0116] Observe the completed distribution contour curves and compare the curve ordinate values ​​corresponding to each weight interval one by one. The interval with the largest ordinate value is the distribution peak interval with the most connecting edges. Extract the minimum value of the distribution peak interval as the base value. Combined with the standard that the parameter correlation strength in the process of pulp regeneration in clinical practice must reach a certain level to have diagnostic and treatment reference significance, it is determined that the minimum reference value of effective correlation strength in clinical practice is 40. Therefore, the screening threshold is determined to be 40. This threshold can retain connecting edges with actual correlation significance and effectively remove weakly correlated connecting edges with no diagnostic and treatment reference value.

[0117] According to the combination order of the endpoint node identifiers corresponding to the connecting edges, examine the weight value of each connecting edge in the initial weighted graph one by one. Connecting edges with a weight value less than 40 are directly identified as weak connections. The line deletion operation is used to completely remove all the lines identified as weak connections from the initial weighted graph. Only connecting edges with a weight value greater than or equal to 40 and the nodes corresponding to these edges are retained. Ensure that the positional relationship between the retained nodes and connecting edges remains unchanged. The graph formed after removing all weak connections is the simplified graph of the initial weighted graph.

[0118] Following the order of the Arabic numerals in the node identifier suffix, each node in the simplified graph is checked one by one to see if each node is connected to any other node in the simplified graph. If a node is not connected to any other node in the simplified graph, it is directly determined to be an isolated node. The operation of deleting icons is used to completely remove all the circular icons of the isolated nodes from the simplified graph. After deleting the isolated nodes, all the remaining nodes and connecting edges in the simplified graph together constitute the candidate core topology of the simplified graph.

[0119] For each node in the candidate core topology, the number of edges with a weight greater than or equal to the screening threshold of 40 is counted one by one according to the order of the connecting edges. Considering the clinical requirement that nodes need to be connected by multiple effective edges to reflect dynamic association characteristics, a connectivity strength standard is set so that each node is connected to at least two edges with the required weight. If a node has fewer than two edges with a weight greater than or equal to 40, the circular icon of that node and the lines of all its connecting edges are deleted. After the above comprehensive verification and deletion operations, the remaining nodes and connecting edges in the candidate core topology constitute the patient's core topology.

[0120] The connecting edges in the core topology are color-coded: edges with weights between 81 and 100 are marked in red, those between 61 and 80 in orange, those between 41 and 60 in yellow, and those equal to 40 in light yellow. All colors use the solid colors corresponding to the standard color chart. Nodes are size-coded: when a node is connected to two edges with a weight greater than or equal to 40, the diameter of the node's circular icon is set to 5mm; for three edges, it is set to 6mm; and for four or more edges, it is set to 7mm. At the same time, 5mm to the right of each node's circular icon, the corresponding time point symbol is marked in black Song typeface, size 12. At the midpoint of the straight line segment of each connecting edge, its weight value is marked in black Song typeface, size 12. After completing all coding and labeling operations, a dynamic data association map of the patient is formed.

[0121] The beneficial effects are as follows: by clearly defining the rules for identifier symbols, the spacing between node distributions, and the style of icons, the uniqueness, orderliness, and standardization of the initial weighted graph are ensured; by refining the clinical basis for weight interval division, curve drawing standards, and threshold settings, the determination of screening thresholds is made more scientific and operable; by clarifying the criteria for judging weak connections and isolated nodes and the deletion methods, combined with specific verification rules for connectivity strength, the core topological structure obtained through layer-by-layer simplification is more relevant and valuable for reference; and by setting specific color, size coding rules, and labeling positions, the dynamic data association graph is made intuitive, clear, and complete, accurately reflecting the distribution of association strength and topological characteristics between time points, providing a high-quality, visual, and reproducibly verifiable topological foundation for subsequent topological clustering structure change pattern recognition, effectively ensuring the accuracy and efficiency of pattern recognition.

[0122] The pattern recognition module 104 is used to perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient.

[0123] In this embodiment of the invention, when the pattern recognition module performs change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient, it is specifically used for:

[0124] Extract the temporal slice structure from the dynamic data association graph to obtain the topological snapshot sequence of the dynamic data association graph;

[0125] Structural continuity is evaluated on the topological snapshot sequence to obtain the stable community evolution path of the topological snapshot sequence;

[0126] Steady-state structure identification is performed on the stable evolution path of the community to obtain the key steady-state topological patterns of the stable evolution path of the community;

[0127] By performing feature mapping on the key steady-state topological patterns, the biological pattern evolution characteristics of the patient can be obtained.

[0128] Based on the time point identification order in the dynamic data association graph, the time span of the time series slice is set to 24 hours, meaning each slice contains the topological structure corresponding to two consecutive time points. The starting time point of the slice is T0, and slices are taken sequentially. The first slice contains the topological structure corresponding to T0 and T1, the second slice contains the topological structure corresponding to T1 and T2, and so on. Each slice is assigned a unique snapshot identifier, which is prefixed with "S" and suffixed with Arabic numerals, starting from 0 and increasing sequentially. The nodes, connecting edges, their weights, and encoding information within each slice are fully preserved. All slices are arranged in the order of their identifiers to obtain the topological structure snapshot sequence of the dynamic data association graph.

[0129] For two adjacent snapshots in the topological snapshot sequence, the number of nodes and edges they share are counted. The ratio of the number of shared nodes to the total number of nodes in the two snapshots is the node overlap rate, and the ratio of the number of shared edges to the total number of edges in the two snapshots is the edge overlap rate. The structural continuity criterion is set as node overlap rate ≥ 80% and edge overlap rate ≥ 80%. This criterion is based on the clinical characteristics of continuous changes in topology during pulp regeneration. The continuity of adjacent snapshots is verified one by one. Snapshots that continuously meet the criterion are sequentially linked to form continuous structural evolution segments. Multiple continuous segments are integrated in chronological order to obtain the stable evolution path of the community in the topological snapshot sequence.

[0130] For each continuous structural evolution segment in the stable evolution path of the community, the number of snapshots contained in the segment is counted. The judgment condition of steady-state structure is that three or more consecutive snapshots satisfy the structural similarity ≥ 90%. The structural similarity is calculated by comprehensively considering the number of shared nodes, the number of shared connecting edges, and the average absolute value of the weight difference of the corresponding connecting edges. Among them, the average absolute value of the weight difference of the corresponding connecting edges is ≤ 5. Each evolution segment is screened one by one, and the continuous snapshot group that meets the judgment condition is extracted. The snapshot in the middle position of each snapshot group is taken as the representative of the steady-state structure. All representative snapshots are arranged in chronological order to obtain the key steady-state topological pattern of the stable evolution path of the community.

[0131] The pre-defined mapping dimensions of biological pattern evolution features include the total number of nodes, the total number of connecting edges, the proportion of connecting edges with a weight ≥80, the number of core nodes, and the average node connectivity. Core nodes are defined as nodes with ≥4 connecting edges, and the average node connectivity is the sum of the number of connecting edges of all nodes divided by the total number of nodes. For each key steady-state topological pattern, the specific values ​​of the above five dimensions are statistically analyzed. The values ​​of the five dimensions are arranged sequentially according to the time order of the key steady-state topological patterns to form a structured feature sequence. This feature sequence is the biological pattern evolution feature of the patient.

[0132] The beneficial effects include: extracting temporal slices through a clear time span and identification rules to ensure the integrity and temporal accuracy of the topological structure snapshot sequence; verifying structural continuity with specific overlap rate standards to ensure that the stable evolution path of the community can truly reflect the continuous change trend of the topological structure; identifying steady-state structures through multi-dimensional structural similarity judgment conditions to accurately capture key topological states in the pulp regeneration process; and performing feature mapping based on preset specific dimensions to transform abstract topological patterns into quantifiable and interpretable biological pattern evolutionary features, providing accurate and structured feature basis for the subsequent prognostic trend judgment of the pulp regeneration process, and ensuring the scientificity and reliability of the judgment results.

[0133] The trend analysis module 105 is used to analyze the prognostic trend of the patient's pulp regeneration process based on the biological pattern evolution characteristics, and obtain the comprehensive efficacy evaluation conclusion of the patient.

[0134] In this embodiment of the invention, when the trend analysis module performs a prognostic trend analysis of the patient's pulp regeneration process based on the biological pattern evolution characteristics to obtain a comprehensive efficacy evaluation conclusion for the patient, it is specifically used for:

[0135] Event detection is performed on the topological pattern transition events in the biological pattern evolution features to obtain the evolutionary events of the biological pattern evolution features;

[0136] Analyze the temporal distribution patterns and intensity variation trends of the evolutionary events to obtain the evolutionary patterns of the events;

[0137] Based on the aforementioned evolutionary pattern, the healing process of the patient is deduced within the current observation window to obtain the patient's regenerative prognosis path;

[0138] The similarity index of the regenerative prognosis path is obtained by comparing it with the typical healing pattern library of the patient.

[0139] Based on the aforementioned consistency index, the efficacy level of the patient's pulp regeneration process was determined, and a comprehensive efficacy evaluation conclusion for the patient was obtained.

[0140] When the trend analysis module analyzes the temporal distribution patterns and intensity variation trends of the evolutionary events to obtain the evolutionary patterns of the events, it is specifically used for:

[0141] Extract the associated time points and association strengths of events in the evolutionary events;

[0142] Based on the associated time points, a temporal clustering analysis is performed on the temporal distribution pattern of the evolutionary events to obtain the temporal distribution law of the evolutionary events;

[0143] Based on the correlation strength, the evolution direction of the intensity change pattern of the evolutionary event is identified to obtain the intensity change trend of the evolutionary event;

[0144] The temporal distribution pattern and the intensity change trend are integrated into the evolutionary pattern of the evolutionary event.

[0145] Key steady-state topological patterns in the biological model evolution characteristics are compared sequentially in chronological order. The pre-set criteria for determining topological pattern transition events are that any one of the five characteristic dimensions of the key steady-state topological pattern undergoes a specific change. Specifically, the total number of nodes changes by ≥2, the total number of connecting edges changes by ≥3, the proportion of connecting edges with a weight of ≥80 changes by ≥10%, the number of core nodes changes by ≥1, and the average connectivity of nodes changes by ≥0.5. As long as one of these conditions is met, it is determined to be a topological pattern transition event. All the determined transition events are organized in chronological order to obtain the evolutionary events of the biological model evolution characteristics.

[0146] For each evolutionary event, the time point identifier marked by the corresponding key steady-state topological pattern is extracted. This identifier is the associated time point of the evolutionary event. The change of the evolutionary event in five feature dimensions is calculated. The change is the absolute value of the difference between the dimension value of the current key steady-state topological pattern and the dimension value of the previous key steady-state topological pattern. The changes in the five dimensions are added together, and the sum is used as the association strength of the evolutionary event, ensuring that each evolutionary event corresponds to a unique associated time point and association strength value.

[0147] All related time points of evolutionary events are sorted in ascending order by the Arabic numerals of the identifier suffix. The time interval for temporal clustering is set to 48 hours, which is determined based on the clinically common occurrence cycle of evolutionary events during pulp regeneration. Starting from the first related time point, the time difference between subsequent related time points and the last time point in the current cluster is determined. If the time difference is ≤48 hours, it is assigned to the current cluster. If it exceeds 48 hours, a new cluster is started. This operation is repeated until all related time points are clustered. The start time, end time, and number of evolutionary events contained in each cluster are counted to form the temporal distribution pattern of evolutionary events.

[0148] All evolutionary events are arranged in chronological order of their associated time points. The correlation strength difference between adjacent evolutionary events is calculated one by one, which is the correlation strength of the later evolutionary event minus the correlation strength of the earlier evolutionary event. The criteria for determining the direction of intensity change are set as follows: when the difference is ≥5, it is determined to be an increase in intensity; when the difference is between -4 and 4, it is determined to be a stable intensity; and when the difference is ≤-5, it is determined to be a decrease in intensity. The direction of intensity change of each pair of adjacent events is recorded in chronological order and integrated to form the trend of intensity change of evolutionary events.

[0149] The start time, end time, number of events, and intensity change direction of each cluster in the temporal distribution pattern are matched one-to-one with the intensity change direction of the corresponding time period. For example, if a cluster has a time range of T4 to T8 and contains 3 evolution events, and the corresponding intensity change direction is increasing, then this combination relationship is included in the evolution pattern. At the same time, the temporal connection relationship and the transition of intensity change direction between different clusters are clarified, and finally a complete evolution pattern of evolution events is formed.

[0150] Starting with the last key steady-state topological pattern in the current observation window, and continuing the timeline at fixed 12-hour intervals, based on the frequency and intensity of events in the clusters of evolutionary patterns, and combined with pulp healing process data corresponding to similar evolutionary patterns in clinical practice, we predict the specific values ​​of the five dimensions of biological pattern evolutionary characteristics for each subsequent time node. This ensures that the predicted values ​​conform to the continuity logic of the evolutionary pattern. The characteristic values ​​of the starting point and all predicted characteristic values ​​are arranged sequentially according to time to form the patient's regenerative prognosis path.

[0151] The typical healing pattern library contains multiple clinically validated reference healing pathways, and the formula for calculating the consistency index is as follows:

[0152] ;

[0153] In the formula, The consistency index is... The total number of healing stages in the aforementioned typical healing pattern library. In the first Within each healing stage, the degree of matching between the evolutionary events of the regenerative prognosis path and reference events in the typical healing pattern library. This is a measure of the structural stability of the regeneration prognosis path. This is the average of the structural stability metrics for all reference paths in the typical healing pattern library.

[0154] Each reference pathway is divided into three healing stages: inflammation resolution, tissue repair, and regeneration and maturation. This refers to the total number of healing stages inherent in this typical healing pattern library, which is 3. For each healing stage, the evolution events of the regeneration prognosis path are checked one by one against the changes in the characteristic dimensions of the reference events for that stage. The number of events with consistent changes in the characteristic dimensions is counted. The result of dividing the number of consistent events by the total number of reference events for that stage is the result corresponding to the i-th healing stage. Calculate the difference between the maximum and minimum values ​​of five feature dimensions in the regeneration prognosis path at all time points: total number of nodes, total number of connecting edges, proportion of connecting edges with a weight ≥80, number of core nodes, and average node connectivity. The sum of these five differences is the result. Simultaneously, the structural stability metric value of each reference path in the typical healing pattern library is calculated using the same method. The sum of the structural stability metrics of all reference paths is then divided by the total number of reference paths to obtain the final result. .

[0155] According to the above formula, using three stages... Sum divided by 3, then multiplied by and The ratio of the two values ​​is used to obtain the consistency index of the regenerative prognostic path. This calculation process reflects the stage fit between the regenerative prognostic path and the typical healing pattern by comprehensively considering the average of the matching degree of all healing stages. Combined with the structural stability of the regenerative prognostic path relative to the typical pattern, the fit level between the two and the stability of the regenerative prognostic path itself are fully quantified.

[0156] The pre-defined criteria for determining the efficacy level are as follows: a consistency index ≥ 0.85 indicates excellent efficacy, 0.70 ≤ consistency index < 0.85 indicates good efficacy, 0.55 ≤ consistency index < 0.70 indicates moderate efficacy, and a consistency index < 0.55 indicates poor efficacy. The calculated consistency index is compared with the above criteria to determine the corresponding efficacy level, which is the patient's comprehensive efficacy evaluation conclusion.

[0157] The beneficial effects include: determining topological pattern transition events by defining clear thresholds for changes in feature dimensions, ensuring the accuracy of evolutionary event detection; defining association strength by the sum of specific dimensional changes, combined with temporal clustering at fixed time intervals and clear intensity difference standards, making the acquisition of temporal distribution patterns and intensity change trends more objective and operable; and inferring regenerative prognostic paths based on evolutionary patterns and referencing clinical data, first providing specific calculation formulas and variable definitions, then clarifying the source of each parameter and the consistency index calculation, and finally classifying efficacy levels through specific thresholds. The entire process is logically coherent and standardized, effectively ensuring the scientific nature, accuracy, and reproducibility of the comprehensive efficacy evaluation conclusions, providing a reliable basis for clinical diagnosis and treatment decisions.

[0158] The report generation module 106 is used to generate a pulp regeneration efficacy evaluation report for the patient based on the comprehensive efficacy evaluation conclusion.

[0159] In this embodiment of the invention, when the report generation module generates a pulp regeneration efficacy evaluation report for the patient based on the comprehensive efficacy evaluation conclusion, it is specifically used for:

[0160] The efficacy level and prognostic direction in the comprehensive efficacy assessment conclusion are taken as the core conclusions for the patient, and the key stages in the biological pattern evolution characteristics are taken as the supporting evidence for the patient.

[0161] The core conclusions and supporting evidence were compiled into a report, resulting in a draft report for the patient.

[0162] In the draft report, the dynamic data correlation map and the regeneration prognosis path are visualized and integrated to obtain the patient's pulp regeneration efficacy evaluation report.

[0163] The comprehensive efficacy evaluation results extract clear efficacy levels and prognostic directions as the core conclusions for patients. Efficacy levels strictly correspond to pre-defined standards: Excellent efficacy is defined as a consistency index ≥0.85, corresponding to a tooth function recovery rate ≥90% and a apical foramen closure rate ≥85%; Good efficacy is defined as 0.70 ≤ consistency index <0.85, corresponding to a function recovery rate of 80% to 89% and a apical foramen closure rate of 75% to 84%; Moderate efficacy is defined as 0.55 ≤ consistency index <0.70, corresponding to a function recovery rate of 70% to 79% and a apical foramen closure rate of 65% to 74%; Poor efficacy is defined as a consistency index <0.55, corresponding to a function recovery rate <70% and a apical foramen closure rate <65%. Prognostic directions are divided into three categories: Continuously improving prognostic directions indicate that the values ​​of all five characteristic dimensions at the subsequent three time points in the regeneration prognostic path conform to the changing trends of the regeneration maturity stage; Stable maintenance indicates that the fluctuation range of subsequent characteristic dimension values ​​is controlled within ±5; Clinical intervention is required indicates that two or more characteristic dimension values ​​show a continuous decline with a single decline of ≥3. At the same time, key stages are extracted from the evolutionary characteristics of biological patterns, including the inflammation resolution stage, the tissue repair stage, and the regeneration and maturation stage. The time range of the inflammation resolution stage is 1 to 2 weeks, the tissue repair stage is 3 to 8 weeks, and the regeneration and maturation stage is 9 to 24 weeks. The corresponding time range and changes in characteristic dimensions of each stage need to be clearly defined. For example, the total number of nodes decreases by 2 to 3 and the proportion of connection edges with a weight ≥80 decreases by ≥10% in the inflammation resolution stage. The specific manifestations of these key stages serve as evidence for patients.

[0164] The report was compiled according to a fixed structure of "core conclusions first, supporting evidence second." The core conclusions section was presented in two parts: the first part clearly stated the efficacy level and the corresponding functional recovery rate and apical foramen closure rate; the second part explained the prognostic direction and expected changes in subsequent characteristic dimensions. The supporting evidence section was divided into sections according to the time sequence of inflammation resolution, tissue repair, and regeneration maturation. Each section recorded in detail the time range of that stage, the specific changes in each characteristic dimension, and the corresponding topological pattern characteristics. The compilation used SimSun font, size 4, with paragraph spacing set to 1.5. The core conclusions and supporting evidence were separated by a first-level heading, "I. Core Conclusions" and "II. Supporting Evidence," ensuring that the content was neatly arranged, logically clear, and without omissions or redundancy, ultimately forming the patient's draft report.

[0165] Following the "Supporting Evidence" section of the draft report, a new "Visual Analysis" section has been added to integrate the dynamic data association map and the regeneration prognosis path. The dynamic data association map is formatted horizontally on A4 paper. Nodes are evenly distributed along the horizontal axis in chronological order, with a 10mm center-to-center spacing between adjacent nodes. Nodes are represented by solid circular icons, with diameters set according to the number of connecting edges: 5mm for 2 edges, 6mm for 3 edges, and 7mm for 4 or more edges. Connecting edges are color-coded according to their weights: red for weights between 81 and 100, orange for 61 to 80, yellow for 41 to 60, and light yellow for weights equal to 40. A time marker is added next to each node, and the weight value is indicated at the midpoint of each connecting edge. The regeneration prognosis path is presented as a line graph. The horizontal axis represents time nodes, marked at 12-hour intervals, and the vertical axis represents feature dimension values ​​ranging from 0 to 100, with a scale marking every 10 units. Five different colored lines correspond to the five feature dimensions, and each node on each line is labeled with a specific value. The graphs and line charts are arranged vertically, each occupying half of the page. Each visualization chart is accompanied by a 20-30 word explanatory text that clarifies the core information reflected by the chart. After integration, a patient's pulp regeneration efficacy evaluation report is obtained.

[0166] The beneficial effects include: by clearly defining specific indicators for efficacy levels, prognostic directions, and key stage characteristics, the core conclusions and supporting evidence are ensured to be accurate and well-supported; the report drafts are compiled according to a fixed structure to ensure logical coherence and standardized layout; and a standardized visualization method is used to integrate atlases and prognostic pathways, making abstract data intuitive and easy to understand. The resulting pulp regeneration efficacy assessment report is complete, well-organized, and highly operable, providing clinicians with clear and reliable reference for developing subsequent treatment plans, while also ensuring the reproducibility of the report generation process.

[0167] Reference Figure 2 The diagram shown is a flowchart illustrating a method for dynamic evaluation of the therapeutic effect after pulp regeneration surgery according to an embodiment of the present invention. In this embodiment, the method for dynamic evaluation of the therapeutic effect after pulp regeneration surgery includes:

[0168] S1. Perform feature scale normalization on the multimodal efficacy monitoring data of the patients to obtain the standardized time-series data of the patients;

[0169] S2. Perform phase synchronization analysis on the temporal correlation between parameters in the standardized time series data to construct the correlation matrix of the patient;

[0170] S3. Based on the correlation matrix, using the time points in the standardized time series data as nodes and the correlation strength in the correlation matrix as edge weights, construct the dynamic data correlation graph of the patient;

[0171] S4. Perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient;

[0172] S5. Based on the evolutionary characteristics of the biological model, the prognostic trend of the patient's pulp regeneration process is analyzed to obtain the comprehensive efficacy evaluation conclusion of the patient;

[0173] S6. Based on the comprehensive efficacy evaluation conclusion, generate the pulp regeneration efficacy evaluation report for the patient.

[0174] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0175] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0176] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A dynamic evaluation system for the efficacy of pulp regeneration surgery, characterized in that, The system includes a data standardization module, a correlation analysis module, a graph construction module, a pattern recognition module, a trend analysis module, and a report generation module, wherein: The data standardization module is used to normalize the feature scale of the patient's multimodal efficacy monitoring data to obtain the patient's standardized time-series data. The correlation analysis module is used to perform phase synchronization analysis on the temporal correlation between parameters in the standardized time series data in order to construct the correlation matrix of the patient. The graph construction module is used to construct a dynamic data association graph of the patient based on the association matrix, using time points in the standardized time series data as nodes and association strength in the association matrix as edge weights. The pattern recognition module is used to perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient. The trend analysis module is used to analyze the prognostic trend of the patient's pulp regeneration process based on the biological pattern evolution characteristics, and to obtain a comprehensive efficacy evaluation conclusion for the patient. The report generation module is used to generate a pulp regeneration efficacy evaluation report for the patient based on the comprehensive efficacy evaluation conclusion.

2. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 1, characterized in that, When the data standardization module performs feature scale normalization on the patient's multimodal efficacy monitoring data to obtain the patient's standardized time-series data, it is specifically used for: Time axis registration was performed on the multimodal efficacy monitoring data of the patients to obtain the registered data of the patients; The registered data is smoothed for outliers to obtain the patient's clean time-series data. Key feature parameters reflecting the pulp regeneration status are extracted from the clean time-series data to obtain the patient's initial feature set; The initial feature set is parameterized dimensionlessly to obtain the standardized time-series data of the patient.

3. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 1, characterized in that, When performing phase synchronization analysis on the temporal correlation relationships between parameters in the standardized time-series data to construct the patient's correlation matrix, the correlation analysis module is specifically used for: Based on the time series length of the standardized time series data, a time window of the time series length is determined, and the window time series data within the time window is extracted; The instantaneous phases between the monitoring parameters in the window time series data are analyzed, and the instantaneous phases are unwrapped to obtain the phase difference sequence of the instantaneous phases; Within the time window, based on the stability of the phase difference sequence, the stability of the phase synchronization between the monitoring parameters is evaluated to obtain a quantitative index of the synchronization of the monitoring parameters. The synchronicity quantification index is reconstructed into a matrix to obtain the correlation matrix of the patient.

4. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 1, characterized in that, When the atlas construction module constructs a dynamic data association atlas of the patient based on the association matrix, using time points in the standardized time-series data as nodes, and the association strength in the association matrix as edge weights, it is specifically used for: Node mapping is performed on the time points in the standardized time series data to obtain the node set of the time points; The elements representing the correlation strength between different time points in the correlation matrix are connected topologically to obtain the connection edges of the correlation matrix. The node set and the connecting edges are integrated into a graph topology to obtain the initial weighted graph of the patient. Based on the weight distribution characteristics of the connecting edges, the initial weighted graph is subjected to salient connection screening to obtain the patient's core topology. The core topology is visualized and encoded to obtain a dynamic data association map of the patient.

5. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 4, characterized in that, When the graph construction module performs salient connection screening on the initial weighted graph based on the weight distribution characteristics of the connection edges to obtain the patient's core topology, it is specifically used for: The weight distribution of the connecting edges is analyzed to obtain the distribution contour of the connecting edges; The saliency threshold of the distribution profile is derived to obtain the screening threshold of the distribution profile; Based on the filtering threshold, weak connections are filtered out from the connecting edges of the initial weighted graph to obtain a simplified graph of the initial weighted graph. By removing isolated nodes from the simplified graph, the candidate core topology of the simplified graph is obtained; Based on the screening threshold, the connectivity strength of the candidate core topology is verified to obtain the patient's core topology.

6. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 1, characterized in that, When the pattern recognition module performs change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient, it is specifically used for: Extract the temporal slice structure from the dynamic data association graph to obtain the topological snapshot sequence of the dynamic data association graph; Structural continuity is evaluated on the topological snapshot sequence to obtain the stable community evolution path of the topological snapshot sequence; Steady-state structure identification is performed on the stable evolution path of the community to obtain the key steady-state topological patterns of the stable evolution path of the community; By performing feature mapping on the key steady-state topological patterns, the biological pattern evolution characteristics of the patient can be obtained.

7. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 1, characterized in that, When the trend analysis module performs a prognostic trend analysis of the patient's pulp regeneration process based on the biological pattern evolution characteristics and obtains a comprehensive efficacy evaluation conclusion for the patient, it is specifically used for: Event detection is performed on the topological pattern transition events in the biological pattern evolution features to obtain the evolutionary events of the biological pattern evolution features; Analyze the temporal distribution patterns and intensity variation trends of the evolutionary events to obtain the evolutionary patterns of the events; Based on the aforementioned evolutionary pattern, the healing process of the patient is deduced within the current observation window to obtain the patient's regenerative prognosis path; The similarity index of the regenerative prognosis path is obtained by comparing it with the typical healing pattern library of the patient. Based on the aforementioned consistency index, the efficacy level of the patient's pulp regeneration process was determined, and a comprehensive efficacy evaluation conclusion for the patient was obtained.

8. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 7, characterized in that, When the trend analysis module analyzes the temporal distribution patterns and intensity variation trends of the evolutionary events to obtain the evolutionary patterns of the events, it is specifically used for: Extract the associated time points and association strengths of events in the evolutionary events; Based on the associated time points, a temporal clustering analysis is performed on the temporal distribution pattern of the evolutionary events to obtain the temporal distribution law of the evolutionary events; Based on the correlation strength, the evolution direction of the intensity change pattern of the evolutionary event is identified to obtain the intensity change trend of the evolutionary event; The temporal distribution pattern and the intensity change trend are integrated into the evolutionary pattern of the evolutionary event.

9. The dynamic evaluation system for postoperative efficacy of pulp regeneration as described in claim 7, characterized in that, When the report generation module generates a pulp regeneration efficacy assessment report for the patient based on the comprehensive efficacy assessment conclusion, it is specifically used for: The efficacy level and prognostic direction in the comprehensive efficacy assessment conclusion are taken as the core conclusions for the patient, and the key stages in the biological pattern evolution characteristics are taken as the supporting evidence for the patient. The core conclusions and supporting evidence were compiled into a report, resulting in a draft report for the patient. In the draft report, the dynamic data correlation map and the regeneration prognosis path are visualized and integrated to obtain the patient's pulp regeneration efficacy evaluation report.

10. A method for dynamic evaluation of the efficacy of pulp regeneration surgery, characterized in that, The method is for using the dynamic evaluation system for postoperative efficacy assessment of pulp regeneration according to claim 1, wherein the method is as follows: S1. Perform feature scale normalization on the multimodal efficacy monitoring data of the patients to obtain the standardized time-series data of the patients; S2. Perform phase synchronization analysis on the temporal correlation between parameters in the standardized time series data to construct the correlation matrix of the patient; S3. Based on the correlation matrix, using the time points in the standardized time series data as nodes and the correlation strength in the correlation matrix as edge weights, construct the dynamic data correlation graph of the patient; S4. Perform change pattern recognition on the topological clustering structure in the dynamic data association map to obtain the biological pattern evolution characteristics of the patient; S5. Based on the evolutionary characteristics of the biological model, the prognostic trend of the patient's pulp regeneration process is analyzed to obtain the comprehensive efficacy evaluation conclusion of the patient; S6. Based on the comprehensive efficacy evaluation conclusion, generate the pulp regeneration efficacy evaluation report for the patient.

Images