Dental pulp postoperative risk prediction method based on data analysis
By obtaining temperature and skin data of the surgical site after endodontic surgery, combining bacterial breeding and multimodal analysis, time sequence images are constructed, risk types are determined and layered early warning is performed, the problem of misjudgment of single indicators of risk prediction after endodontic surgery is solved, and the accuracy and efficiency of prediction are improved.
Patent Information
- Application Number
- CN202510695563.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-28
- Publication Date
- 2025-08-26
AI Technical Summary
The existing endodontic risk prediction methods have low sensitivity and high misjudgment rate. They fail to effectively integrate multiple factors such as local temperature changes, skin status changes and flora types, resulting in insufficient early identification of post-endodontic infections.
By obtaining the temperature and skin surface data of the surgical site, combining bacterial breeding, time sequence images are constructed, and risk types are determined using fluctuating discrete parameters and body temperature data, and based on the prediction results, whether to obtain secondary samples and issue early warnings, and layered early warnings are fused with multimodal indicators.
It improves the accuracy and timeliness of early diagnosis of post-endolal infection, reduces misjudgment and unnecessary waste of monitoring resources, provides detailed risk warning information, and helps medical staff take measures quickly.
Smart Images
Figure CN120544902A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of postoperative risk prediction, and in particular to a method for predicting postoperative risk of endodontic surgery based on data analysis. Background Art
[0002] In the field of clinical medicine, the prediction of postoperative complications is of great significance for improving patient safety and reducing medical costs. In recent years, with the development of medical technology, more quantitative indicators have gradually been introduced into postoperative risk prediction, such as body temperature monitoring and blood inflammatory marker detection. However, these methods still have many problems in their application after endodontic surgery. Although body temperature monitoring is simple and easy, it is difficult to distinguish between normal body temperature fluctuations after surgery and fever caused by infection, which can easily lead to misjudgment; blood inflammatory marker detection requires blood sampling, which is complicated to operate and cannot be monitored in real time. In addition, existing technologies mostly focus on the analysis of a single indicator, ignoring the multi-factor complexity of postoperative risks, such as local temperature changes, changes in skin condition, and bacterial flora types. These factors are of great value in the early identification of postoperative endodontic infection, but have not yet been fully integrated and utilized. Therefore, there is an urgent need for a postoperative risk prediction method that integrates multiple data dimensions to improve the accuracy and timeliness of early diagnosis of postoperative endodontic infection. Summary of the Invention
[0003] The purpose of the present invention is to provide a method for predicting postoperative risk of endodontics based on data analysis, which solves the problems of low sensitivity and high misjudgment rate of single indicators in traditional methods for predicting postoperative risk in the prior art.
[0004] To this end, the present invention provides a method for predicting the risk after endodontic surgery based on data analysis, the method comprising: Step S1, obtaining a test sample of the surgical site, continuously obtaining state characterization data of the surgical site, and periodically obtaining body temperature data of the surgical target; the state characterization data includes temperature data and skin surface data; Step S2, cultivating the bacterial flora of the test sample to obtain the bacterial flora type and density of each bacterial flora; Step S3, determining a characteristic time period based on historical postoperative detection data of surgical targets and the occurrence of postoperative risks, and calling state representation data of the characteristic time period; Step S4, constructing a skin temperature time-series image based on the temperature data, and constructing a skin surface time-series image based on the skin surface data; Step S5, determining the fluctuation discrete parameters of the skin surface time series image, and determining the type of postoperative risk occurrence based on the fluctuation discrete parameters combined with the body temperature data; Step S6, predicting the postoperative risk of the current surgical target based on the postoperative risk occurrence type, the density of each bacterial colony, the temperature data, and the body temperature data in combination with the characteristic time period; Step S7: Determine whether to obtain a secondary sample based on the prediction result of the postoperative risk, issue a type of postoperative risk warning based on the prediction result and the secondary sample analysis result, and issue a warning.
[0005] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, in step S1, the step of obtaining the temperature data and the skin surface data includes: Determine the postoperative affected area of the surgical site; Determine the darkest area in the postoperative affected area as the temperature state representation area, and determine the external affected area corresponding to the surgical site as the skin state representation area; The chromaticity value of the temperature state representation area is recorded as the temperature data, and the deformation amount of the skin state representation area is recorded as the skin surface data.
[0006] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, in step S3, the step of determining the characteristic time period includes: Obtaining all body temperature data within a preset postoperative time period from the historical postoperative detection data of the surgical target and the occurrence of the postoperative risks; Determine the actual time of occurrence of postoperative risks for each historical surgical goal; If a postoperative risk occurs within the preset postoperative time period, the overlap between the actual time of the postoperative risk and the preset postoperative time period is determined as the initial risk time period; All time periods in which the number of coincidences exceeds the coincidence threshold are determined as the characteristic time periods.
[0007] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, the update period of the skin temperature time series image is the same as the update period of the skin surface time series image; The updating period is the same as the acquisition period of the body temperature data of the surgical target.
[0008] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, in step S5, the step of determining the fluctuating discrete parameter includes: Obtaining historical skin surface data corresponding to the current update cycle time from the historical postoperative detection data of the surgical target, and calculating the mean of the historical skin surface data; Calculating an average deviation of state representation data based on the skin surface time-series images; The fluctuation discrete parameter is determined according to a ratio of the average deviation to a mean value of the historical skin surface data.
[0009] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, in step S5, determining the type of postoperative risk based on the fluctuating discrete parameter combined with the body temperature data includes: If the surgical goal meets the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is a latent postoperative risk type; If the surgical goal does not meet the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is an explicit postoperative risk type; The postoperative recovery condition is that the fluctuation discrete parameter is smaller than a preset fluctuation discrete parameter and the body temperature data of the surgical target is within an allowable postoperative fluctuation range.
[0010] As a preferred technical solution of the method for predicting postoperative risk of endodontic surgery based on data analysis, in step S6, the step of predicting the postoperative risk of the current surgical target includes: If the postoperative risk occurrence type is a latent postoperative risk type, predicting the postoperative risk of the current surgical target based on the temperature data and the density of each bacterial colony; If the postoperative risk occurrence type is an explicit postoperative risk type, the postoperative risk of the current surgical target is predicted based on the characteristic time period and the body temperature data.
[0011] As a preferred technical solution for the method of postoperative risk prediction of endodontic surgery based on data analysis, the postoperative risk of the current surgical target is predicted according to the characteristic time period combined with the body temperature data. If the current update cycle is within the characteristic time period, the result of the postoperative risk prediction is high-level postoperative risk; If the current update cycle is not within the characteristic time period, predicting the postoperative risk of the current surgical target based on the body temperature data; If the body temperature data exceeds the allowable postoperative fluctuation range, the result of the postoperative risk prediction is high-level postoperative risk; If the body temperature data is within the allowable postoperative fluctuation range, the result of the postoperative risk prediction is intermediate postoperative risk.
[0012] As a preferred technical solution of the method for predicting postoperative endodontic risk based on data analysis, in step S7, whether to obtain a secondary sample is determined based on the prediction result of the postoperative risk, and the type of postoperative risk warning issued based on the prediction result and the secondary sample analysis result includes: If the predicted result is the intermediate postoperative risk or below, no secondary sample acquisition is performed; If the predicted result is the high postoperative risk, a second sample is obtained; Cluster analysis of secondary samples was performed to obtain secondary bacterial community categories, which were then compared with those obtained from bacterial culture. If the core flora is consistent in the comparison results, an advanced warning will be issued based on the core flora type and the core flora information will be attached; If the comparison results show that the core flora is inconsistent, only a high-level warning will be issued without the core flora information.
[0013] The beneficial effects of the present invention are: The present invention obtains test samples, status characterization data and body temperature data of the surgical site, cultivates the sample flora, determines the characteristic time period, constructs a time series image, determines the risk type through fluctuating discrete parameters and body temperature, and then predicts the postoperative risk. It also decides whether to obtain a second sample and issue an early warning based on the prediction result. By analyzing historical data, it accurately identifies the characteristic time period, provides strong data support for the prediction, improves the prediction accuracy, optimizes the monitoring focus, and improves the monitoring efficiency. The synchronization of data collection ensures the accuracy and reliability of the prediction. The integration of multimodal indicators and layered early warning improves the warning sensitivity and judgment accuracy, avoiding excessive intervention. The acquisition of secondary samples is handled differently according to the risk level, providing detailed information for medical staff, which helps to quickly predict the patient's risk status and take timely measures.
[0014] In particular, the present invention, by analyzing the postoperative test data of historical surgical targets, can accurately identify characteristic time periods when postoperative risks occur, providing data support for postoperative risk prediction, effectively improving the accuracy of postoperative risk prediction, and helping medical staff take preventive measures in advance to reduce the occurrence of postoperative complications. In addition, after determining the characteristic time periods, the monitoring of surgical targets can be focused on these time periods, thereby improving the efficiency of postoperative monitoring. This not only helps to timely detect potential risks, but also reduces unnecessary waste of monitoring resources.
[0015] In particular, in the present invention, the update cycles of the two also match the acquisition cycle of the body temperature data of the surgical target, ensuring the synchronization and consistency of data collection, so that postoperative risk prediction can be analyzed based on more accurate, real-time and comprehensive data, thereby improving the accuracy and reliability of the prediction, and providing a more scientific basis for postoperative patient monitoring and care.
[0016] In particular, the present invention combines temperature fluctuations with bacterial density thresholds to accurately identify early abnormalities. For overt risks (such as systemic infection), the high-risk time period (24 to 72 hours after surgery) defined by historical data and whether the body temperature exceeds the allowable fluctuation range are used to quickly identify high-risk cases. This invention integrates local and systemic indicators in a multimodal manner, and uses thresholds and dynamic time series analysis to provide stratified warnings, improving the sensitivity and accuracy of postoperative warnings. In addition, the classification of latent / overt risks significantly improves the detection rate of early infection (such as bacterial density combined with temperature fluctuations can identify latent infections) while avoiding excessive intervention (such as the misjudgment of physiological fever). The judgment strategy based on characteristic time periods (driven by historical data) can dynamically adjust warning thresholds (such as seasonal or population-specific body temperature fluctuation ranges), enhancing the model's generalization ability.
[0017] In particular, in the present invention, whether to obtain a secondary sample is determined based on the postoperative risk prediction results, and a corresponding postoperative risk warning is issued based on the secondary sample analysis results. The present invention treats different risk levels differently. For intermediate postoperative risks and below, no secondary sample acquisition is performed, thus avoiding unnecessary operations and waste of resources; while obtaining a secondary sample for high-level postoperative risks is helpful for further in-depth analysis. By clustering and analyzing the secondary samples and comparing them with the cultivated flora categories, the core flora situation can be accurately judged. When the core flora is consistent, an advanced warning with core flora information is issued. When it is inconsistent, an early warning can also be issued in time. This provides medical staff with more targeted and detailed information, helping them to quickly understand the patient's postoperative risk status and take corresponding measures in a timely manner, thereby improving the accuracy and effectiveness of postoperative risk monitoring and providing strong support for patient health protection. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] Figure 1 Flowchart of a method for predicting postoperative risk of endodontic surgery based on data analysis in an embodiment of the present invention; Figure 2 This is a flow chart of determining the characteristic time period in an embodiment of the present invention; Figure 3 This is a flow chart for predicting the postoperative risk of the current surgical target in an embodiment of the present invention. DETAILED DESCRIPTION
[0019] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.
[0020] In the description of the present invention, it should be noted that, unless otherwise expressly specified or limited, the terms "mounted," "connected," and "connected" should be understood in a broad sense. For example, they may refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediate medium; and internal communication between two components. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on the specific circumstances.
[0021] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.
[0022] See also Figure 1 As shown, it is a flow chart of a method for predicting postoperative risk of endodontics based on data analysis in an embodiment of the present invention. The present invention provides a method for predicting postoperative risk of endodontics based on data analysis, comprising: Step S1, obtaining a test sample of the surgical site, continuously obtaining state characterization data of the surgical site, and periodically obtaining body temperature data of the surgical target; the state characterization data includes temperature data and skin surface data; Step S2, cultivating the bacterial flora of the test sample to obtain the bacterial flora type and density of each bacterial flora; Step S3, determining a characteristic time period based on historical postoperative detection data of surgical targets and the occurrence of postoperative risks, and calling state representation data of the characteristic time period; Step S4, constructing a skin temperature time-series image based on the temperature data, and constructing a skin surface time-series image based on the skin surface data; Step S5, determining the fluctuation discrete parameters of the skin surface time series image, and determining the type of postoperative risk occurrence based on the fluctuation discrete parameters combined with the body temperature data; Step S6, predicting the postoperative risk of the current surgical target based on the postoperative risk occurrence type, the density of each bacterial colony, the temperature data, and the body temperature data in combination with the characteristic time period; Step S7: Determine whether to obtain a secondary sample based on the prediction result of the postoperative risk, issue a type of postoperative risk warning based on the prediction result and the secondary sample analysis result, and issue a warning.
[0023] In practice, the cultivation of microbial communities on test samples is an existing technology; secondary sample acquisition involves collecting putty near the pulp of the cavity, followed by 16S rDNA sequencing and analysis. Different VPT methods are performed during the operation depending on whether the pulp is exposed and the state of pulp inflammation. The putty samples are sequenced with 16S rDNA to obtain the amplicon sequence variant (ASV) feature sequence and ASV abundance table, and the core microbial community is screened. The appropriate number of clusters is then determined using the Laplace approximation score (Laplace Approximation). The core microbial community of the putty samples is clustered using the Dirichlet Multinomial Mixtures (DMM) to obtain the microbial typing. This process is an existing technology and will not be repeated here. Generally, the manifestations of postoperative risks in terms of bacterial flora include: dominance of aerobic bacteria (such as streptococci and staphylococci), indicating acute suppurative infection, requiring the use of β-lactam antibiotics (such as amoxicillin); dominance of anaerobic bacteria (such as Prevotella and Porphyromonas), indicating chronic or necrotizing infection, requiring combination with metronidazole or clindamycin; mixed infection, requiring broad-spectrum antibiotic coverage (such as amoxicillin + metronidazole), and debridement and drainage when necessary.
[0024] The postoperative detection data of historical surgical targets include body temperature data, status characterization data, postoperative risk occurrence and duration data, bacterial flora categories and the density of each bacterial flora.
[0025] The present invention obtains test samples, status characterization data and body temperature data of the surgical site, cultivates the sample flora, determines the characteristic time period, constructs a time series image, determines the risk type through fluctuating discrete parameters and body temperature, and then predicts the postoperative risk. It also decides whether to obtain a second sample and issue an early warning based on the prediction result. By analyzing historical data, it accurately identifies the characteristic time period, provides strong data support for the prediction, improves the prediction accuracy, optimizes the monitoring focus, and improves the monitoring efficiency. The synchronization of data collection ensures the accuracy and reliability of the prediction. The integration of multimodal indicators and layered early warning improves the warning sensitivity and judgment accuracy, avoiding excessive intervention. The acquisition of secondary samples is handled differently according to the risk level, providing detailed information for medical staff, which helps to quickly predict the patient's risk status and take timely measures.
[0026] Specifically, in step S1, the step of obtaining the temperature data and the skin surface data includes: Determine the postoperative affected area of the surgical site; Determine the darkest area in the postoperative affected area as the temperature state representation area, and determine the external affected area corresponding to the surgical site as the skin state representation area; The chromaticity value of the temperature state representation area is recorded as the temperature data, and the deformation amount of the skin state representation area is recorded as the skin surface data.
[0027] During implementation, the affected area after surgery is determined by using a temperature-changing patch. First, clean the facial skin, select a temperature-changing patch that is suitable for the body temperature change range, cut it, and stick it on the area corresponding to the surgical site on the face. Observe its color changes every 15 to 30 minutes. The area with the darkest color is the area with the most obvious fever, which is the area representing the temperature status. At the same time, record the color change value and time.
[0028] Understandably, tooth extraction is an invasive procedure that can damage tissues such as the gums and alveolar bone, triggering an immune response. Damaged tissue releases inflammatory mediators, such as prostaglandins and histamine, which dilate local blood vessels and increase permeability. Immune cells, such as white blood cells, flock to the wound to eliminate pathogens and repair damaged tissue. During this process, metabolism in the inflamed area accelerates, increasing heat production and causing facial fever. Fever caused by this inflammatory response is generally common within one to two days after tooth extraction, with body temperatures typically below 38°C. Temperature-changing patches can intuitively reflect temperature differences through color changes, making them a more intuitive way to identify areas with fever. This helps to initially determine the extent and severity of the fever, providing a reference for assessing postoperative infections and other conditions.
[0029] In addition, when postoperative infection occurs, bacteria and other pathogens multiply in the wound, releasing toxins and stimulating the nerve endings around the wound, which will make the pain receptors more sensitive and cause the pain to increase significantly. The surgical target will be biting the teeth, and the corresponding skin condition representation area, that is, the masseter area, will bulge outward and deform. It is understandable that the greater the bite force, the greater the deformation.
[0030] A wireless displacement sensor is placed between the temperature-changing patch and the skin, or a wireless displacement sensor is set on the symmetrical face to obtain the deformation of the skin condition characterization area. The acquisition method is the existing technology and is not specifically limited here.
[0031] The skin state representation region is located inside the temperature state representation region.
[0032] See also Figure 2 As shown, it is Figure 2 This is a flowchart of determining the characteristic time period in an embodiment of the present invention. In step S3, the step of determining the characteristic time period includes: Obtaining all body temperature data within a preset postoperative time period from the historical postoperative detection data of the surgical target and the occurrence of the postoperative risks; Determine the actual time of occurrence of postoperative risks for each historical surgical goal; If a postoperative risk occurs within the preset postoperative time period, the overlap between the actual time of the postoperative risk and the preset postoperative time period is determined as the initial risk time period; All time periods in which the number of coincidences exceeds the coincidence threshold are determined as the characteristic time periods.
[0033] In practice, the preset postoperative time period is 24h to 72h after surgery. The specific time selection is determined according to the type of surgery. The overlap threshold is not less than 2 times. Preferably, the overlap threshold is 3 times. Understandably, in medical data analysis, a single data point is often insufficient to support reliable conclusions. Setting a threshold of three times ensures that the characteristic time period is determined based on multiple overlapping times, effectively reducing interference from random factors and improving data reliability and stability.
[0034] In this invention, by analyzing the postoperative test data of historical surgical targets, characteristic time periods when postoperative risks occur can be accurately identified, providing data support for postoperative risk prediction, effectively improving the accuracy of postoperative risk prediction, and helping medical staff to take preventive measures in advance to reduce the occurrence of postoperative complications. In addition, after determining the characteristic time periods, the monitoring of surgical targets can be focused on these time periods, thereby improving the efficiency of postoperative monitoring. This not only helps to timely detect potential risks, but also reduces unnecessary waste of monitoring resources.
[0035] Specifically, the update period of the skin temperature time-series image is the same as the update period of the skin surface time-series image; The updating period is the same as the acquisition period of the body temperature data of the surgical target.
[0036] In practice, the acquisition period of the surgical target body temperature data is not less than 0.5 hours, preferably 1 hour, and the update period of the skin temperature time series image and the update period of the skin surface time series image are also 1 hour.
[0037] Continuously obtaining the surgical target's body temperature data can be achieved by wearing a temperature tag on the patient, a Bluetooth patch that continuously monitors the body's temperature, or a wearable thermometer. These methods usually have wireless transmission capabilities. This is existing technology and is not specifically limited.
[0038] It is understandable that after endodontic surgery, the surgical wound can easily become a breeding ground for bacteria, especially when oral hygiene is poor or aseptic techniques are not strictly followed during surgery. Postoperative infection may cause local or systemic fever. By monitoring body temperature, signs of infection can be detected in a timely manner. After early detection of fever, antibiotic treatment and other measures can be taken in a timely manner to prevent the spread of infection and reduce the occurrence of complications. In addition, whether the postoperative body temperature is normal or not can reflect the patient's recovery. Normal body temperature indicates that the postoperative recovery is relatively smooth, while abnormal body temperature indicates the risk of postoperative infection. For example, low body temperature indicates an increased risk of postoperative infection, affects drug metabolism and anesthesia recovery, and high body temperature indicates inflammation or infection.
[0039] In the present invention, the update cycles of both also match the acquisition cycle of the body temperature data of the surgical target, ensuring the synchronization and consistency of data collection, so that postoperative risk prediction can be analyzed based on more accurate, real-time and comprehensive data, thereby improving the accuracy and reliability of the prediction, and providing a more scientific basis for postoperative patient monitoring and care.
[0040] Specifically, in step S5, the step of determining the fluctuation discrete parameter includes: Obtaining historical skin surface data corresponding to a current update cycle time from the historical postoperative detection data of the surgical target, and calculating a mean value of the historical skin surface data; Calculating an average deviation of state representation data based on the skin surface time-series images; The fluctuation discrete parameter is determined according to a ratio of the average deviation to a mean value of the historical skin surface data.
[0041] In practice, the fluctuating discrete parameter is determined at the end of each update period.
[0042] Specifically, in step S5, determining the type of postoperative risk based on the fluctuation discrete parameter combined with the body temperature data includes: If the surgical goal meets the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is a latent postoperative risk type; If the surgical goal does not meet the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is an explicit postoperative risk type; The postoperative recovery condition is that the fluctuation discrete parameter is smaller than a preset fluctuation discrete parameter and the body temperature data of the surgical target is within an allowable postoperative fluctuation range.
[0043] In practice, the preset fluctuation discrete parameter is selected in the interval [0.15, 0.25], and the allowable fluctuation range after surgery is [36.5℃, 37.5℃]. Understandably, within the first 72 hours after surgery, body temperature may rise slightly due to surgical trauma and other factors. Postoperative fever is the body's response to surgical trauma, known as "absorption fever." Generally, fever does not exceed 38°C and is a normal postoperative reaction. If the temperature continues to rise or exceeds 38.5°C, other issues such as infection should be considered.
[0044] See also Figure 3 As shown, it is a flowchart of predicting the postoperative risk of the current surgical target in an embodiment of the present invention. In step S6, the step of predicting the postoperative risk of the current surgical target includes: If the postoperative risk occurrence type is a latent postoperative risk type, predicting the postoperative risk of the current surgical target based on the temperature data and the density of each bacterial colony; If the postoperative risk occurrence type is an explicit postoperative risk type, the postoperative risk of the current surgical target is predicted based on the characteristic time period and the body temperature data.
[0045] In practice, the postoperative risk of the surgical target is predicted based on temperature data. If the temperature fluctuation is less than 1°C and the total bacterial density is less than 10 4 If the CFU / mL is lower, it is judged as no postoperative risk at present; otherwise, it is judged as low postoperative risk.
[0046] Temperature fluctuations are obtained based on the maximum and minimum values in the skin temperature time series image.
[0047] Specifically, the postoperative risk of the current surgical target is predicted based on the characteristic time period combined with the body temperature data. If the current update cycle is within the characteristic time period, the result of the postoperative risk prediction is high-level postoperative risk; If the current update cycle is not within the characteristic time period, predicting the postoperative risk of the current surgical target based on the body temperature data; If the body temperature data exceeds the allowable postoperative fluctuation range, the result of the postoperative risk prediction is high-level postoperative risk; If the body temperature data is within the allowable postoperative fluctuation range, the result of the postoperative risk prediction is intermediate postoperative risk.
[0048] This method combines temperature fluctuations with bacterial density thresholds to accurately identify early abnormalities. For overt risks (such as systemic infection), high-risk cases are quickly identified based on historically defined high-risk time periods (24 to 72 hours after surgery) and whether the body temperature exceeds the allowable fluctuation range. This method integrates local and systemic indicators in a multimodal manner, and uses thresholds and dynamic time series analysis to provide stratified warnings, improving the sensitivity and accuracy of postoperative warnings. Furthermore, latent / overt risk classification significantly improves the detection rate of early infection (for example, bacterial density combined with temperature fluctuations can identify latent infections) while avoiding excessive intervention (such as misjudging physiological fever). A judgment strategy based on characteristic time periods (driven by historical data) can dynamically adjust warning thresholds (such as seasonal or population-specific temperature fluctuation ranges), enhancing the model's generalization capabilities.
[0049] Specifically, in step S7, whether to obtain a secondary sample is determined based on the prediction result of the postoperative risk, and the type of postoperative risk warning issued based on the prediction result and the secondary sample analysis result includes: If the predicted result is the intermediate postoperative risk or below, no secondary sample acquisition is performed; If the predicted result is the high postoperative risk, a second sample is obtained; Cluster analysis of secondary samples was performed to obtain secondary bacterial community categories, which were then compared with those obtained from bacterial culture. If the core flora is consistent in the comparison results, an advanced warning will be issued based on the core flora type and the core flora information will be attached; If the comparison results show that the core flora is inconsistent, only a high-level warning will be issued without the core flora information.
[0050] In practice, the secondary sample is obtained without damaging the surgical site, and the sample is obtained near the surgical tooth.
[0051] In the present invention, whether to obtain a secondary sample is determined based on the postoperative risk prediction results, and a corresponding postoperative risk warning is issued based on the secondary sample analysis results. The present invention treats different risk levels differently. For intermediate postoperative risks and below, no secondary sample acquisition is performed, thus avoiding unnecessary operations and waste of resources; while obtaining a secondary sample for high-level postoperative risks is helpful for further in-depth analysis. By clustering and analyzing the secondary samples and comparing them with the cultivated flora categories, the core flora situation can be accurately judged. When the core flora is consistent, an advanced warning with core flora information is issued. When it is inconsistent, an early warning can also be issued in a timely manner. This provides medical staff with more targeted and detailed information, helping them to quickly understand the patient's postoperative risk status and take corresponding measures in a timely manner, thereby improving the accuracy and effectiveness of postoperative risk monitoring and providing strong support for the patient's health protection.
[0052] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the devices, methods and computer program products according to various embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, program segment or part of the code, which contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, as well as the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based device that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.
[0053] Obviously, the above embodiments of the present invention are merely examples for the purpose of clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. A person skilled in the art would be able to make other variations or modifications based on the above description. It is not necessary and impossible to enumerate all embodiments here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention shall be included within the scope of protection of the claims of the present invention.
Claims
1. A method for predicting postoperative risk of endodontic surgery based on data analysis, characterized in that: include: Step S1, obtaining a test sample of the surgical site, continuously obtaining state representation data of the surgical site, and periodically obtaining body temperature data of the surgical target; The state characterization data is temperature data and skin surface data; Step S2, cultivating the bacterial flora of the test sample to obtain the bacterial flora type and density of each bacterial flora; Step S3, determining a characteristic time period based on historical postoperative detection data of surgical targets and the occurrence of postoperative risks, and calling state representation data of the characteristic time period; Step S4, constructing a skin temperature time-series image based on the temperature data, and constructing a skin surface time-series image based on the skin surface data; Step S5, determining the fluctuation discrete parameters of the skin surface time series image, and determining the type of postoperative risk occurrence based on the fluctuation discrete parameters combined with the body temperature data; Step S6, predicting the postoperative risk of the current surgical target based on the postoperative risk occurrence type, the density of each bacterial colony, the temperature data, and the body temperature data in combination with the characteristic time period; Step S7: Determine whether to obtain a secondary sample based on the prediction result of the postoperative risk, issue a type of postoperative risk warning based on the prediction result and the secondary sample analysis result, and issue a warning.
2. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 1, characterized in that: In step S1, the step of obtaining the temperature data and the skin surface data includes: Determine the postoperative affected area of the surgical site; Determine the darkest area in the postoperative affected area as the temperature state representation area, and determine the external affected area corresponding to the surgical site as the skin state representation area; The chromaticity value of the temperature state representation area is recorded as the temperature data, and the deformation amount of the skin state representation area is recorded as the skin surface data.
3. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 2, characterized in that: In step S3, the step of determining the characteristic time period includes: Obtaining all body temperature data and postoperative risk occurrence information within a preset postoperative time period from the historical postoperative test data of the surgical target; Determine the actual time of occurrence of postoperative risks for each historical surgical goal; If a postoperative risk occurs within the preset postoperative time period, the overlap between the actual time of the postoperative risk and the preset postoperative time period is determined as the initial risk time period; All time periods in which the number of coincidences exceeds the coincidence threshold are determined as the characteristic time periods.
4. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 3, characterized in that: The updating period of the skin temperature time-series image is the same as the updating period of the skin surface time-series image; The updating period is the same as the acquisition period of the body temperature data of the surgical target.
5. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 4, characterized in that: In step S5, the step of determining the fluctuation discrete parameter includes: Obtaining historical skin surface data corresponding to a current update cycle time from the historical postoperative detection data of the surgical target, and calculating a mean value of the historical skin surface data; Calculating an average deviation of state representation data based on the skin surface time-series images; The fluctuation discrete parameter is determined according to a ratio of the average deviation to a mean value of the historical skin surface data.
6. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 5, characterized in that: In step S5, determining the type of postoperative risk based on the fluctuation discrete parameter combined with the body temperature data includes: If the surgical goal meets the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is a latent postoperative risk type; If the surgical goal does not meet the postoperative recovery conditions, the postoperative risk occurrence type of the surgical goal is an explicit postoperative risk type; The postoperative recovery condition is that the fluctuation discrete parameter is smaller than a preset fluctuation discrete parameter and the body temperature data of the surgical target is within an allowable postoperative fluctuation range.
7. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 6, characterized in that: In step S6, the step of predicting the postoperative risk of the current surgical target includes: If the postoperative risk occurrence type is a latent postoperative risk type, predicting the postoperative risk of the current surgical target based on the temperature data and the density of each bacterial colony; If the postoperative risk occurrence type is an explicit postoperative risk type, the postoperative risk of the current surgical target is predicted based on the characteristic time period and the body temperature data.
8. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 7, characterized in that: Predict the postoperative risk of the current surgical target based on the characteristic time period combined with the body temperature data, If the current update cycle is within the characteristic time period, the result of the postoperative risk prediction is high-level postoperative risk; If the current update cycle is not within the characteristic time period, predicting the postoperative risk of the current surgical target based on the body temperature data; If the body temperature data exceeds the allowable postoperative fluctuation range, the result of the postoperative risk prediction is high-level postoperative risk; If the body temperature data is within the allowable postoperative fluctuation range, the result of the postoperative risk prediction is intermediate postoperative risk.
9. The method for predicting postoperative risk of endodontic surgery based on data analysis according to claim 8, characterized in that: In step S7, whether to obtain a secondary sample is determined based on the prediction result of the postoperative risk, and the type of postoperative risk warning issued based on the prediction result and the secondary sample analysis result includes: If the predicted result is the intermediate postoperative risk or below, no secondary sample acquisition is performed; If the predicted result is the high postoperative risk, a second sample is obtained; Cluster analysis of secondary samples was performed to obtain secondary bacterial community categories, which were then compared with those obtained from bacterial culture. If the core flora is consistent in the comparison results, an advanced warning will be issued based on the core flora type and the core flora information will be attached; If the comparison results show that the core flora is inconsistent, only a high-level warning will be issued without the core flora information.