Patient risk dynamic assessment system in medical and nursing education

Abstract

This invention relates to the field of medical and nursing technology, and proposes a dynamic patient risk assessment system for medical and nursing education. The system includes: a data acquisition module for acquiring multimodal image data of the postoperative gastric anastomosis area; a data processing module for preprocessing the multimodal image data, establishing a three-dimensional model of the organs surrounding the anastomosis, and analyzing the anastomosis edge features; and a risk assessment module for constructing an anastomosis status assessment model, outputting a comprehensive score based on anastomosis blood perfusion status, probability of abnormal tissue metabolism, risk of anastomosis leakage, anastomosis scar hyperplasia, and healing trend prediction. By enhancing quantum dot fluorescence imaging, Raman spectroscopy imaging, CT scan images, and intraoperative high-definition white light images, and combining these with the opinions of medical staff to arrive at a nursing plan, the system achieves higher accuracy and feasibility compared to manual assessment by medical staff, thus facilitating its use in medical and nursing care.

Description

Technical Field

[0001] This invention relates to the field of medical and nursing technology, and in particular to a dynamic patient risk assessment system for medical and nursing education. Background Technology

[0002] In the field of medical care, accurate risk assessment and the development of personalized care plans are crucial for improving the quality of care and ensuring patient safety. This is especially true in the care of anastomotic sites after gastric cancer surgery, where traditional methods have revealed numerous problems.

[0003] In the imaging assessment stage, traditional single-modal imaging techniques are insufficient to comprehensively reflect the anastomosis condition, and various images are often subject to noise interference, leading to significant errors in the extraction of anastomosis features by medical staff, which in turn affects risk assessment. In the risk assessment stage, traditional methods rely heavily on the subjective experience of medical staff, lacking quantification and standardization. These methods have limitations such as large subjective variability, low specificity, neglect of the interrelationships between indicators, and inability to conduct real-time dynamic assessments. Consequently, in the past, standardized protocols were often used in nursing plan development, without fully considering individual patient differences, resulting in poor nursing outcomes and a high incidence of postoperative complications. Furthermore, traditional nursing education mainly imparts experience through nursing records, lacking realism and interactivity. Medical students find it difficult to deeply understand the logic and basis behind nursing plans, which is not conducive to cultivating their clinical decision-making abilities. Summary of the Invention

[0004] In view of the problems existing in the prior art, the purpose of this invention is to provide a dynamic patient risk assessment system for medical and nursing education, so as to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, this invention provides a dynamic patient risk assessment system for medical and nursing education, comprising:

[0006] The data acquisition module is used to acquire multimodal imaging data of the gastric organ anastomosis area after surgery, including quantum dot fluorescence images, Raman spectroscopy images, CT tomography images and intraoperative high-definition white light images, and to acquire the patient's vital signs data.

[0007] The data processing module is used to preprocess multimodal image data, establish a three-dimensional model of organs surrounding the anastomosis, and analyze the edge features of the anastomosis.

[0008] The risk assessment module is used to build an anastomosis status assessment model and output a comprehensive score of anastomosis blood perfusion status, probability of abnormal tissue metabolism, risk of anastomosis leakage, anastomosis scar hyperplasia, and healing trend prediction.

[0009] The early warning decision-making module is used to classify risk levels based on risk assessment results, generate early warning information, and generate personalized intervention plans by combining current patient vital sign data.

[0010] The interactive feedback module is used to organize the acquired data, generated personalized plans, and actual treatment effects into a dataset, establish a teaching case library, and create virtual operation training scenarios.

[0011] Preferably, the data acquisition module acquires multimodal image data by including the following steps:

[0012] S11. Establish a model for predicting the anastomosis length in patients undergoing gastrectomy, inject targeted quantum dot probes preoperatively, and calculate the probe dose based on the patient's weight.

[0013] S12. Based on the surgical procedure, collect the patient's quantum dot fluorescence images, Raman spectroscopy images, CT tomography images, and intraoperative high-definition white light images.

[0014] S13. Add a timestamp, patient ID, and acquisition device information to each frame of image and upload it to the database for storage.

[0015] Preferably, the data processing module preprocesses the acquired image data by including the following steps:

[0016] S21. Denoise the quantum dot fluorescence image and perform baseline correction on the Raman spectral image;

[0017] S22. Perform three-dimensional reconstruction of CT tomographic images to generate a triangular mesh model of the anastomosis, and equalize the contrast of the high-definition white light images during the operation to enhance the details of the anastomosis suture.

[0018] S23. Calculate the fluorescence intensity distribution at the anastomosis site, extract the edge features of the anastomosis degree, and analyze the anastomosis status;

[0019] S24. Construct the anastomosis feature vector, register the multimodal images, and fuse the extracted geometric features, intensity features, and texture features.

[0020] Preferably, the risk assessment module constructs an anastomosis status assessment model and outputs a comprehensive score, including the following steps:

[0021] S31. Input the processed multimodal image features, perform layer-by-layer feature extraction and transformation on the processed multimodal image features, mine image depth features, and provide a high-dimensional, abstract image representation for subsequent risk assessment.

[0022] S32. Treat the geometric, intensity, and texture features of the anastomosis as graph nodes, pass and update features based on the attention mechanism between nodes, capture the dependencies between features, strengthen the weight of effective features, and improve the accuracy of risk prediction.

[0023] S33. Establish models for anastomotic blood perfusion, probability of abnormal tissue metabolism, risk of anastomotic leakage, anastomotic scar hyperplasia, and healing trend prediction. Statistically fuse the output results of each model to calculate the comprehensive risk value of anastomotic anastomosis after gastric cancer surgery.

[0024] Preferably, in step S33, calculating the comprehensive risk value of anastomosis after gastric cancer surgery includes the following steps:

[0025] S331. Fit the probability distribution of each model output to accurately characterize its uncertainty and fluctuation pattern, distinguish the model reliability and optimize the comprehensive scoring fusion logic.

[0026] S332. By calculating the entropy of the probability distribution output by the calculation module, the reliability of the calculation results of each model is quantified, providing a basis for the reasonable allocation of module weights when comprehensively assessing the risk of anastomosis.

[0027] S333: By allocating weights through normalized entropy weights and combining the distribution mean and variance penalty, we focus on reliable model outputs, suppress fluctuation interference, and accurately and comprehensively assess the anastomotic risk after gastric cancer surgery.

[0028] Preferably, in step S12, the order of acquiring image data is as follows: during the operation, a high-definition white light camera is used to acquire detailed images of the anastomosis suture; within 24 hours after the operation, a quantum dot fluorescence imager is used to acquire microvascular perfusion images of the anastomosis; within 24-48 hours, a Raman spectrometer is used to acquire tissue metabolism images of the anastomosis, with the scanning range covering the anastomosis and surrounding 2cm of tissue; and within 48-72 hours, a CT tomographic scan is performed with a slice thickness set to 0.5mm.

[0029] Preferably, the early warning decision module generates early warning information and personalized intervention plans by including the following steps:

[0030] S41. Set an early warning threshold for anastomotic-related complications. When the real-time assessment results exceed the threshold, trigger a multi-channel early warning based on the early warning priority.

[0031] S42. Conduct risk source analysis to trace the root causes of anastomotic complications after gastric cancer surgery from abnormal indicators, and assist in precise nursing decisions.

[0032] S43. Generate a risk propagation path diagram, display the causal relationship between abnormal indicators, and automatically adjust the monitoring frequency according to the warning level;

[0033] S44. Synchronize the early warning information to the electronic medical record system, mark it as a priority matter, generate a personalized intervention plan, and upload it to the database for storage;

[0034] S45. Analyze the generated personalized plan and use state transition probability, immediate reward and discount factor to determine the anastomotic status value under different nursing intervention actions.

[0035] S46. Adjust the intervention plan based on feedback from medical staff and record the intervention effect for model updates.

[0036] In the preferred step S44, the personalized protocol generation includes local drug release dosage control at the anastomosis site, nutritional support formulation, and early mobilization plan.

[0037] The patient risk dynamic assessment system for medical and nursing education provided by this invention has the following beneficial effects:

[0038] By enhancing quantum dot fluorescence imaging, Raman spectroscopy imaging, CT scan imaging, and intraoperative high-definition white light imaging, and evaluating these images, the resulting nursing plans, combined with the opinions of medical staff, are more accurate and feasible than manual assessment by medical staff. This is beneficial for medical and nursing use. Furthermore, the system can be used for simulation teaching, providing a more realistic experience compared to traditional methods of medical staff reading relevant nursing records. It also makes it easier for medical staff to understand the purpose of the nursing plans and evaluate the nursing plans generated by medical staff, which is beneficial for medical teaching. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is a schematic diagram of the system modules of the dynamic patient risk assessment system provided in this application for medical and nursing education. Detailed Implementation

[0041] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings and examples. The following examples are for illustrative purposes only and should not be construed as limiting the scope of the invention.

[0042] like Figure 1 As shown, this embodiment proposes a dynamic patient risk assessment system for medical and nursing education, including:

[0043] The data acquisition module is used to acquire multimodal imaging data of the gastric organ anastomosis area after surgery, including quantum dot fluorescence images, Raman spectroscopy images, CT tomography images and intraoperative high-definition white light images, and to acquire the patient's vital signs data.

[0044] The data processing module is used to preprocess multimodal image data, establish a three-dimensional model of organs surrounding the anastomosis, and analyze the edge features of the anastomosis.

[0045] The risk assessment module is used to build an anastomosis status assessment model and output a comprehensive score of anastomosis blood perfusion status, probability of abnormal tissue metabolism, risk of anastomosis leakage, anastomosis scar hyperplasia, and healing trend prediction.

[0046] The early warning decision-making module is used to classify risk levels based on risk assessment results, generate early warning information, and generate personalized intervention plans by combining current patient vital sign data.

[0047] The interactive feedback module is used to organize the acquired data, generated personalized plans, and actual treatment effects into a dataset, establish a teaching case library, and create virtual operation training scenarios.

[0048] This invention enhances quantum dot fluorescence imaging, Raman spectroscopy imaging, CT scan imaging, and intraoperative high-definition white light imaging through a systematic approach, and evaluates these images. The resulting nursing plans, combined with feedback from medical staff, offer higher accuracy and feasibility compared to manual assessment by medical personnel, thus facilitating medical and nursing care. Furthermore, this system can be used for simulated teaching, providing a more realistic experience than traditional methods of medical staff reading nursing records. It also makes it easier for medical staff to understand the intent of the nursing plans and evaluate the plans generated by them, which is beneficial for medical teaching.

[0049] In this embodiment, the data acquisition module acquires multimodal image data through the following steps:

[0050] S11. Establish a model for predicting the anastomosis length in patients undergoing gastrectomy, using the following formula: In the formula, For the estimated anastomosis length, Clinical characteristic variables The number of clinical characteristics variables included, but were not limited to, the patient's BMI, age, maximum tumor diameter, gastric tissue resection volume, and intraoperative anastomosis method coefficient. The coefficient for manual anastomosis was 1, and the coefficient for stapled anastomosis was 0.7. To estimate the area of ​​blood flow perfusion at the anastomosis site, To predict the mean fluorescence intensity at the anastomosis site, This represents the tumor volume. The proportion of gastric resection. , , as well as To set the model parameters, a targeted quantum dot probe is injected preoperatively, and the probe dose is calculated based on the patient's weight using the following formula: In the formula, For dosage, For the patient's weight;

[0051] S12. Based on the surgical procedure, collect the patient's quantum dot fluorescence images, Raman spectroscopy images, CT tomography images, and intraoperative high-definition white light images.

[0052] S13. Add a timestamp, patient ID, and acquisition device information to each frame of image and upload it to the database for storage.

[0053] In this embodiment, in step S12, the order of acquiring image data is as follows: during the operation, a high-definition white light camera is used to acquire detailed images of the anastomosis suture; within 24 hours after the operation, a quantum dot fluorescence imager is used to acquire microvascular perfusion images of the anastomosis; within 24-48 hours, a Raman spectrometer is used to acquire tissue metabolism images of the anastomosis, with the scanning range covering the anastomosis and surrounding 2cm of tissue; and within 48-72 hours, a CT tomographic scan is performed with a slice thickness set to 0.5mm.

[0054] Specifically, a multi-dimensional data acquisition system was constructed around the assessment of anastomosis after gastric cancer surgery. By integrating the prediction model of patient characteristics, tumor burden and blood supply status, the prediction error of anastomosis length was controlled within a high-precision range, and the pathophysiological process of anastomosis was accurately matched.

[0055] In this embodiment, the data processing module preprocesses the acquired image data, including the following steps:

[0056] S21. Denoise the quantum dot fluorescence image and perform baseline correction on the Raman spectroscopy image, using the following formula: In the formula, For pixels The denoising value of the sum, For pixels The original value, For similarity weights, and These are the corresponding normalization factors. For pixels Belongs to the neighborhood pixels, For the original signal The wavelet coefficients obtained after performing a continuous wavelet transform wavelet basis functions The complex conjugate, For scale parameters, For translation parameters, For integration variables, It is a differential unit;

[0057] S22. Perform three-dimensional reconstruction of the CT tomographic images to generate a triangular mesh model of the anastomosis. Then, equalize the contrast of the intraoperative high-definition white light images to enhance the details of the anastomosis suture. The formula is: In the formula, These are the three-dimensional coordinates of the surface intersection point located on the line connecting two vertices in the CT volume data, calculated using linear interpolation. and Vertices and The three-dimensional coordinates The threshold set during the thresholding process of CT volume data. and They are respectively and The original grayscale value of the CT scan at the location, and The coordinates of the white light image before and after equalization are respectively: The grayscale value of a pixel. It is the inverse function of the cumulative distribution function. The grayscale value within the image block belongs to The number of pixels;

[0058] S23. Calculate the fluorescence intensity distribution at the anastomosis site, extract the edge features of the anastomosis, and analyze the anastomosis condition. The formula is: In the formula, and These are the weak edges and strong edges that have been filtered out, respectively. It is a proportionality coefficient, and , and These are the image grayscale mean and standard deviation, respectively. To adjust the coefficient, and , The fluorescence intensity is The probability distribution at time, The number of Gaussian distributions, , and The first Mixing coefficients, mean, and covariance matrix of a Gaussian distribution;

[0059] S24. Construct the anastomosis feature vector, register the multimodal images, and fuse the extracted geometric features, intensity features, and texture features using the following formula: In the formula, For the final decision result after integration, The number of types of multimodal images participating in the fusion. For the first The predicted probabilities obtained after analyzing the features of various image modalities. For the first The weight coefficients of each modality in decision fusion.

[0060] Specifically, through multimodal image preprocessing and feature fusion, the reliability and analytical depth of anastomosis assessment data were significantly improved. By enhancing the signal-to-noise ratio of fluorescence images, controlling Raman spectral baseline drift error, and using CT 3D reconstruction and enhanced white light images, the surface error of the anastomosis triangular mesh model was minimized. The entire process achieved full-chain optimization from image denoising to feature abstraction, providing high-dimensional, low-noise quantitative features for the risk assessment model, resulting in higher accuracy of subsequent comprehensive scoring.

[0061] In this embodiment, the risk assessment module constructs an anastomosis status assessment model and outputs a comprehensive score through the following steps:

[0062] S31. Input the processed multimodal image features, perform layer-by-layer feature extraction and transformation on the processed multimodal image features, mine image depth features, and provide a high-dimensional, abstract image representation for subsequent risk assessment. The formula is as follows: In the formula, For the first The first layer Each feature map Given a set of input feature maps, For the first The first layer One input feature map, For convolution kernel, This is the bias value. For activation functions;

[0063] S32. Treat the geometric, intensity, and texture features of the anastomosis joint as graph nodes. Use the attention mechanism between nodes to pass and update features, capture the dependencies between features, strengthen the weights of effective features, and improve the accuracy of risk prediction. The formula is: In the formula, For nodes The output characteristics, For nodes The set of neighboring points, For nodes For nodes Attention coefficient This is the weight matrix. For the parameter vector of the attention mechanism, It is a multimodal feature;

[0064] S33. Establish models for anastomotic blood perfusion, probability of abnormal tissue metabolism, risk of anastomotic leakage, anastomotic scar hyperplasia, and healing trend prediction. Statistically fuse the output results of each model to calculate the comprehensive risk value of anastomotic anastomosis after gastric cancer surgery.

[0065] Specifically, the formula for the quantitative model of anastomotic blood flow perfusion is as follows: In the formula, The blood perfusion index, Peak systolic blood flow velocity, As a resistance index, Hemoglobin concentration, The diameter of the anastomotic vessel. This refers to the number of days after surgery. , , , , The formula for the probability model of abnormal tissue metabolism, corresponding to the weighting coefficients, is as follows: In the formula, This represents the probability of metabolic abnormalities. For linear predictors, Lactate level, The pH value of the interstitial fluid. This refers to venous blood oxygen saturation. This refers to the glucose concentration. It is C-reactive protein. , , , , The formula for the anastomosis leakage risk scoring model, corresponding to the regression coefficients, is as follows: In the formula, To score the risk, Number of risks Risk weighting As a risk variable, the formula for the anastomotic scar hyperplasia prediction model is: In the formula, The scar hyperplasia index, Transforming growth factor, The ratio of type I to type III collagen. For wound tension, For the patient's age, History of diabetes , , , , This is the corresponding proportion coefficient.

[0066] In this embodiment, step S33, calculating the comprehensive risk value of the anastomosis status after gastric cancer surgery, includes the following steps:

[0067] S331. Fit the probability distribution of each model's output to accurately characterize its uncertainty and fluctuation patterns, distinguish model reliability, and optimize the comprehensive scoring fusion logic. The formula is as follows: In the formula, For the first The output of each model, The sign for normal distribution. and The first The output mean and variance of each model;

[0068] S332. By calculating the entropy of the probability distribution output by the calculation module, the reliability of each model's calculation results is quantified, providing a basis for rationally allocating module weights when comprehensively assessing the risk of anastomosis. The formula is as follows: In the formula, For the first The information entropy output by the model For the first The model output results The probability density function, For the first The model output results The integral variable;

[0069] S333. By allocating weights using normalized entropy weights and combining the distribution mean and variance penalty, the system focuses on reliable model output and suppresses fluctuation interference to accurately and comprehensively assess the anastomotic risk after gastric cancer surgery. The formula is as follows: In the formula, To obtain the comprehensive risk score of the anastomosis after integrating information from various models, For the number of models, For the first The entropy of the output distribution of each model, and , For the first The mean of the output distribution of each model. This is the variance penalty coefficient. For the first The variance of the output distribution of each model.

[0070] Specifically, a high-precision anastomotic risk assessment system was constructed through multi-layer feature extraction, graph attention mechanism, and multi-model fusion. By mining image depth features layer by layer, the correlation between multiple indicators such as fluorescence intensity and texture and anastomotic leakage was extracted. The accuracy of capturing the dependence between geometric features was improved through the attention mechanism, and the weight of key features such as abnormal blood perfusion was strengthened, so that the Kappa value of the comprehensive risk score and clinical outcome was highly consistent. The whole process realizes intelligent assessment from feature abstraction to uncertainty quantification, improves the prediction accuracy of complications such as anastomotic leakage and scar hyperplasia, and provides a reliable quantitative basis for personalized intervention.

[0071] In this embodiment, the early warning decision module generates early warning information and personalized intervention plans through the following steps:

[0072] S41. Set an early warning threshold for anastomotic-related complications. When the real-time assessment result exceeds the threshold, trigger a multi-channel early warning based on the early warning priority. The formula is as follows: In the formula, As a priority for early warning, This represents the probability of leakage risk. For blood perfusion scoring, This is a metabolic abnormality index. , and These are all corresponding weighting coefficients;

[0073] S42. Conduct risk source analysis to trace the root causes of anastomotic complications after gastric cancer surgery from abnormal indicators, and assist in precise nursing decisions.

[0074] S43. Generate a risk propagation path diagram, display the causal relationship between abnormal indicators, and automatically adjust the monitoring frequency according to the warning level;

[0075] S44. Synchronize the early warning information to the electronic medical record system, mark it as a priority matter, generate a personalized intervention plan, and upload it to the database for storage;

[0076] S45. Analyze the generated personalized plan, and use state transition probability, immediate reward, and discount factor to determine the anastomotic state value under different nursing intervention actions. The formula is as follows: In the formula, For the value of the matching mouth state, Let be the state transition probability. To provide timely rewards, As a discount factor, To implement nursing intervention actions Then from the current state The next state to which it transitions;

[0077] S46. Adjust the intervention plan based on feedback from medical staff and record the intervention effect for model updates.

[0078] In this embodiment, step S44, the personalized treatment plan generation includes local drug release dosage control at the anastomosis site, nutritional support formula, and early mobilization plan, as shown in the formula: In the formula, and These are the final and baseline doses of the local drug-releasing agent at the anastomosis site, respectively. The percentage of protein in a nutritional support formula. This is the albumin level test value. The intensity level of early activity for patients after gastric cancer surgery. This is the boundary constraint function.

[0079] Specifically, by integrating leakage risk, perfusion score, and metabolic abnormalities into quantitative indicators, the early warning response time is shortened compared to manual early warning. Through risk tracing and propagation path mapping, medical staff can quickly locate the root cause of complications, improve the accuracy of causal relationship identification, and the generated personalized intervention plan is dynamically optimized through a reinforcement learning framework, thereby improving the effectiveness of personalized nursing plans, reducing the incidence of serious postoperative complications, and improving the efficiency of nursing decision-making.

[0080] The above embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Although the invention has been described in detail with reference to the embodiments, those skilled in the art should understand that various combinations, modifications, or equivalent substitutions of the technical solutions of the invention do not depart from the spirit and scope of the invention and should be covered within the scope of the claims of the invention.

Claims

1. A dynamic patient risk assessment system for medical and nursing education, characterized in that, include: The data acquisition module is used to acquire multimodal imaging data of the gastric organ anastomosis area after surgery, including quantum dot fluorescence images, Raman spectroscopy images, CT tomography images and intraoperative high-definition white light images, and to acquire the patient's vital signs data. The data processing module is used to preprocess multimodal image data, establish a three-dimensional model of organs surrounding the anastomosis, and analyze the edge features of the anastomosis. The risk assessment module is used to build an anastomosis status assessment model and output a comprehensive score of anastomosis blood perfusion status, probability of abnormal tissue metabolism, risk of anastomosis leakage, anastomosis scar hyperplasia, and healing trend prediction. The early warning decision-making module is used to classify risk levels based on risk assessment results, generate early warning information, and generate personalized intervention plans by combining current patient vital sign data. The interactive feedback module is used to organize the acquired data, generated personalized plans, and actual treatment effects into a dataset, establish a teaching case library, and create virtual operation training scenarios. The data processing module preprocesses the acquired image data, including the following steps: S21. Denoise the quantum dot fluorescence image and perform baseline correction on the Raman spectral image; S22. Perform three-dimensional reconstruction of CT tomographic images to generate a triangular mesh model of the anastomosis, and equalize the contrast of the high-definition white light images during the operation to enhance the details of the anastomosis suture. S23. Calculate the fluorescence intensity distribution at the anastomosis site, extract the edge features of the anastomosis degree, and analyze the anastomosis status; S24. Construct the anastomosis feature vector, register the multimodal images, and fuse the extracted geometric features, intensity features, and texture features; The risk assessment module constructs an anastomosis status assessment model and outputs a comprehensive score, including the following steps: S31. Input the processed multimodal image features, perform layer-by-layer feature extraction and transformation on the processed multimodal image features, mine image depth features, and provide a high-dimensional, abstract image representation for subsequent risk assessment. S32. Treat the geometric, intensity, and texture features of the anastomosis as graph nodes, pass and update features based on the attention mechanism between nodes, capture the dependencies between features, strengthen the weight of effective features, and improve the accuracy of risk prediction. S33. Establish models for anastomotic blood perfusion, probability of abnormal tissue metabolism, risk of anastomotic leakage, anastomotic scar hyperplasia, and healing trend prediction. Statistically fuse the output results of each model to calculate the comprehensive risk value of anastomotic anastomosis after gastric cancer surgery.

2. The dynamic patient risk assessment system for medical and nursing education according to claim 1, characterized in that, The data acquisition module acquires multimodal image data through the following steps: S11. Establish a model for predicting the anastomosis length in patients undergoing gastrectomy, inject targeted quantum dot probes preoperatively, and calculate the probe dose based on the patient's weight. S12. Based on the surgical procedure, collect the patient's quantum dot fluorescence images, Raman spectroscopy images, CT tomography images, and intraoperative high-definition white light images. S13. Add a timestamp, patient ID, and acquisition device information to each frame of image and upload it to the database for storage.

3. The dynamic patient risk assessment system for medical and nursing education according to claim 1, characterized in that, In step S33, calculating the comprehensive risk value of anastomosis after gastric cancer surgery includes the following steps: S331. Fit the probability distribution of each model output to accurately characterize its uncertainty and fluctuation pattern, distinguish the model reliability and optimize the comprehensive scoring fusion logic. S332. By calculating the entropy of the probability distribution output by the calculation module, the reliability of the calculation results of each model is quantified, providing a basis for the reasonable allocation of module weights when comprehensively assessing the risk of anastomosis. S333: By allocating weights through normalized entropy weights and combining the distribution mean and variance penalty, we focus on reliable model outputs, suppress fluctuation interference, and accurately and comprehensively assess the anastomotic risk after gastric cancer surgery.

4. The dynamic patient risk assessment system for medical and nursing education according to claim 2, characterized in that, In step S12, the sequence of image data acquisition is as follows: during the operation, a high-definition white light camera is used to acquire detailed images of the anastomosis suture; within 24 hours after the operation, a quantum dot fluorescence imager is used to acquire microvascular perfusion images of the anastomosis; within 24-48 hours, a Raman spectrometer is used to acquire tissue metabolism images of the anastomosis, with the scanning range covering the anastomosis and surrounding 2cm of tissue; and within 48-72 hours, a CT tomographic scan is performed with a slice thickness set to 0.5mm.

5. The dynamic patient risk assessment system for medical and nursing education according to claim 1, characterized in that, The early warning decision-making module generates early warning information and personalized intervention plans through the following steps: S41. Set an early warning threshold for anastomotic-related complications. When the real-time assessment results exceed the threshold, trigger a multi-channel early warning based on the early warning priority. S42. Conduct risk source analysis to trace the root causes of anastomotic complications after gastric cancer surgery from abnormal indicators, and assist in precise nursing decisions. S43. Generate a risk propagation path diagram, display the causal relationship between abnormal indicators, and automatically adjust the monitoring frequency according to the warning level; S44. Synchronize the early warning information to the electronic medical record system, mark it as a priority matter, generate a personalized intervention plan, and upload it to the database for storage; S45. Analyze the generated personalized plan and use state transition probability, immediate reward and discount factor to determine the anastomotic status value under different nursing intervention actions. S46. Adjust the intervention plan based on feedback from medical staff and record the intervention effect for model updates.

6. The dynamic patient risk assessment system for medical and nursing education according to claim 5, characterized in that, In step S44, the personalized plan generation includes local drug release dosage control at the anastomosis site, nutritional support formulation, and early activity plan.

Images