Management method and system for nursing safety of blood purification center

Abstract

The invention discloses a blood purification center nursing safety management method and system, and relates to the technical field of medical data analysis, and the method comprises the steps: collecting and preprocessing initial physiological data of a patient, dialysis initial parameters and vascular access state data at the beginning of dialysis, and generating a multi-modal data set; constructing a multi-modal blood safety model, predicting physiological indexes according to the multi-modal data set, and evaluating the blood safety of the patient according to the physiological indexes; a dialysis scheme is generated according to the evaluation result, and dialysis parameters are dynamically adjusted; collecting dialysis effect data, analyzing the effectiveness of a dialysis scheme through a statistical analysis method, and generating a dialysis effect report; according to the dialysis effect report, the multi-modal blood safety model is updated, and an optimized multi-modal blood safety model is generated. According to the method, the blood safety state of the patient is accurately evaluated by constructing the multi-modal blood safety model, and the safety and effectiveness of dialysis treatment are remarkably enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of medical data analysis, and in particular to a method and system for managing nursing safety in a blood purification center. Background Art

[0002] With the continuous advancement of medical technology, blood purification has become an important treatment for acute and chronic renal failure, acute poisoning, and various metabolic disorders. As the primary location for centralized hemodialysis treatment, the safety of care provided by blood purification centers is directly related to patients' quality of life and prognosis. In recent years, with the widespread adoption of blood purification technology, improving the safety of care in blood purification centers has become a research hotspot.

[0003] Management methods for nursing safety in blood purification centers fail to fully utilize the correlation between patients' multimodal data (such as physiological data, dialysis parameters, and vascular access status data), resulting in insufficient accuracy and reliability of assessment results. Secondly, management methods for nursing safety in blood purification centers often rely on fixed rules or simple empirical models, lacking dynamic consideration of individual patient characteristics and unable to adjust dialysis parameters in real time based on patients' physiological changes, resulting in insufficient safety and effectiveness of dialysis programs. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a management method for nursing safety in a blood purification center to solve the problems of insufficient utilization of multimodal data and lack of dynamic adjustment capability of dialysis schemes.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for managing nursing safety in a blood purification center, which includes collecting the patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis and preprocessing them to generate a multimodal data set; constructing a multimodal blood safety model, predicting physiological indicators based on the multimodal data set, and evaluating the patient's blood safety based on the physiological indicators; generating a dialysis plan based on the evaluation results, and dynamically adjusting the dialysis parameters; collecting dialysis effect data, analyzing the effectiveness of the dialysis plan through statistical analysis methods, and generating a dialysis effect report; updating the multimodal blood safety model based on the dialysis effect report to generate an optimized multimodal blood safety model.

[0008] As a preferred embodiment of the method for managing nursing safety in a blood purification center according to the present invention, the following is mentioned:

[0009] The preprocessing includes missing value filling, outlier removal and data normalization;

[0010] The preprocessed initial physiological data, initial dialysis parameters, and vascular access status data were aligned according to timestamps and merged into a multimodal dataset;

[0011] Health status categories are added to the multimodal data at each time point in the multimodal dataset according to standardized medical assessment methods.

[0012] As a preferred solution of the blood purification center nursing safety management method of the present invention, wherein: the multimodal blood safety model is constructed, the specific steps are as follows:

[0013] The input layer is defined using a fully connected layer, and the feature extraction layer is defined using a convolutional neural network model and a long short-term memory network model;

[0014] The health status prediction layer is defined by the fully connected layer and the Softmax activation function, and the output layer is defined by the result integration method;

[0015] A multimodal blood safety model is constructed based on the input layer, feature extraction layer, health status prediction layer and output layer.

[0016] As a preferred solution of the blood purification center nursing safety management method of the present invention, wherein: the physiological indicators are predicted based on the multimodal data set, and the patient's blood safety is evaluated based on the physiological indicators. The specific steps are as follows:

[0017] The multimodal dataset is input into the multimodal blood safety model, and physiological indicators are generated through forward propagation of the multimodal blood safety model;

[0018] Assess the patient's blood safety status through direct assessment method.

[0019] As a preferred solution of the blood purification center nursing safety management method of the present invention, wherein: generating a dialysis plan based on the evaluation results and dynamically adjusting the dialysis parameters, the specific steps are as follows:

[0020] Based on the evaluation results, a dialysis plan is generated through the dialysis parameter derivation method;

[0021] The patient's physiological indicators are monitored in real time during the dialysis process. When the physiological indicators are abnormal, the dialysis parameters are dynamically adjusted according to the abnormal situation.

[0022] As a preferred solution of the blood purification center nursing safety management method of the present invention, wherein: the dialysis effect data is collected, the effectiveness of the dialysis program is analyzed by statistical analysis method, and a dialysis effect report is generated. The specific steps are as follows:

[0023] Statistical analysis methods were used to analyze the changes in toxin clearance, fluid balance, and electrolyte balance before and after dialysis, and the dialysis effect was classified according to the changes in each indicator.

[0024] A dialysis effect report is generated based on the changes in each indicator and the dialysis effect classification results.

[0025] As a preferred solution of the blood purification center nursing safety management method of the present invention, wherein: the multimodal blood safety model is updated according to the dialysis effect report to generate an optimized multimodal blood safety model, the specific steps are as follows:

[0026] Based on the changes in various indicators in the dialysis effect report, the structure and hyperparameters of the multimodal blood safety model and the weights of the feature extraction layer are updated;

[0027] The performance of the multimodal blood safety model is evaluated by regularly collecting dialysis effect reports and continuously optimized based on the evaluation results.

[0028] In a second aspect, the present invention provides a management system for nursing safety in a blood purification center, comprising a data acquisition module, a model building module, a dialysis plan generation module, a dialysis effect analysis module, and a model update module; the data acquisition module is used to collect the patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis and perform preprocessing to generate a multimodal data set; the model building module is used to construct a multimodal blood safety model, predict physiological indicators based on the multimodal data set, and evaluate the patient's blood safety based on the physiological indicators; the dialysis plan generation module is used to generate a dialysis plan based on the evaluation results and dynamically adjust the dialysis parameters; the dialysis effect analysis module is used to collect dialysis effect data, analyze the effectiveness of the dialysis plan through statistical analysis methods, and generate a dialysis effect report; the model update module is used to update the multimodal blood safety model based on the dialysis effect report to generate an optimized multimodal blood safety model.

[0029] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for managing nursing safety in a blood purification center as described in the first aspect of the present invention is implemented.

[0030] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for managing nursing safety in a blood purification center as described in the first aspect of the present invention is implemented.

[0031] The beneficial effects of the present invention are as follows: through multimodal data fusion and construction of a multimodal blood safety model, an accurate assessment of the patient's blood safety status is achieved, and a personalized dialysis plan is generated, thereby improving the accuracy of blood safety assessment and the personalization of dialysis plans, and significantly enhancing the safety and effectiveness of dialysis treatment; through multi-indicator comprehensive evaluation and classification methods, a comprehensive analysis and intuitive expression of dialysis effects are achieved, providing reliable data support for the dynamic adjustment of dialysis plans, and improving the scientificity and operability of dialysis treatment. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0033] Figure 1 Flowchart of the management approach for nursing safety in blood purification centers;

[0034] Figure 2 Schematic diagram of the management system for nursing safety in blood purification centers;

[0035] Figure 3 Flowchart constructed for the multimodal blood safety model;

[0036] Figure 4 Flowchart updated for the multimodal blood safety model. DETAILED DESCRIPTION

[0037] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0038] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0039] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0040] Reference Figures 1 to 4, is an embodiment of the present invention, which provides a method for managing nursing safety in a blood purification center, comprising the following steps:

[0041] S1. Collect the patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis and preprocess them to generate a multimodal dataset.

[0042] Initial physiological data include vital signs and blood biochemistry data;

[0043] It should be noted that vital signs data refers to blood pressure, heart rate, blood oxygen saturation, respiratory rate and body temperature; blood biochemical data refers to blood urea nitrogen, blood creatinine, blood potassium, blood sodium and hemoglobin; vital signs data and blood biochemical data can be collected through corresponding equipment.

[0044] Initial dialysis parameters include blood flow, ultrafiltration volume, dialysate flow, and anticoagulant dosage;

[0045] It should be noted that the initial dialysis parameters refer to the control parameters of the dialysis equipment and dialysis process that are pre-set before the start of hemodialysis treatment based on the patient's condition, weight, dialysis goals and other individualized information.

[0046] Vascular access status data include blood flow velocity, vessel wall pressure, and vascular access type;

[0047] It should be noted that ultrasound equipment is used to measure the blood flow velocity in the vascular access. During dialysis, the blood flow velocity is indirectly monitored by the pressure sensor of the dialysis machine. The vascular wall pressure includes arterial pressure and venous pressure, which are monitored in real time using the pressure sensor of the dialysis machine. The vascular access type refers to the vascular connection method used by the patient for hemodialysis treatment, which usually includes arteriovenous fistula, arteriovenous graft fistula and central venous catheter. The type of vascular access is determined by combining blood flow velocity and vascular wall pressure data.

[0048] Preprocessing includes missing value filling, outlier removal, and data normalization;

[0049] It should be noted that the linear interpolation method was used to fill missing values. For example, for the time series of physiological data, the average value of the previous and next time points can be used to fill missing values. The isolation forest method was used to detect and eliminate outliers. The initial physiological data, initial dialysis parameters, and vascular access status data were normalized to the interval [0, 1].

[0050] The preprocessed initial physiological data, initial dialysis parameters, and vascular access status data were aligned according to timestamps and merged into a multimodal dataset;

[0051] It should be noted that the database management tool MySQL was used to extract timestamp columns from the initial physiological data, dialysis initial parameters, and vascular access status data. The nearest neighbor timestamp alignment method was used to find records with the same or similar timestamps in the initial physiological data, dialysis initial parameters, and vascular access status data, and the records with the same or similar timestamps were merged into one row. The aligned initial physiological data, dialysis initial parameters, and vascular access status data were merged into one data table. When merging, all initial physiological data, dialysis initial parameters, and vascular access status data with the same timestamp were integrated into one row based on the timestamp to generate a multimodal dataset.

[0052] Add health status categories to the multimodal data at each time point in the multimodal dataset based on standardized medical assessment methods;

[0053] It should be noted that the classification standards for health status categories are formulated, and the health status categories include normal, hypotension risk, hypertension risk, and imbalance syndrome risk. If the patient's initial physiological data and initial dialysis parameters are within the normal range (systolic blood pressure 90-140 mmHg, heart rate 60-100 bpm), it is marked as normal; if the systolic blood pressure is <90 mmHg or the heart rate is <60 bpm, it is marked as hypotension risk; if the systolic blood pressure is >140 mmHg or the heart rate is >100 bpm, it is marked as hypertension risk; if the change in urea nitrogen clearance before and after dialysis exceeds 20%, it is marked as imbalance syndrome risk.

[0054] S2. Build a multimodal blood safety model, predict physiological indicators based on multimodal data sets, evaluate the patient's blood safety based on physiological indicators, generate a dialysis plan based on the evaluation results, and dynamically adjust dialysis parameters.

[0055] The input layer is defined using a fully connected layer, and the feature extraction layer is defined using a convolutional neural network model and a long short-term memory network model;

[0056] It should be noted that the fully connected layer is used to receive the input multimodal data set and map it to a high-dimensional feature space. The output dimension of the fully connected layer can be set according to the requirements of subsequent feature extraction, for example, mapping the input to a 64-dimensional feature space. The convolutional neural network (CNN) is used to extract multimodal data features. The output of the fully connected layer is adjusted to a shape suitable for the convolution operation. A one-dimensional convolution layer (Conv1 D) is used to extract local multimodal data features. The convolution kernel size can be set to 3, the number of filters is set to 32, and the activation function is ReLU. The maximum pooling layer (MaxPooling1 D) is used to downsample the convolution result to reduce the dimension of the multimodal data features. The pooling window size can be set to 2, and the pooled multimodal data features are flattened into a one-dimensional vector. The long short-term memory network (LSTM) is used to extract long-term dependencies in time series data. The output of the fully connected layer is adjusted to a shape suitable for the LSTM input. The LSTM layer is used to extract temporal features, and the number of LSTM units can be set to 64. The features extracted by CNN and LSTM are spliced ​​together to form a comprehensive multimodal data feature representation.

[0057] The health status prediction layer is defined by the fully connected layer and the Softmax activation function, and the output layer is defined by the result integration method;

[0058] It should be noted that the input of the health status prediction layer is the multimodal data features output by the feature extraction layer. The fully connected layer is used to map the input multimodal data features to the number of health status categories. The number of neurons in the fully connected layer is set to the number of health status categories. The activation function Softmax is used to generate the probability distribution of health status categories and extract physiological indicators based on the number of health status categories. The output layer integrates the health status prediction results into a complete output.

[0059] Construct a multimodal blood safety model based on the input layer, feature extraction layer, health status prediction layer, and output layer;

[0060] It should be noted that the multimodal dataset is divided into a training set (70%), a validation set (15%), and a test set (15%). The health status prediction layer is a multi-classification task, and the classification cross entropy is used as the loss function, which is expressed as follows:

[0061]

[0062] Among them, L1 represents the health state loss value, N represents the number of samples, C represents the number of health state categories, and y ij represents the true label of the i-th sample in the j-th category, p ij represents the predicted probability of the i-th sample in the j-th category, i represents the index variable of the number of samples, with a value of 1-N, and j represents the index variable of the number of health status categories, with a value of 1-C;

[0063] It should also be noted that the number of training rounds and batch size of the multimodal blood safety model are set, the training set data is input into the multimodal blood safety model, the loss function value is calculated, and the Adam optimizer is used to optimize the number of training rounds and batch size to minimize the loss function; after each training round, the validation set is used to evaluate the performance of the multimodal blood safety model; after the performance verification of the multimodal blood safety model meets the standards, the multimodal blood safety model is saved to a file.

[0064] The test set is input into the multimodal blood safety model, and physiological indicators are generated through forward propagation of the multimodal blood safety model;

[0065] It should be noted that physiological indicators include systolic blood pressure, heart rate, blood oxygen saturation, blood flow, ultrafiltration volume and dialysate flow. The test set is completely not involved in the training phase of the multimodal blood safety model to ensure that the test set data distribution and characteristics are not learned by the multimodal blood safety model; the test set data is input into the input layer of the multimodal blood safety model, and passes through the feature extraction layer and the health status prediction layer to generate the number of health status categories. Based on the number of health status categories, the physiological indicator prediction results of each sample are extracted.

[0066] Assess the patient's blood safety status through direct assessment methods;

[0067] It should be noted that the healthy range of each physiological indicator is defined according to medical standards, including systolic blood pressure 90-140 mmHg (normal range), diastolic blood pressure 60-90 mmHg (normal range); heart rate 60-100 bpm (normal range); blood oxygen saturation >95% (normal range); blood flow velocity >20 cm / s (normal range); each physiological indicator is compared with the medical standards one by one to determine whether it exceeds the normal range. Based on the results of the item-by-item comparison, the patient's blood safety status is comprehensively judged; low risk: all physiological indicators are within the normal range; medium risk: some physiological indicators exceed the normal range, but not to a serious degree; high risk: multiple physiological indicators exceed the normal range, or one indicator deviates seriously from the normal range.

[0068] S3. Generate a dialysis plan based on the evaluation results and dynamically adjust dialysis parameters.

[0069] Based on the evaluation results, a dialysis plan is generated through the dialysis parameter derivation method;

[0070] It should be noted that the patient's blood safety status and risk type are extracted from the evaluation results, and the dialysis parameters are dynamically adjusted according to the dialysis parameter adjustment rules. For example, when the systolic blood pressure and blood flow velocity are lower than normal levels, the dialysate flow rate can be increased, the ultrafiltration volume can be decreased, and the anticoagulant dosage can be increased. The final dialysis plan is generated based on the adjusted dialysis parameters.

[0071] Monitor the patient's physiological indicators in real time during dialysis. When the physiological indicators are abnormal, dynamically adjust the dialysis parameters according to the abnormal situation.

[0072] It should be noted that the patient's physiological indicator data is collected every five minutes to determine whether each physiological indicator exceeds the normal range. If it is within the normal range, monitoring will continue; if it exceeds the normal range, it will be marked as abnormal; each physiological indicator is divided into mild abnormality, moderate abnormality and severe abnormality, and dialysis parameters are dynamically adjusted according to the abnormal situation.

[0073] S4. Collect dialysis effect data, analyze the effectiveness of the dialysis plan through statistical analysis methods, and generate a dialysis effect report.

[0074] Statistical analysis methods were used to analyze the changes in toxin clearance, fluid balance, and electrolyte balance before and after dialysis, and the dialysis effect was classified according to the changes in each indicator.

[0075] It should be noted that toxin clearance indicators include urea clearance rate (URR) and changes in serum urea nitrogen (BUN); fluid balance indicators include weight changes and ultrafiltration volume (UFV); electrolyte balance indicators include changes in serum potassium ion concentration, serum sodium ion concentration, and serum calcium ion concentration; the changes in the values ​​of each indicator before and after dialysis are calculated, and the reasonable range of changes in each indicator is defined according to the Kidney Disease Involved in Dialysis Guidelines (KDIGO) standards: URR greater than 65%, BUN reduced by 65%-70%, weight change value close to dry weight (within ±0.5kg), ultrafiltration volume needs to be combined with the specific situation of the patient, usually 1-2L, and serum potassium ion concentration changes between 3.5-5.5mmo l / L, serum sodium ion concentration changes between 135-145mmol / L, serum calcium ion concentration changes between 8.5-10.5mg / dL; according to the change value of each index, the dialysis effect is classified as follows: if the toxin clearance index reaches the target value (the change value of urea clearance rate is ≥10%, and the serum urea nitrogen is significantly reduced), the fluid balance index is within a reasonable range (the weight change is close to the dry weight, and the UFV meets the requirements), and the electrolyte balance index is stable within the normal range, the dialysis effect is good; if the toxin clearance index partially reaches the target value (urea clearance rate is ≥10%, and the serum urea nitrogen is significantly reduced), the dialysis effect is good. The dialysis effect is general if the change is 8%-10%, serum urea nitrogen decreases to a certain extent), the fluid balance index deviates slightly (weight change deviates from dry weight by 0.5-1.0 kg), and the electrolyte balance index deviates partially from the normal range; the dialysis effect is poor if the toxin clearance index does not reach the target value (urea clearance rate change <8%, serum urea nitrogen does not decrease significantly), the fluid balance index deviates seriously from the target value (weight change deviates from dry weight by >1 kg, UFV is too high or too low), and the electrolyte balance index deviates seriously from the normal range.

[0076] Generate a dialysis effect report based on the changes in each indicator and the dialysis effect classification results;

[0077] It should be noted that the dialysis effect report needs to record the patient number, dialysis date, change in urea clearance rate, change in serum urea nitrogen, change in body weight, ultrafiltration volume, change in serum potassium ion concentration, change in serum sodium ion concentration, change in serum calcium ion concentration, and dialysis effect classification; for patients with average or poor dialysis effect, dialysis improvement suggestions are put forward.

[0078] S5. Update the multimodal blood safety model based on the dialysis effect report and generate an optimized multimodal blood safety model.

[0079] Based on the changes in various indicators in the dialysis effect report, the structure, hyperparameters, and feature extraction layer weights of the multimodal blood safety model are updated;

[0080] It should be noted that based on the changes in various indicators in the dialysis effect report, we analyzed which indicators are most critical for predicting dialysis effects. The change in urea clearance rate indicates the change in urea clearance rate before and after dialysis, which directly reflects the ability of dialysis treatment to clear toxins from the patient's body. Serum potassium ion concentration is highly sensitive to dialysis treatment, and changes in serum potassium ion concentration can quickly reflect the electrolyte regulation effect during dialysis. Weight change reflects the fluid management effect of dialysis and ensures water balance in the patient's body. Therefore, urea clearance rate change, serum potassium ion concentration change, and weight change are key indicators for predicting dialysis effects. Hidden layers are added to capture the complex relationship between indicators. The learning rate is reduced to improve the convergence of the multimodal blood safety model, the regularization parameter is increased to prevent overfitting of the multimodal blood safety model, and the training iterations and batch size of the multimodal blood safety model are increased. The weights of the multimodal blood safety model are updated using the gradient descent method, the gradient of the cross-entropy loss function with respect to the weights of the feature extraction layer is calculated, and the weights are updated based on the gradient and learning rate.

[0081] The performance of the multimodal blood safety model is evaluated by regularly collecting dialysis effect reports and continuously optimized based on the evaluation results.

[0082] This embodiment also provides a management system for blood purification center nursing safety, including: a data acquisition module, a model building module, a dialysis plan generation module, a dialysis effect analysis module, and a model update module;

[0083] The data acquisition module is used to collect the patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis and perform preprocessing to generate a multimodal data set;

[0084] A model building module is used to build a multimodal blood safety model, predict physiological indicators based on multimodal data sets, and evaluate patient blood safety based on physiological indicators;

[0085] A dialysis plan generation module is used to generate a dialysis plan based on the evaluation results and dynamically adjust dialysis parameters;

[0086] The dialysis effect analysis module is used to collect dialysis effect data, analyze the effectiveness of the dialysis plan through statistical analysis methods, and generate a dialysis effect report;

[0087] The model updating module is used to update the multimodal blood safety model according to the dialysis effect report and generate an optimized multimodal blood safety model.

[0088] This embodiment also provides a computer device suitable for the management method of nursing safety in a blood purification center, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the management method of nursing safety in a blood purification center proposed in the above embodiment.

[0089] The computer device may be a terminal, comprising a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device comprises a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner may be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a button, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse.

[0090] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, it implements the management method for achieving nursing safety in a blood purification center as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0091] In summary, the present invention achieves accurate assessment of the patient's blood safety status and generates personalized dialysis plans through: multimodal data fusion and construction of a multimodal blood safety model, thereby improving the accuracy of blood safety assessment and the personalization of dialysis plans, and significantly enhancing the safety and effectiveness of dialysis treatment; through multi-indicator comprehensive evaluation and classification methods, a comprehensive analysis and intuitive expression of dialysis effects are achieved, providing reliable data support for the dynamic adjustment of dialysis plans, and improving the scientific nature and operability of dialysis treatment.

[0092] 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 the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for managing nursing safety in a blood purification center, characterized by: include, The patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis are collected and preprocessed to generate a multimodal dataset; Build a multimodal blood safety model, predict physiological indicators based on multimodal data sets, and evaluate patient blood safety based on physiological indicators; Generate a dialysis plan based on the evaluation results and dynamically adjust dialysis parameters; Collect dialysis effect data, analyze the effectiveness of the dialysis plan through statistical analysis methods, and generate a dialysis effect report; According to the dialysis effect report, the multimodal blood safety model is updated to generate an optimized multimodal blood safety model.

2. The method for managing nursing safety in a blood purification center according to claim 1, characterized in that: The preprocessing includes missing value filling, outlier removal and data normalization; The preprocessed initial physiological data, initial dialysis parameters, and vascular access status data were aligned according to timestamps and merged into a multimodal dataset; Health status categories are added to the multimodal data at each time point in the multimodal dataset according to standardized medical assessment methods.

3. The method for managing nursing safety in a blood purification center according to claim 2, characterized in that: The specific steps of constructing the multimodal blood safety model are as follows: The input layer is defined using a fully connected layer, and the feature extraction layer is defined using a convolutional neural network model and a long short-term memory network model; The health status prediction layer is defined by the fully connected layer and the Softmax activation function, and the output layer is defined by the result integration method; A multimodal blood safety model is constructed based on the input layer, feature extraction layer, health status prediction layer and output layer.

4. The method for managing nursing safety in a blood purification center according to claim 3, characterized in that: The specific steps of predicting physiological indicators based on multimodal data sets and evaluating the patient's blood safety based on the physiological indicators are as follows: The multimodal dataset is input into the multimodal blood safety model, and physiological indicators are generated through forward propagation of the multimodal blood safety model; Assess the patient's blood safety status through direct assessment method.

5. The method for managing nursing safety in a blood purification center according to claim 4, characterized in that: The dialysis plan is generated according to the evaluation results, and the dialysis parameters are dynamically adjusted. The specific steps are as follows: Based on the evaluation results, a dialysis plan is generated through the dialysis parameter derivation method; The patient's physiological indicators are monitored in real time during the dialysis process. When the physiological indicators are abnormal, the dialysis parameters are dynamically adjusted according to the abnormal situation.

6. The method for managing nursing safety in a blood purification center according to claim 5, characterized in that: The dialysis effect data is collected, the effectiveness of the dialysis program is analyzed by statistical analysis methods, and a dialysis effect report is generated. The specific steps are as follows: Statistical analysis methods were used to analyze the changes in toxin clearance, fluid balance, and electrolyte balance before and after dialysis, and the dialysis effect was classified according to the changes in each indicator. A dialysis effect report is generated based on the changes in each indicator and the dialysis effect classification results.

7. The method for managing nursing safety in a blood purification center according to claim 6, characterized in that: The multimodal blood safety model is updated according to the dialysis effect report to generate an optimized multimodal blood safety model. The specific steps are as follows: Based on the changes in various indicators in the dialysis effect report, the structure and hyperparameters of the multimodal blood safety model and the weights of the feature extraction layer are updated; The performance of the multimodal blood safety model is evaluated by regularly collecting dialysis effect reports and continuously optimized based on the evaluation results.

8. A blood purification center nursing safety management system, based on the blood purification center nursing safety management method according to any one of claims 1 to 7, characterized in that: Including data acquisition module, model building module, dialysis plan generation module, dialysis effect analysis module and model update module; The data acquisition module is used to collect the patient's initial physiological data, initial dialysis parameters, and vascular access status data at the start of dialysis and perform preprocessing to generate a multimodal data set; A model building module is used to build a multimodal blood safety model, predict physiological indicators based on multimodal data sets, and evaluate patient blood safety based on physiological indicators; A dialysis plan generation module is used to generate a dialysis plan based on the evaluation results and dynamically adjust dialysis parameters; The dialysis effect analysis module is used to collect dialysis effect data, analyze the effectiveness of the dialysis plan through statistical analysis methods, and generate a dialysis effect report; The model updating module is used to update the multimodal blood safety model according to the dialysis effect report and generate an optimized multimodal blood safety model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for managing nursing safety in a blood purification center according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for managing nursing safety in a blood purification center according to any one of claims 1 to 7 are implemented.

Images