Novel intelligent dialysis machine system

Abstract

The invention relates to the field of intelligent dialysis, in particular to a novel intelligent dialysis machine system which comprises a dialysis machine, a handset and a monitoring board which are connected through hospital intranet wifi or the Internet or a server. The dialysis machine comprises a blood center management terminal, a data acquisition unit and a central processing module; the data acquisition unit records patient information from a hospital API (Application Program Interface); the central processing module adopts a support vector machine to extract patient features and analyze dialysis parameters of a patient; the monitoring board is used for displaying the information of the patient; according to the novel intelligent dialysis machine system, a quantitative basis is provided for intelligent regulation and control of dialysis parameters on the basis of a dialysis machine parameter optimization algorithm of a support vector machine, the use efficiency and diagnosis and treatment accuracy of medical equipment are improved, medical errors caused by improper parameter setting are reduced, and the service life of the medical equipment is prolonged. And the development of medical equipment towards the intelligentization and individuation directions is promoted, and the overall medical service quality is improved.

Description

Technical Field

[0001] The present invention belongs to the field of intelligent dialysis, and in particular relates to a novel intelligent dialysis machine system. Background Art

[0002] In the modern medical field, medical equipment is a key tool for diagnosis and treatment. The rationality of its parameter settings is directly related to the effectiveness of diagnosis and treatment. Taking the hemodialysis machine as an example, the blood flow rate is usually 200-400mL / min. It is adjusted according to the patient's vascular access and tolerance. The initial setting is generally 200-250mL / min, and gradually increased to 300-400mL / min. The dialysate flow rate is usually 500-800mL / min, with a standard setting of 500mL / min. It can be increased to 800mL / min as needed to improve dialysis efficiency. The ultrafiltration rate is usually 5-15mL / kg / h, and is adjusted according to the patient's dryness and blood pressure. The setting is based on weight and fluid retention, avoiding exceeding 15mL / kg / h to prevent hypotension and muscle cramps; the dialysis time is usually 3-5 hours / time, 3 times a week, and the standard dialysis time is 4 hours / time, 3 times a week, which can be adjusted according to the patient's specific situation; the dialysate temperature is usually 36.5℃ to 37.5℃, with the standard setting of 37℃, which can be fine-tuned according to the patient's body temperature and comfort; the dialysate is composed of 135-145mmol / L of sodium, 2.0-4.0mmol / L of potassium, 1.25-1.75mmol / L of calcium, 0.5-0.75mmol / L of magnesium, and 30-35mmol / L of bicarbonate. Adjustments are made based on the patient's electrolyte levels and acid-base balance to avoid electrolyte disturbances and acid-base imbalances;

[0003] Traditional medical equipment parameter setting methods mostly rely on manual operations and experience of medical staff. Although this method is feasible to a certain extent, it has obvious limitations. On the one hand, experience judgment lacks standardization and quantitative basis. The operations of different medical staff may lead to differences in parameter settings, which in turn affects the consistency of diagnosis and treatment; on the other hand, faced with complex and changeable clinical conditions, experience is difficult to fully cover and cannot accurately meet the personalized needs of each patient, resulting in low diagnosis and treatment efficiency and increased risks of misdiagnosis and missed diagnosis. Summary of the Invention

[0004] The purpose of the present invention is to overcome the defects in the background technology, such as the lack of standardization and quantitative basis for judgment, the operations of different medical staff may lead to differences in parameter settings, affecting the consistency of diagnosis and treatment, and facing complex and changeable clinical conditions, resulting in low diagnosis and treatment efficiency and increased risks of misdiagnosis and missed diagnosis, thereby realizing a new type of intelligent dialysis machine system.

[0005] To achieve the above-mentioned purpose, the technical solution of the present invention is: a new intelligent dialysis machine system, comprising

[0006] Dialysis machines, handheld devices, and monitoring dashboards are connected via the hospital's intranet Wi-Fi, the Internet, or a server;

[0007] The dialysis machine includes a blood center management terminal, a data acquisition unit and a central processing module;

[0008] The data collection unit inputs patient information from the hospital API interface;

[0009] The central processing module uses support vector machines to extract patient features and analyze the patient's dialysis parameters;

[0010] Monitoring board, used to display patient information;

[0011] The handheld device is used by medical staff to identify the patient's facial information and match it one-to-one with the above-mentioned entered patient information, as well as to display and adjust the above-mentioned dialysis machine parameters.

[0012] Specifically, the blood center management terminal includes a doctor terminal, a nurse terminal, and an engineer terminal;

[0013] The doctor side includes consultation module, case module, bed reservation module, inventory module, publicity management module and statistical analysis module;

[0014] The consultation module allows medical staff to search and view the dialysis status of dialysis patients on the same day, modify the parameters of the dialysis machine and upload the modified prescription;

[0015] Case module, used for patient file creation, tag screening, name search, case information import or export, and case card making;

[0016] The bed reservation module is used to reserve long-term or short-term beds for patients in the ward and query patient bed reservation information; the inventory module is used to query consumables information and manage consumables estimates and medicines;

[0017] Promotion management module, used to add and cancel promotional content and promotional records;

[0018] The statistical analysis module is used to collect statistics on patients' dialysis treatment, medicines and consumables, patient information, and hospital infection information. The nurse side includes the consultation module, case module, bed reservation module, and statistical analysis module:

[0019] The consultation module is used by nurses to verify patient information, dialysis prescriptions and dialysis drugs, and start the dialysis machine. The case module is used to create patient files, filter types, filter tags, search names, import and export case information, and create case cards.

[0020] The bed reservation module is used to query the bed availability in different wards and arrange beds for patients;

[0021] The statistical analysis module is used to collect statistics on patients' dialysis treatments, medicines and consumables, patient information, and hospital infection information. The engineer side includes the statistical analysis module, business management module, system management module, basic tool module, and monitoring module.

[0022] Statistical analysis module, used to record the maintenance status of dialysis machines;

[0023] Business management module, used for maintenance personnel to manage shifts in wards and set business parameters;

[0024] System management module, used for management of personnel in upper and lower levels of the maintenance department;

[0025] Basic tool module, used for task scheduling of maintenance personnel;

[0026] The monitoring module is used to view all online users in the user list.

[0027] Specifically, the handheld device includes a login module, a message display module, a scheduling module, a case module, a verification module, a ward round module, a shift handover module, a statistics module, and a management module that are communicatively connected to the blood center management terminal;

[0028] Login module, used for medical staff to log in to the above terminal login interface by entering a username and a login password that matches the username;

[0029] Message display module, used to search and display the content of scheduling, case, verification, ward rounds, handover, statistics and management on the above terminal login interface;

[0030] The scheduling module is used to schedule different medical staff. Medical staff can filter and query their own shifts according to wards and shifts.

[0031] Case module, used to collect and retrieve patient records;

[0032] Verification module, used to verify the treatment information entered by patients in different treatment ward areas;

[0033] The ward round module is used by doctors to query the ward round and examination records of patients in different wards;

[0034] The handover module is used for medical staff to browse and register the content of the handover;

[0035] Statistics module, used to record and analyze the conditions of patients on different dates;

[0036] The management module is used to register disinfection records and disinfection maintenance of different wards.

[0037] Specifically, the support vector machine uses radial basis kernel function and polynomial kernel to construct a support vector machine model to capture the nonlinear relationship between dialysis machine parameters and clinical conditions. The expression of the support vector machine model is:

[0038] Among them, α i , α i * is the Lagrange multiplier, K(x i ,x) is the kernel function, b is the bias term;

[0039] The kernel function adopts radial basis function, and its specific form is:

[0040] K(x i ,x j )=exp(-γ||x i -x j || 2 ) (2)

[0041] Among them, γ is the bandwidth parameter of the kernel function, which determines the scope of the kernel function and the similarity measure between data points; |x i -x j | represents sample x i and x j The Euclidean distance between

[0042] The mathematical expression of the polynomial kernel is:

[0043]

[0044] Where: γ is the scaling factor that controls the weight of the monomial; r is the bias term (constant term) that increases the flexibility of the model; x i T is x i The transpose of ; d is the degree of the polynomial, which determines the dimension of the feature space after mapping.

[0045] Specifically, the Z-score method is used to standardize the vector machine model data, and its formula is:

[0046]

[0047] Among them, x is the original data, μ is the mean of the data, σ is the standard deviation of the data, and x′ is the standardized data.

[0048] Specifically, a five-fold cross-validation method is used to adjust the hyperparameters of the support vector machine.

[0049] Specifically, a stratified sampling method is used to train the vector machine model and adjust its parameters.

[0050] Specifically, confusion matrix, precision and F1 score were used to evaluate the performance of the vector machine model in dialysis machine parameter adjustment.

[0051] Compared with the prior art, the novel intelligent dialysis machine system of the present invention has at least the following beneficial effects:

[0052] The algorithm for optimizing dialysis machine parameters based on a support vector machine in the present invention provides a quantitative basis for the intelligent regulation of dialysis parameters, improves the utilization efficiency and diagnosis and treatment accuracy of medical equipment, reduces medical errors caused by improper parameter settings, and promotes the development of medical equipment in the direction of intelligence and personalization, thereby improving the overall quality of medical services. The user enters an account name and a password matching the account name according to the user role to facilitate logging into the matching doctor, nurse, or engineer terminal in the blood center management terminal. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flow chart of the blood center management terminal of the present invention;

[0054] Figure 2 It is a doctor-side process diagram of the present invention;

[0055] Figure 3 It is a schematic diagram of the nurse side process of the present invention;

[0056] Figure 4 It is a schematic diagram of the engineer side process of the present invention;

[0057] Figure 5 It is a schematic diagram of the handheld device flow of the present invention;

[0058] Figure 6 Schematic diagram of confusion matrix in Model 1 of the present invention;

[0059] Figure 7 is a schematic diagram showing training and validation scores of a high learning curve in Model 1 of the present invention;

[0060] Figure 8 Schematic diagram of the confusion matrix in Model 2 of the present invention;

[0061] Figure 9 is a schematic diagram showing training and validation scores of the high learning curve in Model 2 of the present invention;

[0062] Figure 10 It is a schematic diagram of parameter sensitivity analysis of the vector machine model using the SHAP value of the present invention. DETAILED DESCRIPTION

[0063] The novel intelligent dialysis machine system of the present invention will be described in more detail below with reference to the accompanying drawings and through specific implementation methods.

[0064] In the description of the present invention, it should be understood that the terms "upper", "lower", "front", "back", "left", "right", "top", "bottom", "inside", "outside", etc., indicating directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0065] This embodiment discloses a new intelligent dialysis machine system, such as Figure 1-10 Shown, including

[0066] The dialysis machine, handheld device and monitoring board are connected via the hospital intranet wifi or the Internet or a server; the dialysis machine includes a blood center management terminal, a data acquisition unit and a central processing module;

[0067] The data collection unit inputs patient information from the hospital API interface, imports the patient information into the blood center management terminal and stores it;

[0068] The central processing module uses support vector machines to extract patient features and analyze the patient's dialysis parameters;

[0069] Monitoring board, used to display patient information;

[0070] The handheld device is used by medical staff to identify the patient's facial information and match it with the above-mentioned entered patient information and display dialysis parameters. The patient's facial information is collected by a camera in the existing technology and compared with the patient information entered by the above-mentioned blood center management terminal, so that the facial information of the user scanned by the handheld device is consistent with the patient information entered by the management terminal, which facilitates subsequent operations of the handheld device.

[0071] The blood center management terminal includes a doctor terminal, a nurse terminal, and an engineer terminal. Use Google Chrome and enter the corresponding URL to enter the login interface of the dialysis machine blood center management terminal. Enter the account name and password matching the account name according to the user role to log in to the matching doctor terminal, nurse terminal, or engineer terminal.

[0072] Log in to the doctor's page to display the doctor's personal information, dialysis status overview information and handover records; the dialysis status overview information includes the number of patients who have not signed in, are waiting for dialysis treatment, are undergoing dialysis treatment, and have completed dialysis treatment.

[0073] The doctor side includes consultation module, case module, bed reservation module, inventory module, publicity management module and statistical analysis module;

[0074] The consultation module allows medical staff to search and view the dialysis status of patients on that day's shift, modify dialysis machine parameters, and upload modified prescriptions;

[0075] The case module is used for patient file creation, label screening, name search, case information import or export, and case card making; name search is performed by setting up a search bar in the case module, and the patient's name is entered to query the patient's case; type screening is performed by setting up a filter drawer on one side of the search bar, and the searched patients are quickly filtered by filtering conditions; the above-mentioned patient information after file creation is stored together with the above-mentioned input information, and the patient's facial information is matched one by one with the patient information after file creation and the above-mentioned input information through scanning with the handheld camera.

[0076] The bed reservation module is used to reserve long-term or short-term beds for patients in the ward and query patient bed reservation information;

[0077] Inventory module, used to query consumables information and manage consumables estimates and medicines;

[0078] Promotion management module, used to add and cancel promotional content and promotional records;

[0079] The statistical analysis module is used to collect statistics on patients' dialysis treatment, medicines and consumables, patient information, and hospital infection information. The nurse side includes the consultation module, case module, bed reservation module, and statistical analysis module:

[0080] The consultation module is used by nurses to verify patient information, dialysis prescriptions and dialysis drugs, and start the dialysis machine. The case module is used to create patient files, filter types, filter tags, search names, import and export case information, and create case cards.

[0081] The bed reservation module is used to query the bed availability in different wards and arrange beds for patients;

[0082] The statistical analysis module is used to collect statistics on patients' dialysis treatments, medicines and consumables, patient information, and hospital infection information. The engineer side includes the statistical analysis module, business management module, system management module, basic tool module, and monitoring module.

[0083] Statistical analysis module, used to record the maintenance status of dialysis machines;

[0084] Business management module, used for maintenance personnel to manage shifts in wards and set business parameters;

[0085] System management module, used for management of personnel in upper and lower levels of the maintenance department;

[0086] Basic tool module, used for task scheduling of maintenance personnel;

[0087] The monitoring module is used to view all online users in the user list.

[0088] The handheld device includes a login module, a message display module, a scheduling module, a case module, a verification module, a ward round module, a shift handover module, a statistics module, and a management module that are communicatively connected to the terminal;

[0089] Login module, used for medical staff to log in to the above terminal login interface by entering a username and a login password that matches the username;

[0090] Message display module, used to search and display the content of scheduling, case, verification, ward rounds, handover, statistics and management on the above terminal login interface;

[0091] The scheduling module is used to schedule different medical staff. Medical staff can filter and query their own shifts according to wards and shifts.

[0092] Case module, used to collect and retrieve patient records;

[0093] Verification module, used to verify the treatment information entered by patients in different treatment ward areas;

[0094] The ward round module is used by doctors to query the ward round and examination records of patients in different wards;

[0095] The handover module is used for medical staff to browse and register the content of the handover;

[0096] Statistics module, used to record and analyze the conditions of patients on different dates;

[0097] The management module is used to register disinfection records and disinfection maintenance of different wards.

[0098] The support vector machine uses radial basis kernel function and polynomial kernel to construct a support vector machine model to capture the nonlinear relationship between dialysis machine parameters and clinical conditions. The expression of the support vector machine model is:

[0099]

[0100] Among them, α i , α i * is the Lagrange multiplier, K(x i ,x) is the kernel function, b is the bias term;

[0101] The kernel function adopts radial basis function, and its specific form is:

[0102] K(x i ,x j )=exp(-γ||x i -x j || 2 ) (2)

[0103] Among them, γ is the bandwidth parameter of the kernel function, which determines the scope of the kernel function and the similarity measure between data points; |x i -x j | represents sample x i and x j The Euclidean distance between

[0104] The mathematical expression of the polynomial kernel is:

[0105]

[0106] Where: γ is the scaling factor that controls the weight of the monomial; r is the bias term (constant term) that increases the flexibility of the model; x i T is x i The transpose of ; d is the degree of the polynomial, which determines the dimension of the feature space after mapping;

[0107] The polynomial kernel maps the original features to a high-dimensional space through polynomial expansion.

[0108] Since the dimensions and value ranges of different characteristic parameters vary greatly, the blood flow value range may be between 100 and 300, while the ultrafiltration volume range is between 0.01 and 5, and the value range is relatively small. This difference will affect the training effect and convergence speed of the vector model. Therefore, the Z-score method is used to standardize the vector machine model data. The formula is:

[0109]

[0110] Among them, x is the original data, μ is the mean of the data, σ is the standard deviation of the data, and x′ is the standardized data.

[0111] The five-fold cross-validation method was used to adjust the hyperparameters of the above support vector machine.

[0112] Using stratified sampling method, vector machine model training and parameter adjustment:

[0113] Confusion matrix, precision and F1 score are used to evaluate the performance of the vector machine model in dialysis machine parameter adjustment:

[0114] Assuming the category labels are A, B, and C, the confusion matrix is as follows

[0115]

[0116] Diagonal elements : The number of correctly classified samples;

[0117] Off-diagonal elements : The number of samples that are actually class X but are misclassified as class Y;

[0118] Accuracy:

[0119]

[0120] The above formula represents the proportion of samples predicted to be A that are actually A in the F1 score:

[0121]

[0122] The performance of the trained vector machine model is evaluated using the test set. The results are as follows: Figure 6-9 As shown:

[0123] Model 1

[0124] like Figure 6 He Ru Figure 7 As shown in the two figures, the model shows excellent accuracy. The high learning curve shows both training and validation scores exceeding 0.9, and the overall confusion matrix accuracy is approximately 91% ((7+5+27) / 43). The majority class (good) is perfectly classified, with all 27 examples correctly predicted with no confusion. Generalization is stable, with convergence of training and validation scores in the learning curve and minimal fluctuation in the shaded area, indicating that the model is not overfitting. The best cross-validation scores are RBF and F1 scores of 0.91, respectively.

[0125] Model 2

[0126] like Figure 8 He Ru Figure 9 As shown in the two figures, the results that can be drawn from the two figures are that the accuracy of the second model is high. The learning curve shows that the training and validation scores are both 0.9+, and the overall accuracy of the confusion matrix is about 86.3% (number of correctly predicted samples / total number of samples = (14+52+116) / (14+1+2+52+15+0+8+116) = 182 / 211); the majority class (good) classification effect is good, all 116 samples are correctly predicted without confusion; the generalization ability is stable; the training and validation scores in the learning curve converge, and the shadow area fluctuates little, indicating that the model is not overfitting, and the training score is The value (red) was always higher than 0.93 and finally stabilized at 0.94±0.01 (narrow shaded area), indicating that the model had a strong ability to fit the training data. The validation score (green) increased from 0.78 to 0.87±0.02 (wide shaded area), indicating good generalization ability and no signs of overfitting. The best cross-validation score RBF was 0.88, the F1 score was 0.87, and the test set precision was 0.88. The support vector machine model showed good performance in the regression task of medical device parameters and clinical image quality scores, and can provide a reliable basis for parameter optimization.

[0127] The SHAP value is used to perform parameter sensitivity analysis on the support vector machine model. The SHAP value provides an explanatory value for each feature by calculating the contribution of each feature to the model prediction result. The larger the absolute value of the SHAP value, the greater the influence of the feature on the model prediction result. For model 1: Figure 10 As shown, ultrafiltration volume is the most important feature in the model prediction, directly affecting the patient's fluid balance management or complication risk; the significant impact of vascular access type reflects the importance of access stability on treatment efficacy. Vascular access types in this case include but are not limited to autologous fistulas; the number of dialysis sessions and age indicate that the frequency and age of treatment are key factors in long-term prognosis; the influence of treatment mode and dialysis duration is relatively small.

[0128] For Model 2: Figure 10 As shown, ultrafiltration volume and blood pressure parameters are the core features across the two types of analysis, suggesting the global importance of fluid management and blood pressure control in model prediction; the significance of tube access type and treatment mode may reflect the impact of long-term treatment strategies on patient prognosis; the impact of treatment mode and dialysis time is relatively small; Model 2 is used to predict complications (such as hypotension, infection) or treatment response in dialysis patients, and medical staff need to pay attention to monitoring and intervention of high SHAP value features; optimization of autologous fistulas (such as surgical quality and maintenance) can directly improve the clinical outcomes predicted by Model 2.

[0129] The dialysis machine is connected to a variety of sensors, including pressure sensors, flow sensors, temperature sensors, and blood component sensors, which are used to monitor the data parameters of the patient's blood purification.

[0130] When the patient starts dialysis, the dialysis machine obtains data from Model 1, determines the patient's condition, and adjusts the corresponding data parameter settings; during the patient's dialysis process, the patient's blood purification record data is obtained in real time through the sensors on the dialysis machine, and Model 2 is used to determine the current patient's dialysis status. The dialysis machine parameters are adjusted in a timely manner to ensure that the dialysis patient is in good condition.

[0131] When the present invention is used, it comprises the following steps:

[0132] S1. Check the hemodialyzer and tubing for damage, install the tubing according to aseptic principles, and pre-flush the dialyzer and tubing with normal saline to remove all air and gas;

[0133] S2. Set dialysis parameters, select appropriate vascular access, perform puncture, and secure the puncture needle; connect the arterial and venous ends, start the blood pump, and begin dialysis treatment;

[0134] S3. Continuously monitor the patient's response and the parameters of the dialysis machine during dialysis, and record important data during dialysis and the patient's response;

[0135] S4. After dialysis is completed, perform closed blood return according to existing standards and clean and disinfect the dialysis machine.

[0136] It should be noted that the structures depicted in the drawings herein are not fixed, unchangeable implementations of the present invention in practice. The components of the embodiments of the present invention generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Furthermore, the drawings in this specification and the abstract are schematic only and do not represent the specific structure or actual quantities of the components in practice.

[0137] Unless otherwise defined, the technical or scientific terms used herein shall have the ordinary meaning understood by persons of ordinary skill in the field to which the invention belongs. The use of "one" or "an" and other similar words in the specification and claims of this application does not necessarily indicate a quantitative limitation. "Include" or "comprising" and other similar words mean that the elements or parts preceding the word include the elements or parts listed after the word and their equivalents, without excluding other elements or parts. "Connect" or "connected" and other similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect.

[0138] The exemplary embodiments of the present invention are described in detail above with reference to preferred embodiments. However, it will be understood by those skilled in the art that, without departing from the concept of the present invention, various variations and modifications may be made to the above-mentioned specific embodiments, and various combinations of the various technical features and structures proposed in the present invention may be made without exceeding the scope of protection of the present invention.

Claims

1. A new intelligent dialysis machine system, characterized by: include Dialysis machines, handheld devices, and monitoring dashboards are connected via the hospital's intranet Wi-Fi, the Internet, or a server; The dialysis machine includes a blood center management terminal, a data acquisition unit and a central processing module; The data collection unit inputs patient information from the hospital API interface; The central processing module uses support vector machines to extract patient features and analyze the patient's dialysis parameters; Monitoring board, used to display patient information; The handheld device is used by medical staff to identify the patient's facial information and match it one-to-one with the above-mentioned entered patient information, as well as to display and adjust the above-mentioned dialysis machine parameters.

2. The novel intelligent dialysis machine system according to claim 1, characterized in that: The blood center management terminal includes a doctor terminal, a nurse terminal and an engineer terminal; The doctor side includes consultation module, case module, bed reservation module, inventory module, publicity management module and statistical analysis module; The consultation module allows medical staff to search and view the dialysis status of dialysis patients on the same day, modify the parameters of the dialysis machine and upload the modified prescription; Case module, used for patient file creation, tag screening, name search, case information import or export, and case card making; The bed reservation module is used to reserve long-term or short-term beds for patients in the ward and query patient bed reservation information; Inventory module, used to query consumables information and manage consumables estimates and medicines; Promotion management module, used to add and cancel promotional content and promotional records; The statistical analysis module is used to collect statistics on patients' dialysis treatment, drugs and consumables, patient information, and hospital infection information. The nurse side includes the consultation module, case module, bed reservation module, and statistical analysis module. The consultation module is used by nurses to verify patient information, dialysis prescriptions, dialysis drugs, and start the dialysis machine. Case module, used for patient file creation, type screening, tag screening, name search, case information import and export, and case card creation; The bed reservation module is used to query the bed availability in different wards and arrange beds for patients; The statistical analysis module is used to collect statistics on patients' dialysis treatments, medicines and consumables, patient information, and hospital infection information. The engineer side includes the statistical analysis module, business management module, system management module, basic tool module, and monitoring module. Statistical analysis module, used to record the maintenance status of dialysis machines; Business management module, used for maintenance personnel to manage shifts in wards and set business parameters; System management module, used for management of personnel in upper and lower levels of the maintenance department; Basic tool module, used for task scheduling of maintenance personnel; The monitoring module is used to view all online users in the user list.

3. The novel intelligent dialysis machine system according to claim 1, characterized in that: The handheld device includes a login module, a message display module, a scheduling module, a case module, a verification module, a ward round module, a shift handover module, a statistics module, and a management module that are communicatively connected to the blood center management terminal; Login module, used for medical staff to log in to the above terminal login interface by entering a username and a login password that matches the username; Message display module, used to search and display the content of scheduling, case, verification, ward rounds, handover, statistics and management on the above terminal login interface; The scheduling module is used to schedule different medical staff. Medical staff can filter and query their own shifts according to wards and shifts. Case module, used to collect and retrieve patient records; Verification module, used to verify the treatment information entered by patients in different treatment ward areas; The ward round module is used by doctors to query the ward round and examination records of patients in different wards; The handover module is used for medical staff to browse and register the content of the handover; Statistics module, used to record and analyze the conditions of patients on different dates; The management module is used to register disinfection records and disinfection maintenance of different wards.

4. The novel intelligent dialysis machine system according to claim 1, characterized in that: The support vector machine uses radial basis kernel function and polynomial kernel to construct a support vector machine model to capture the nonlinear relationship between dialysis machine parameters and clinical conditions. The expression of the support vector machine model is: Among them, α i , α i * is the Lagrange multiplier, K(x i ,x) is the kernel function, b is the bias term; Kernel function K(x i ,x j ) uses radial basis function, the specific form is: K(x i ,x j )=exp(-γ||x i -x j || 2 ) (2) Among them, γ is the bandwidth parameter of the kernel function, which determines the scope of the kernel function and the similarity measure between data points; |x i -x j | represents sample x i and x j The Euclidean distance between The mathematical expression of the polynomial kernel is: Where: γ is the scaling factor that controls the weight of the monomial; r is the bias term (constant term) that increases the flexibility of the model; x i T is x i The transpose of ; d is the degree of the polynomial, which determines the dimension of the feature space after mapping; the polynomial kernel maps the original features to a high-dimensional space through polynomial expansion.

5. The novel intelligent dialysis machine system according to claim 4 is characterized in that: The Z-score method is used to standardize the vector machine model data, and the formula is: Among them, x is the original data, μ is the mean of the data, σ is the standard deviation of the data, and x′ is the standardized data.

6. The novel intelligent dialysis machine system according to claim 4, characterized in that: The five-fold cross-validation method was used to adjust the hyperparameters of the above support vector machine.

7. The novel intelligent dialysis machine system according to claim 4, characterized in that: The stratified sampling method is used to train the vector machine model and adjust its parameters.

8. The novel intelligent dialysis machine system according to claim 4, characterized in that: Confusion matrix, precision and F1 score were used to evaluate the performance of the vector machine model after the dialysis machine parameters were adjusted.

Images