Alarm method and device based on telemedicine conference, medium and program product

Abstract

The embodiment of the invention discloses an alarm method and device based on a telemedicine conference, a medium and a program product. The method comprises the following steps: performing data extraction and classification on consultation records of a telemedicine conference, collected historical medical records and environment detection records, and obtaining medical data and context awareness data of a patient; acquiring disease probability distribution based on the noise gate Bayesian network and the medical data; acquiring a feature vector matrix for the context awareness data and the disease probability distribution through an interface layer; acquiring a disease type, a severity parameter and an emergency degree parameter of the patient based on the connection heuristic network and the feature vector matrix so as to determine whether the patient has a health risk, and if yes, selecting an optimal receiver; and sending alarm information to the optimal receiver. The method can dynamically adapt to complex and changeable medical situations and environmental factors in combination with the use of a remote medical conference, and carries out intelligent alarm, so as to improve the accuracy and response speed of patient health risk identification.

Description

Technical Field

[0001] This disclosure relates to the field of medical and health technology, and in particular to an alarm method, device, medium, and program product based on telemedicine conferencing. Background Technology

[0002] With the rapid development of telemedicine technology, telemedicine conferencing has become a common medical service model, especially in situations where medical resources are limited or patients are unable to visit the hospital, where telemedicine can provide efficient and convenient medical services. However, in telemedicine conferencing, how to quickly and accurately identify patients' health risks and promptly notify relevant medical staff remains a challenge. Traditional alerting methods are usually based on single-dimensional data, such as patients' vital signs or medical history, lacking comprehensive analysis of multi-source data, leading to untimely or inaccurate alerts. Summary of the Invention

[0003] In view of this, the present disclosure provides an alarm method, device, medium, and program product based on telemedicine conferencing, which can be combined with the use of telemedicine conferencing to dynamically adapt to complex and ever-changing medical situations and environmental factors, and perform intelligent alarms to improve the accuracy and response speed of patient health risk identification.

[0004] In a first aspect, the present disclosure provides an alarm method based on remote medical conferencing, employing the following technical solution: Collect consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records; Data extraction and classification are performed on the consultation records, historical medical records, and environmental monitoring records to obtain the patient's medical data and context-aware data; Construct a noise-gated Bayesian network, an interface layer, and a connection-heuristic network; Based on the noisy gate Bayesian network and the medical data, the patient's disease risk is assessed to obtain the disease probability distribution; The interface layer is used to standardize the context-aware data and the disease probability distribution to obtain a feature vector matrix. Based on the connection heuristic network and the feature vector matrix, the patient's disease is analyzed to obtain the patient's disease type, severity parameters, and urgency parameters. When a patient is determined to have a health risk based on the disease type, the severity parameter, and the urgency parameter, the optimal recipient is selected from the medical staff participating in the telemedicine conference. Send an alarm message to the optimal recipient.

[0005] Optionally, the step of assessing the patient's disease risk based on the noisy gate Bayesian network and the medical data to obtain the disease probability distribution includes: The medical data is preprocessed; The time window analysis algorithm is used to divide the preprocessed medical data into multiple datasets; Feature extraction and dimensionality reduction are performed on each dataset to obtain multiple sets of variables; The noisy gate Bayesian network was adjusted using each set of variables to obtain multiple adjusted noisy gate Bayesian networks; Multiple disease probability distributions are obtained based on several adjusted noise gate Bayesian networks.

[0006] Optionally, the interface layer includes an input module, a data processing module, a standardization module, a feature fusion module, and an output module; The input module is used to receive the context-aware data and the multiple disease probability distributions; The data processing module is used to convert the context-aware data into a first time series and the multiple disease probability distributions into a second time series. The standardization module is used to standardize the first time series to obtain a first matrix, and to standardize the second time series to obtain a second matrix; The feature fusion module is used to fuse the first matrix and the second matrix to obtain the feature vector matrix; The output module is used to output the feature vector matrix to the connection heuristic network.

[0007] Optionally, feature extraction is performed on the context-aware data to obtain a feature set; The feature set is divided into multiple feature subsets using a time window analysis algorithm. Each feature subset is converted into a vector sequence, and multiple vector sequences constitute the first time series. The expression for the first time series is: ; Where S is the first time series; Let be the vector sequence at time point t; T is the total number of time points in the time series, t∈[1,T]; The expression for the second time series is: ; Where P is the second time series; Let be the disease probability distribution at time point t.

[0008] Optionally, the expression for the eigenvector matrix is: ; Where F is the eigenvector matrix; represents the components of the eigenvector matrix at time point t; ; in, Let be the attention weight at time point t; For standardization ; For standardization ; ; in, This is the normalization function; for transpose; for Dimensions.

[0009] Optionally, a risk score is obtained based on the disease type, the severity parameter, and the urgency parameter; If the risk score reaches a preset risk threshold, the patient is determined to have a health risk. If the risk score does not reach the preset risk threshold, the patient is determined to have no health risk. The formula for calculating the risk score is as follows: ; Where A is the risk score; The preset weight values; For the severity parameter of the patient's i-th disease; For the patient's i-th disease, this is a parameter indicating the urgency level. The weights of the severity parameter for the i-th disease; is the weight of the urgency parameter of the i-th disease; I is the total number of disease types suffered by the patient, i∈[1,I].

[0010] Optionally, selecting the optimal recipient from among the healthcare professionals participating in the telemedicine conference includes: Obtain the comprehensive matching score between each medical staff member and patient, and select the medical staff member with the highest comprehensive matching score as the optimal recipient; The formula for calculating the comprehensive matching score is as follows: ; Where Z is the overall matching score; , , , A is the preset weight value; B is the distance matching score; C is the professional matching score; D is the workload matching score; E is the interaction matching score. ; Where d represents the distance between medical staff and patients; ; in, Let the professional matching degree between medical staff and patients for the i-th disease be defined as follows: if there is an element in the medical staff's expertise vector sequence that matches the i-th disease, then... If the sequence of medical staff's expertise vectors does not contain an element that matches the i-th disease, then ; ; in, The number of future working hours until the day of the telemedicine meeting; For the workload of medical staff in the future working hours, k1 and k2 are preset adjustment coefficients; ; in, The quality score for the m-th interaction between healthcare workers and patients; Let m be the weight of the m-th interaction; M is the total number of interactions between medical staff and patients, m∈[1,M]; These are preset weight values.

[0011] Secondly, this disclosure also provides an alarm system based on remote medical conferencing, which adopts the following technical solution: The record collection module is used to collect consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records. The data extraction module is used to extract and classify the consultation records, the historical medical records and the environmental monitoring records to obtain the patient's medical data and context-aware data. The network building module is used to build noisy gate Bayesian networks, interface layers, and connection-heuristic networks. The disease risk assessment module is used to assess the disease risk of patients based on the noise gate Bayesian network and the medical data, and to obtain the disease probability distribution. The data standardization module is used to standardize the context-aware data and the disease probability distribution through the interface layer to obtain a feature vector matrix; The disease analysis module is used to perform disease analysis on patients based on the connection heuristic network and the feature vector matrix, and to obtain the patient's disease type, severity parameters and urgency parameters. The personnel selection module is used to select the optimal recipient from among the medical staff participating in the telemedicine conference when it is determined that a patient has a health risk based on the disease type, the severity parameter, and the urgency parameter. The alarm information sending module is used to send alarm information to the optimal recipient.

[0012] Thirdly, this disclosure also provides a computer device, which adopts the following technical solution: The computer device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor, which enables the at least one processor to perform any of the above-described alarm methods based on telemedicine conferencing.

[0013] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium storing computer instructions for causing a computer to execute any of the above-described alarm methods based on telemedicine conferencing.

[0014] Fifthly, embodiments of this disclosure also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of any of the methods described above.

[0015] The alarm method based on telemedicine conferences provided in this disclosure achieves multi-source data fusion by collecting consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records. This helps to more comprehensively understand the patient's health status and environment, thereby improving the accuracy of disease risk assessment. Utilizing a noisy gate Bayesian network to analyze medical data enables dynamic acquisition of disease probability distributions. Data standardization of context-aware data and disease probability distributions is performed at the interface layer to generate a feature vector matrix, laying the foundation for subsequent disease analysis using a connection-heuristic network and ensuring unified processing and effective utilization of different types of data. Disease analysis based on the connection-heuristic network and feature vector matrix can acquire the patient's disease type, severity parameters, and urgency parameters, concretizing the patient's specific situation and the severity and urgency of their needs. This facilitates subsequent detection of whether the patient has health risks. Once a health risk is identified, the optimal recipient can be intelligently selected, ensuring that alarm information is received and processed promptly by the most suitable medical personnel, improving response speed and efficiency.

[0016] The above description is merely an overview of the technical solution disclosed herein. In order to better understand the technical means of this disclosure and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this disclosure more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of this disclosure, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating an alarm method based on telemedicine conferencing provided in an embodiment of this disclosure; Figure 2 A flowchart illustrating the method for constructing a noisy gate Bayesian network provided in this embodiment of the disclosure; Figure 3 A schematic diagram of an alarm system based on remote medical conferencing provided in this embodiment of the disclosure; Figure 4 This is a schematic diagram of the structure of a computer device provided in an embodiment of the present disclosure. Detailed Implementation

[0019] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0020] It should be understood that the following specific examples illustrate the implementation of this disclosure, and those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0021] It should be noted that various aspects of embodiments within the scope of the appended claims are described below. It will be apparent that the aspects described herein can be embodied in a wide variety of forms, and any particular structure and / or function described herein is merely illustrative. Based on this disclosure, those skilled in the art will understand that one aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number of aspects set forth herein can be used to implement the device and / or practice the method. Additionally, this device and / or method can be implemented using structures and / or functionalities other than one or more of the aspects set forth herein.

[0022] It should also be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this disclosure. The drawings only show the components related to this disclosure and are not drawn according to the number, shape and size of the components in actual implementation. In actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0023] Furthermore, specific details are provided in the following description to facilitate a thorough understanding of the examples. However, those skilled in the art will understand that the described aspects can be practiced without these specific details.

[0024] Reference Figure 1 This disclosure provides an alarm method based on telemedicine conferencing, including the following steps: S1: Collect consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records; S2: Extract and classify data from consultation records, historical medical records, and environmental monitoring records to obtain patient medical data and context-aware data; S3: Construct a noisy gate Bayesian network, an interface layer, and a connection-heuristic network; S4: Based on noisy gate Bayesian networks and medical data, assess patients' disease risk and obtain disease probability distributions; S5: Standardize context-aware data and disease probability distribution through the interface layer to obtain a feature vector matrix; S6: Based on connection-heuristic networks and feature vector matrices, perform disease analysis on patients to obtain the patient's disease type, severity parameters, and urgency parameters; S7: When a patient is determined to have a health risk based on disease type, severity parameters, and urgency parameters, select the optimal recipient from among the healthcare professionals participating in the telemedicine conference; S8: Send alarm information to the optimal recipient.

[0025] This disclosed alarm method based on telemedicine conferences achieves multi-source data fusion by collecting consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records. This helps to gain a more comprehensive understanding of the patient's health status and environment, thereby improving the accuracy of disease risk assessment. Utilizing a noisy gate Bayesian network to analyze medical data enables the dynamic acquisition of disease probability distributions. Through an interface layer, context-aware data and disease probability distributions are standardized to generate a feature vector matrix, laying the foundation for subsequent disease analysis using connection-heuristic networks and ensuring unified processing and effective utilization of different data types. Disease analysis based on connection-heuristic networks and feature vector matrices can acquire the patient's disease type, severity parameters, and urgency parameters, concretizing the patient's specific situation and the severity and urgency of their needs. This facilitates subsequent detection of health risks and, upon confirming a health risk, intelligently selects the optimal recipient, ensuring that alarm information is received and processed promptly by the most suitable medical personnel, improving response speed and efficiency.

[0026] In summary, intelligent data processing and analysis, along with the combination of telemedicine conference recordings and patient historical data, improve the accuracy and timeliness of disease diagnosis and alerts, thereby enhancing the overall quality of medical services and patient satisfaction. This also helps reduce potential risks during the medical process and safeguards patient health and safety.

[0027] In S1, during telemedicine consultations, patients and multiple medical professionals can communicate in real time via video conferencing. The system monitors data and call quality throughout the entire consultation process to ensure smooth communication and accurate diagnosis. Simultaneously, utilizing sensors and environmental monitoring equipment, the system can detect environmental factors such as temperature, humidity, altitude, and Air Quality Index (AQI) of the patient's environment and automatically record this data in the environmental monitoring log. During or after the consultation, medical professionals will complete a detailed consultation record, including the doctor's diagnostic report, discussion content, treatment recommendations, and the results of inquiries about the patient's current condition. When the doctor submits the consultation record, the system also collects the patient's historical medical records, including electronic medical records, CT scan results, laboratory test reports, past medical history, family medical history, and allergy history, to provide patients with more comprehensive medical services.

[0028] In S2, biometric parameters (such as heart rate, blood pressure, and blood sugar), laboratory test results (such as blood test results), medication records (drug name, dosage, and frequency of use), genetic diseases, allergens, patient activity status (resting / mild activity / moderate activity / vigorous activity), recent dietary habits (fasting / light meals / normal / large meals), sleep quality (poor / fair / good), work hours, and mental state (relaxed / high stress) are extracted from consultation records and historical medical records. Environmental factor data is extracted from environmental monitoring records. Edge computing technology is used to perform preliminary processing and screening of the extracted data at the data source, such as using medical data standards like HL7 FHIR, to ensure data interoperability between different systems and reduce network transmission pressure.

[0029] The extracted data are classified according to type, and the patient's medical data and context-aware data are constructed according to the classification results. The medical data includes at least one of the following: vital signs (heart rate, blood pressure, blood sugar, etc.), laboratory test results, and medication records. The context-aware data includes at least one of the following: physiological indicators (heart rate, blood pressure, blood oxygen saturation, body temperature, respiratory rate, etc.), environmental factor data, and context factor data (patient activity status, recent diet, working hours, mental state, etc.).

[0030] In S3, refer to Figure 2 The flowchart illustrating the method for constructing a noisy gate Bayesian network shows the following steps: S31: Collect disease data from multiple confirmed patients and create multiple network nodes based on the disease data; S32: Construct a directed acyclic graph based on the dependencies between multiple network nodes; S33: Establish a noise gate conditional probability table for each network node; S34: The network framework is composed of a directed acyclic graph and a noise gate conditional probability table; S35: Fine-tune the network framework using labeled datasets to obtain a noisy gate Bayesian network.

[0031] In S31, electronic health records of confirmed patients are collected from medical institutions such as hospitals and clinics, as well as from publicly available medical databases. Disease data, including information on confirmed diseases such as heart disease, diabetes, and cancer, symptom data such as fever, cough, chest pain, and shortness of breath, and risk factors such as age, gender, family medical history, lifestyle (smoking, drinking, exercise), and laboratory test results, are extracted from these electronic health records. These data are preprocessed to remove missing values ​​and outliers, and data from different sources are standardized and normalized to ensure data consistency. After processing, network nodes are defined according to data type, including disease nodes, symptom nodes, and risk factor nodes.

[0032] In S32, the dependencies, i.e. causal relationships, between network nodes are determined through consultation with medical experts and data mining techniques (such as PC algorithm and Hill-Climbing algorithm). A directed acyclic graph (DAG) is constructed using the determined dependencies to ensure that there are no loops in the graph. Graph theory algorithms (such as topological sorting) are used to verify and adjust the DAG to ensure its rationality.

[0033] In S33, conditional probabilities are estimated using historical frequency statistics based on medical expert knowledge and data, thus constructing a Conditional Probability Table (CPT) for each network node. A noise gate model is used to further simplify the CPT. Specifically, each CPT includes several parent nodes, each parent node connects to multiple original child nodes, and each original child node has a probability value. The noise gate model assumes that each parent node independently influences its original child nodes, assigning an independent influence probability to each parent node, and introducing leakage probabilities to represent the influence of unknown factors not captured by the model. This means adding a leakage node to each parent node, and the probability value of this leakage node is called the leakage probability. The sum of the probability values ​​of the original child nodes and the leakage node of each parent node equals 1. This approach reduces the storage and computational complexity of the CPT while maintaining the model's predictive accuracy.

[0034] In S34, DAG and CPT are combined to form a preliminary network framework. Simulated data or small-scale real data are used to verify the correctness and stability of the network framework and make timely adjustments.

[0035] In S35, a labeled medical dataset is collected, ensuring that each sample in the dataset has a clear disease label and related symptoms and risk factor information. Bayesian network learning algorithms (such as maximum likelihood estimation and Bayesian estimation) are used to fine-tune the probability values ​​in CPT. Cross-validation or hold-out validation methods are employed to evaluate the accuracy, sensitivity, and specificity of the fine-tuned network framework. Based on the evaluation results, the network framework is further optimized to obtain the final noisy-gate Bayesian network.

[0036] In S4, medical data is preprocessed, including standardization and normalization. Time window analysis techniques (such as sliding windows) are used to divide the preprocessed medical data into multiple datasets, each representing a set of data within a time window.

[0037] Principal component analysis (PCA) or autoencoders are used to extract features and reduce dimensionality for each dataset, resulting in multiple concise and representative variable sets. These variable sets retain the main information from the original dataset while reducing redundancy and noise.

[0038] The noisy-gate Bayesian network is adjusted using each set of variables, including adding or removing nodes in the directed acyclic graph (DAG), updating dependencies between nodes, and re-estimating the conditional probability tables for each node in the DAG, particularly the independent influence probability and leakage probability of the noisy gate model. These probability values ​​can be updated using maximum likelihood estimation (MLE) or Bayesian estimation to reflect the latest data features and medical knowledge. Based on the adjusted noisy-gate Bayesian network, effective disease risk assessment can be performed, obtaining the patient's disease probability distribution, i.e., the probability that the patient may have the disease. Since the multiple sets of variables are temporally ordered, the multiple disease probability distributions obtained from these sets can reflect the probability that the patient may have the disease at different times.

[0039] In S3, an interface layer is created, which includes an input module, a data processing module, a standardization module, a feature fusion module, and an output module. The input module receives context-aware data and multiple disease probability distributions. The data processing module converts the context-aware data into a first time series and the multiple disease probability distributions into a second time series. The standardization module standardizes the first time series to obtain a first matrix and the second time series to obtain a second matrix. The feature fusion module fuses the first and second matrices to obtain a feature vector matrix. The output module outputs the feature vector matrix for use by connecting the heuristic network.

[0040] In S5, correlation analysis or principal component analysis (PCA) algorithms are used to extract features from context-aware data to obtain a feature set. A time window analysis algorithm (such as a fixed window or sliding window algorithm) is used to divide the feature set into multiple feature subsets. It is required that the time series formed by multiple feature subsets be consistent with the time series formed by multiple disease probability distributions. Each feature subset contains multiple characteristic vectors. Each feature subset is converted into a vector sequence. The multiple vector sequences obtained are used to form a time series, which is denoted as the first time series.

[0041] The expression for the first time series is: ; In the formula, S is the first time series; Let be the vector sequence at time point t; T is the total number of time points in the time series, t∈[1,T].

[0042] ; In the formula, Let J be the j-th feature vector at time t; J is the total number of feature vectors at time t, j∈[1, J].

[0043] The expression for the second time series is: ; In the formula, P is the second time series; Let be the disease probability distribution at time point t.

[0044] ; In the formula, Let N be the probability that a patient has the nth disease in the disease probability distribution at time t; N is the total number of disease types in the disease probability distribution at time t, which is the number of different diseases that a patient may have at time t, where n∈[1,N].

[0045] Using the z-score normalization method, the first time series is transformed into the first matrix, which is expressed as follows: ; In the formula, This is the first matrix; For standardization .

[0046] ; In the formula, For standardization .

[0047] ; In the formula, for The mean; for The standard deviation.

[0048] The z-score normalization method is used to transform the second time series into a second matrix, the expression of which is: ; In the formula, This is the second matrix; For standardization .

[0049] ; In the formula, For standardization .

[0050] ; In the formula, for The mean; for The standard deviation.

[0051] Calculate the attention weight at each time point, and fuse the first and second matrices based on the attention weights to obtain the feature vector matrix. The expression for the feature vector matrix is ​​as follows: ; In the formula, F is the eigenvector matrix; Let be the components of the eigenvector matrix at time point t.

[0052] ; In the formula, Let be the attention weight at time point t, that is, at time point t. and Their relative importance.

[0053] ; In the formula, This is the normalization function; for transpose; for Dimensions.

[0054] In S3, a connection-inspired network is created, consisting of an input layer, a multi-layer perceptron (MLP) structure, and an output layer. The input layer receives the feature vector matrix. The MLP structure includes three hidden layers (fully connected layers), each followed by a ReLU activation function. The second hidden layer has 128 neurons, the third has 64 neurons, and the first has 32 neurons. The output layer has three output branches, each corresponding to a classification task: the first branch outputs the types of diseases the patient suffers from, the second branch outputs the severity parameters of the disease types, and the third branch outputs the urgency parameters of the disease types.

[0055] Optionally, when it is necessary to capture key features or key time points, an attention layer can be added to the connection-inspired network according to actual needs.

[0056] In S6, the feature vector matrix is ​​input into the connection-heuristic network to obtain several disease types of the patient, as well as severity and urgency parameters of several disease types. The severity and urgency parameters are specific score values.

[0057] In S7, in one possible implementation, if the severity parameter of any disease type reaches a preset severity threshold or the urgency parameter reaches a preset urgency threshold, then the patient is determined to have a health risk; if the severity parameter of each disease type does not reach the preset severity threshold and the urgency parameter does not reach the preset urgency threshold, then the patient is determined not to have a health risk.

[0058] In another possible implementation, a risk score is calculated based on the disease type, severity parameter, and urgency parameter. If the risk score reaches a preset risk threshold, the patient is determined to have a health risk; if the risk score does not reach the preset risk threshold, the patient is determined not to have a health risk.

[0059] The formula for calculating the risk score is as follows: ; In the formula, A is the risk score; These are preset weight values, specifically representing the proportion of severity in the overall risk assessment; This indicates the proportion of urgency in the overall risk assessment; For the severity parameter of the patient's i-th disease; For the patient's i-th disease, this is a parameter indicating the urgency level. Let be the weight of the severity parameter for the i-th disease. When the severity parameter falls within different numerical ranges... The value of will also be different; Let be the weight of the urgency parameter for the i-th disease. When the urgency parameter falls within different numerical ranges... The value of i will also vary; I is the total number of disease types suffered by the patient, i∈[1,I].

[0060] When a patient is identified as having health risks, the optimal recipient is selected from the medical staff participating in the telemedicine conference. Specifically, the comprehensive matching score between each medical staff member and the patient is obtained, and the medical staff member with the highest comprehensive matching score is selected as the optimal recipient.

[0061] The formula for calculating the overall matching score is as follows: ; In the formula, Z is the comprehensive matching score; , , , A is the preset weight value; B is the distance matching score; C is the professional matching score; D is the workload matching score; E is the interaction matching score.

[0062] ; In the formula, d represents the distance between medical staff and patients, and the unit used in this disclosure is kilometers; ; In the formula, Let the professional matching degree between medical staff and patients for the i-th disease be defined as follows: if there is an element in the medical staff's expertise vector sequence that matches the i-th disease, then... This indicates that a medical professional possesses expertise in treating the i-th disease. If there is no element in the medical professional's expertise vector sequence that matches the i-th disease, then... This indicates that the medical staff have expertise in treating the i-th disease.

[0063] ; In the formula, The number of future working hours until the day of the telemedicine meeting; For the workload of medical staff in the future working hours, k1 and k2 are preset adjustment coefficients. For example, if the telemedicine conference is on January 1, 2024, the following calculations are performed: If the timeframe is 30 hours, then the workload of medical staff in the next 30 hours is calculated starting from the day of the telemedicine conference. .

[0064] ; In the formula, The quality score for the m-th interaction between medical staff and patients represents the patient's satisfaction with the interaction; Let m be the weight of the m-th interaction. The value is determined by the type of interaction. For example, if the interaction involves developing a treatment plan for a patient or performing surgery, then... The value is relatively high, especially if it's for ward rounds or similar activities. The value of m is relatively low; M is the total number of interactions between medical staff and patients, m∈[1,M]; These are preset weight values.

[0065] Furthermore, the priority of alarms is determined based on the risk score, and different alarm sending channels are used for different priorities. For example, priority 1 is sent via email, priority 2 via SMS, and priority 3 via telephone. If a telemedicine conference is in progress, the telemedicine conference sending channel can also be selected, and the alarm information will be displayed on the telemedicine conference interface.

[0066] In S8, the optimal alarm method is selected based on the alarm priority, and the alarm information is sent to the optimal recipient according to the optimal alarm method. For example, if the optimal alarm method is SMS notification and the optimal recipient is the patient's attending physician, then the alarm information is sent to the attending physician in the form of SMS. The alarm information includes the patient's basic information, several types of diseases, severity parameters, urgency parameters, etc., so that the attending physician can formulate a medical plan for the patient.

[0067] Furthermore, a unified healthcare system is set up for medical staff. Within this system, all alarm information received by the medical staff is compiled into charts for easy access. Similarly, a unified patient system is set up for patients. Within this system, all alarm information for each patient, along with corresponding medical staff information, is compiled into charts, allowing patients to easily understand their health status and consult with relevant medical personnel.

[0068] Reference Figure 3 This disclosure provides an alarm system based on telemedicine conferencing, including: The record collection module 101 is used to collect consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records. Data extraction module 102 is used to extract and classify consultation records, historical medical records and environmental monitoring records to obtain patients' medical data and context-aware data; Network building module 103 is used to build noisy gate Bayesian networks, interface layers, and connection heuristic networks; The disease risk assessment module 104 is used to assess the disease risk of patients based on a noisy gate Bayesian network and medical data, and to obtain the disease probability distribution. The data standardization module 105 is used to standardize context-aware data and disease probability distribution through the interface layer to obtain a feature vector matrix. The disease analysis module 106 is used to perform disease analysis on patients based on connection heuristic networks and feature vector matrices, and to obtain the patient's disease type, severity parameters and urgency parameters. The personnel selection module 107 is used to select the optimal recipient from among the medical staff participating in the telemedicine conference when a patient is determined to have a health risk based on the disease type, severity parameters, and urgency parameters. The alarm information sending module 108 is used to send alarm information to the optimal recipient.

[0069] The various variations and specific examples of the alarm method based on telemedicine conferences provided above are also applicable to the alarm system based on telemedicine conferences provided in this disclosure. Through the foregoing detailed description of the alarm method based on telemedicine conferences, those skilled in the art can clearly understand the implementation method of the alarm system based on telemedicine conferences. For the sake of brevity, it will not be described in detail here.

[0070] A computer device according to embodiments of the present disclosure includes a memory and a processor. The memory is used to store non-transitory computer-readable instructions. Specifically, the memory may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may, for example, include random access memory (RAM) and / or cache memory. The non-volatile memory may, for example, include read-only memory (ROM), hard disk, flash memory, etc.

[0071] The processor may be a central processing unit (CPU) or other form of processing unit with data processing capabilities and / or instruction execution capabilities, and may control other components in the computer device to perform desired functions. In one embodiment of this disclosure, the processor is used to execute computer-readable instructions stored in the memory, causing the computer device to perform all or part of the steps of the alarm method based on telemedicine conferencing described in the foregoing embodiments of this disclosure.

[0072] Those skilled in the art will understand that, in order to solve the technical problem of how to achieve a good user experience, this embodiment may also include well-known structures such as communication buses and interfaces, and these well-known structures should also be included within the protection scope of this disclosure.

[0073] like Figure 4 This is a schematic diagram of a computer device provided for an embodiment of the present disclosure. It illustrates a structural schematic diagram suitable for implementing the computer device in the embodiments of the present disclosure. Figure 4 The computer device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0074] like Figure 4 As shown, a computer device may include a processor (such as a central processing unit, graphics processing unit, etc.), which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) or programs loaded from storage devices into random access memory (RAM). The RAM also stores various programs and data required for the operation of the computer device. The processor, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0075] Typically, the following devices can be connected to the I / O interface: input devices, such as sensors or visual information acquisition devices; output devices, such as displays; storage devices, such as magnetic tapes or hard drives; and communication devices. Communication devices allow the computer device to communicate wirelessly or wiredly with other devices (such as edge computing devices) to exchange data. Although Figure 4 A computer apparatus with various devices is shown, but it should be understood that it is not required to implement or have all of the devices shown. More or fewer devices may be implemented or included alternatively.

[0076] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from a storage device, or installed from a ROM. When the computer program is executed by a processor, all or part of the steps of the alarm method based on telemedicine conferencing according to embodiments of this disclosure are performed.

[0077] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0078] A computer-readable storage medium according to embodiments of the present disclosure stores non-transitory computer-readable instructions. When these non-transitory computer-readable instructions are executed by a processor, all or part of the steps of the alarm method based on telemedicine conferencing described in the foregoing embodiments of the present disclosure are performed.

[0079] The aforementioned computer-readable storage media include, but are not limited to: optical storage media (e.g., CD-ROM and DVD), magneto-optical storage media (e.g., MO), magnetic storage media (e.g., magnetic tape or portable hard drive), media with built-in rewritable non-volatile memory (e.g., memory card), and media with built-in ROM (e.g., ROM cartridge).

[0080] For a detailed description of this embodiment, please refer to the corresponding descriptions in the foregoing embodiments, which will not be repeated here.

[0081] The basic principles of this disclosure have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this disclosure are merely examples and not limitations, and should not be considered as essential features of each embodiment of this disclosure. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the scope of this disclosure to the necessity of employing the aforementioned specific details for implementation.

[0082] In this disclosure, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The block diagrams of devices, apparatuses, devices, and systems involved in this disclosure are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as "comprising," "including," "having," etc., are open-ended terms meaning "including but not limited to," and are used interchangeably with them. The terms "or" and "and" as used herein refer to the terms "and / or," and are used interchangeably with them unless the context clearly indicates otherwise. The term "such as" as used herein refers to the phrase "such as but not limited to," and is used interchangeably with it.

[0083] Additionally, as used herein, the "or" used in a list of items beginning with "at least one" indicates a separate list, such that a list of, for example, "at least one of A, B, or C" means A or B or C, or AB or AC or BC, or ABC (i.e., A and B and C). Furthermore, the word "exemplary" does not imply that the described example is preferred or better than other examples.

[0084] It should also be noted that in the systems and methods of this disclosure, the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered as equivalent solutions to this disclosure.

[0085] Various changes, substitutions, and modifications can be made to the technology described herein without departing from the teachings defined by the appended claims. Furthermore, the scope of the claims of this disclosure is not limited to the specific aspects of the processes, machines, manufactures, events, means, methods, and actions described above. Currently existing or later-developed processes, machines, manufactures, events, means, methods, or actions that perform substantially the same function or achieve substantially the same result as the corresponding aspects described herein can be utilized. Therefore, the appended claims include such processes, machines, manufactures, events, means, methods, or actions within their scope.

[0086] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use this disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of this disclosure. Therefore, this disclosure is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0087] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this disclosure to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. An alarm method based on remote medical conferencing, characterized in that, include: Collect consultation records from telemedicine conferences, patients' historical medical records, and environmental monitoring records; Data extraction and classification are performed on the consultation records, historical medical records, and environmental monitoring records to obtain the patient's medical data and context-aware data; Construct a noise-gated Bayesian network, an interface layer, and a connection-heuristic network; Based on the noisy gate Bayesian network and the medical data, the patient's disease risk is assessed to obtain the disease probability distribution; The interface layer is used to standardize the context-aware data and the disease probability distribution to obtain a feature vector matrix. Based on the connection heuristic network and the feature vector matrix, the patient's disease is analyzed to obtain the patient's disease type, severity parameters, and urgency parameters. When a patient is determined to have a health risk based on the disease type, the severity parameter, and the urgency parameter, the optimal recipient is selected from the medical staff participating in the telemedicine conference. Send an alarm message to the optimal recipient.

2. The alarm method based on remote medical conferencing according to claim 1, characterized in that, The process of assessing patient disease risk and obtaining disease probability distribution based on the noisy gate Bayesian network and the medical data includes: The medical data is preprocessed; The time window analysis algorithm is used to divide the preprocessed medical data into multiple datasets; Feature extraction and dimensionality reduction are performed on each dataset to obtain multiple sets of variables; The noisy gate Bayesian network was adjusted using each set of variables to obtain multiple adjusted noisy gate Bayesian networks; Multiple disease probability distributions are obtained based on several adjusted noise gate Bayesian networks.

3. The alarm method based on remote medical conferencing according to claim 2, characterized in that, The interface layer includes an input module, a data processing module, a standardization module, a feature fusion module, and an output module; The input module is used to receive the context-aware data and the multiple disease probability distributions; The data processing module is used to convert the context-aware data into a first time series and the multiple disease probability distributions into a second time series. The standardization module is used to standardize the first time series to obtain a first matrix, and to standardize the second time series to obtain a second matrix; The feature fusion module is used to fuse the first matrix and the second matrix to obtain the feature vector matrix; The output module is used to output the feature vector matrix to the connection heuristic network.

4. The alarm method based on remote medical conferencing according to claim 3, characterized in that, Feature extraction is performed on the context-aware data to obtain a feature set; The feature set is divided into multiple feature subsets using a time window analysis algorithm. Each feature subset is converted into a vector sequence, and multiple vector sequences constitute the first time series. The expression for the first time series is: ； Where S is the first time series; Let be the vector sequence at time point t; T is the total number of time points in the time series, t∈[1,T]; The expression for the second time series is: ； Where P is the second time series; Let be the disease probability distribution at time point t.

5. The alarm method based on remote medical conferencing according to claim 4, characterized in that, The expression for the eigenvector matrix is: ； Where F is the eigenvector matrix; represents the components of the eigenvector matrix at time point t; ； in, Let be the attention weight at time point t; For standardization ; For standardization ; ； in, This is the normalization function; for transpose; for Dimensions.

6. The alarm method based on remote medical conferencing according to claim 1, characterized in that, A risk score is obtained based on the disease type, the severity parameter, and the urgency parameter; If the risk score reaches a preset risk threshold, the patient is determined to have a health risk. If the risk score does not reach the preset risk threshold, the patient is determined to have no health risk. The formula for calculating the risk score is as follows: ； Where A is the risk score; The preset weight values; For the severity parameter of the patient's i-th disease; For the patient's i-th disease, this is a parameter indicating the urgency level. The weights of the severity parameter for the i-th disease; The weights of the urgency parameter for the i-th disease; I represents the total number of disease types the patient has, i∈[1,I].

7. The alarm method based on remote medical conferencing according to claim 6, characterized in that, The selection of the optimal recipient from among the medical personnel participating in the telemedicine conference includes: Obtain the comprehensive matching score between each medical staff member and patient, and select the medical staff member with the highest comprehensive matching score as the optimal recipient; The formula for calculating the comprehensive matching score is as follows: ； Where Z is the overall matching score; , , , A is the preset weight value; B is the distance matching score; C is the professional matching score; D is the workload matching score; E is the interaction matching score. ； Where d represents the distance between medical staff and patients; ； in, Let the professional matching degree between medical staff and patients for the i-th disease be defined as follows: if there is an element in the medical staff's expertise vector sequence that matches the i-th disease, then... If the sequence of medical staff's expertise vectors does not contain an element that matches the i-th disease, then ; ； in, The number of future working hours until the day of the telemedicine meeting; For the workload of medical staff in the future working hours, k1 and k2 are preset adjustment coefficients; ； in, The quality score for the m-th interaction between healthcare workers and patients; Let m be the weight of the m-th interaction; M is the total number of interactions between medical staff and patients, m∈[1,M]; These are preset weight values.

8. A computer device, characterized in that, The computer device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the alarm method based on telemedicine conferencing as described in any one of claims 1-7.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the alarm method based on telemedicine conferencing as described in any one of claims 1-7.

10. A computer program product comprising computer instructions, characterized in that, When executed by a processor, the computer instructions implement the steps of the method according to any one of claims 1-7.

Images