Internet-based health care service system

Abstract

The invention relates to a health care service system based on the Internet, and relates to the technical field of health care, and the system comprises the steps: obtaining user voice data, carrying out the voice recognition through combining a private semantic library, and generating a hospital guide information flow; the hospital guide information flow is uploaded to a hybrid cloud architecture network for pathology screening, the current pathology state is obtained, and the hybrid cloud architecture network comprises a public cloud architecture and a private cloud architecture; based on the current pathological state, pathological state development prediction is carried out on a historical load response level, if a prediction result exceeds an early warning threshold value, continuous alarm is carried out, and nursing assistance is carried out based on alarm signal linkage positioning. The technical problems that the health care service data processing capacity is insufficient, the pathology screening precision is not high, personalized services are lacked, and nursing rescue linkage is not smooth are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of health care, and in particular to a health care service system based on the Internet. Background Art

[0002] With the rapid development and popularization of the Internet, people's health awareness is also increasing. The traditional health care service model often relies on offline medical institutions and medical staff, and there are problems such as uneven resource allocation, low service efficiency and poor patient medical experience. Especially in the face of scenarios such as chronic disease management and emergency medical assistance, the traditional model often cannot meet patients' needs for timely, accurate and personalized health care services. In recent years, with the continuous development and application of technologies such as cloud computing, big data, and artificial intelligence, new technical means and solutions have been provided for health care services. By using these advanced technologies, rapid processing and analysis of medical and health data can be achieved, providing a more scientific and accurate basis for disease prevention, diagnosis and treatment. At the same time, the introduction of Internet technology has also broken the time and space limitations of traditional health care services, allowing patients to obtain the required health care services anytime and anywhere. However, most of the health care service systems on the market currently have problems such as insufficient data processing capabilities, low pathology screening accuracy, lack of personalized services, and poor nursing and rescue linkage. Therefore, how to make full use of Internet technology and medical health data to build an efficient, convenient and personalized health care service system has become a technical problem that needs to be solved urgently. Summary of the invention

[0003] The present invention aims to solve the technical problems in the prior art of insufficient data processing capability of health care services, low pathology screening accuracy, lack of personalized services and poor nursing and rescue linkage, and provides an Internet-based health care service system to solve them.

[0004] The technical solution of the present invention to solve the above technical problems is as follows: The present invention provides an Internet-based health care service system, the execution steps of which include: obtaining user voice data, performing voice recognition in combination with a private semantic library, and generating a guidance information flow; uploading the guidance information flow to a hybrid cloud architecture network for pathological screening to obtain a current pathological state, wherein the hybrid cloud architecture network includes a public cloud architecture and a private cloud architecture; based on the current pathological state, predicting the development of the pathological state at a historical load response level, and issuing a continuous alarm if the prediction result exceeds a warning threshold, and providing nursing assistance based on alarm signal linkage positioning.

[0005] The beneficial effects of the present invention are: by acquiring user voice data to generate a guidance information flow, which is then uploaded to the hybrid cloud architecture network for pathology screening, and then development predictions are made based on the pathological state and an alarm is issued when necessary, and finally linkage positioning is performed for nursing assistance, which can improve the efficiency and accuracy of health care services and realize timely and accurate personalized health care services. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] Figure 1 A schematic flow chart of the execution steps of an Internet-based health care service system provided by the present invention.

[0007] Figure 2 A schematic flow chart of the execution steps for generating a guidance information flow in an Internet-based health care service system provided by the present invention. DETAILED DESCRIPTION

[0008] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0009] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0010] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or explanation". Any embodiment described as "for example" in the present invention is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any technician in the field to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes will not be elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed in the present invention.

[0011] Example: like Figure 1As shown, the embodiment of the present invention provides an Internet-based health care service system, and the execution steps include: S10: Obtain user voice data, perform voice recognition in combination with a private semantic library, and generate a guidance information flow.

[0012] S20: Upload the guidance information flow to the hybrid cloud architecture network for pathology screening to obtain the current pathology status, wherein the hybrid cloud architecture network includes a public cloud architecture and a private cloud architecture.

[0013] S30: Based on the current pathological state, the development of the pathological state is predicted at the historical load response level. If the prediction result exceeds the warning threshold, a continuous alarm is issued, and nursing assistance is performed based on the alarm signal linkage positioning.

[0014] Exemplarily, the user inputs voice data through a microphone to describe his or her symptoms, and this interactive process triggers the voice recognition function of the system. The system first obtains these voice data and uses a locally trained dialect database (i.e., a private semantic library) for voice recognition. The private database is optimized and trained for a specific dialect, so it can more accurately understand the user's voice input, especially those dialect words and expressions that may be difficult to be understood by a general voice recognition system. By combining the private semantic library for efficient and accurate semantic recognition, the system can convert the user's voice description into text data, and then generate a guidance information flow containing preliminary judgments on symptoms, causes of disease, and recommended examination items. This process not only improves the user experience, but also enables the system to provide users with subsequent health care services more quickly. For example, if a user describes his or her headache in dialect with vomiting symptoms, the system can accurately identify and generate the corresponding guidance information flow, providing strong support for subsequent medical diagnosis, especially for elderly people with slurred speech, interactive recognition based on a targeted and simple trained private semantic library can increase recognition efficiency and user experience, and improve their self-care ability.

[0015] Furthermore, after obtaining the guidance information flow of the user's symptoms, the system will upload it to a hybrid cloud architecture network for pathological screening to determine the current pathological status. This hybrid cloud architecture network integrates public cloud architecture and private cloud architecture. The public cloud architecture has powerful computing power and rich pathological data resources across regions and populations, which can process large-scale data and provide comprehensive pathological analysis; while the private cloud architecture provides higher data security and privacy protection, and is particularly suitable for storing and processing users' sensitive health information. The system will intelligently choose to upload the guidance information flow to the public cloud architecture or private cloud architecture for pathological screening based on the data volume, urgency, and current network load. For example, for non-urgent and small-data guidance information flows, the system may give priority to private cloud architecture for processing, so as to use localized pathological recognition models to quickly give a preliminary diagnosis. For urgent or large-data situations, the system will upload the guidance information flow to the public cloud architecture, use its powerful computing power for more in-depth analysis, and more accurately determine the current pathological status. This flexible selection mechanism ensures that the system can efficiently handle various complex health care needs while ensuring data security.

[0016] Finally, after determining the user's current pathological state, the system will further predict the development of the pathological state. This step is based on the user's historical load response level, that is, the physical response and treatment conditions under similar pathological states in the past. The system will collect and analyze these historical data, use the collected historical pathological data as the horizontal coordinate, and the pathological state development trend as the vertical coordinate to perform curve fitting on the pathological state development trend. In this way, a prediction function curve is obtained, which is actually an intuitive presentation of a mathematical model that reflects the development law of the disease that is adapted to the user based on historical data. For example, in the case of diabetes, the curve shows the trend of blood sugar levels changing over time and with different treatment methods, allowing the system to clearly see the development pattern of the disease under different conditions. However, since the adaptability of different individuals is different, individualized analysis is required when performing curve fitting. Afterwards, the prediction function curve is used as a constraint to predict the possible trend of the user's current state at the curve level of the original body mechanism response fluctuation, and obtain the prediction result. If the prediction result shows that the user's pathological state is likely to deteriorate or exceed the preset warning threshold, the system will immediately activate the continuous alarm mechanism. This alarm mechanism sends emergency signals to users in a variety of ways (such as sound, vibration, pop-up windows, etc.), reminding users to pay attention and take corresponding measures. At the same time, the system will automatically match nearby medical institutions based on the user's geographic location information, and recommend the most suitable medical institutions for nursing assistance to users based on the current pathological state development prediction results. For example, if the user is diagnosed with subsequent shock or the body enters the emergency stage, and the system predicts that his body may deteriorate rapidly in a short period of time, the system will immediately activate the alarm and recommend nearby hospitals with strong emergency departments to the user. At the same time, the system will automatically generate navigation path information to help users quickly reach the hospital, and send nursing assistance requests to the hospital in advance if necessary, so that the hospital can prepare for reception. In this way, users can get timely and effective medical assistance in a short time, thereby greatly improving the rescue effect and survival rate.

[0017] In a preferred embodiment, Figure 2 As shown, user voice data is obtained, voice recognition is performed in combination with a private semantic library, and a guidance information flow is generated, including: voice data is collected through a user terminal device, and the collected user voice data is pre-processed to obtain voice data to be converted; the voice data to be converted is input into a semantic recognition model, and in combination with a pre-built private semantic library, semantic recognition is performed on the voice data to be converted and converted into text data to obtain semantic request information; symptom list matching is performed based on the semantic request information, and a guidance information flow is generated according to the matching results, wherein the guidance information flow includes a preliminary judgment of the disease, the cause involved, and recommended examination items.

[0018] Optionally, when users need health care services, they can interact with the user through their own terminal devices (such as smartphones, tablets, or smart speakers). The system first starts the voice data collection function and accurately captures the symptoms described by the user through the microphone of the device. In order to ensure the accuracy and clarity of the voice data, the system pre-processes the collected raw voice data, including noise reduction, noise removal, and voice signal enhancement, so as to obtain the voice data to be converted. After that, the voice data to be converted is input into the semantic recognition model. This model is deeply learned and trained to efficiently convert voice into text. At the same time, in order to improve the accuracy and adaptability of recognition, the system also combines a pre-built private semantic library. This private semantic library is pre-trained in the user's local language. For example, after the initial communication through guided dialogue, the local dialect is easier to be recognized. The training contains a large number of medical and health-related vocabulary and local expressions, especially professional terms for common diseases, causes, and examination items. By combining the private semantic library, the semantic recognition model can more accurately understand the user's voice description and convert it into detailed text data, that is, semantic request information. With the semantic request information, the system will match a symptom list corresponding to its own record information based on the user's care records and information links. This symptom list contains typical symptoms, possible causes, and recommended examination items for various diseases. By matching the user's semantic request information with the symptom list, the system can quickly generate a guide information flow. This guide information flow not only includes the preliminary judgment of the disease, but also involves the possible causes and the examination items recommended for diagnosis. For example, if the user describes in the postoperative care stage through voice: "I always feel headache and a little nauseous recently." After voice data collection, preprocessing, semantic recognition and symptom list matching, the system will generate a guide information flow like this: "Preliminary judgment: postoperative concurrent headache with nausea; possible causes: migraine, increased intracranial pressure, etc.; recommended examination items: head CT, electroencephalogram, etc." In this way, users can have a preliminary understanding of their health status based on the guide information flow, and further interact or decide whether to seek further medical examination based on this.

[0019] In a preferred embodiment, the guidance information flow is uploaded to a hybrid cloud architecture network for pathology screening to obtain the current pathology status, wherein the hybrid cloud architecture network includes a public cloud architecture and a private cloud architecture, including: encrypting the guidance information flow, dynamically classifying it according to a preset diversion strategy, and uploading it to the hybrid cloud architecture network to complete pathology identification and screening to obtain an identification result; and using the identification result as the current pathology status.

[0020] Furthermore, after the user generates a guidance information flow containing a preliminary diagnosis of the disease, the cause of the disease, and the recommended examination items through voice interaction, the system will immediately upload this guidance information flow to an efficient and secure hybrid cloud architecture network for pathological screening. In order to ensure the security and privacy of user data, the system will use a hash function to encrypt the guidance information flow before uploading. This encryption technology can ensure the confidentiality and integrity of the data during transmission and prevent it from being accessed or tampered by unauthorized personnel. Subsequently, the guidance information flow is dynamically classified according to the preset diversion strategy. This diversion strategy takes into account multiple factors, including the data volume, urgency, and load of the current hybrid cloud architecture network of the guidance information flow. Through a comprehensive evaluation of these factors, it can be intelligently determined whether to upload the guidance information flow to the public cloud architecture or the private cloud architecture for processing. For example, if the guidance information flow contains an urgent and serious description of the disease, the system may give priority to uploading it to the public cloud architecture. This is because the public cloud architecture usually has more powerful computing power and a richer database of medical resources, which can complete pathological identification and screening more quickly and give accurate identification results. On the contrary, if the guidance information flow describes a relatively mild or common disease, and the current public cloud architecture is under a high load, the system may choose to upload it to the private cloud architecture for processing to reduce the burden on the public cloud architecture and improve processing efficiency. Once the guidance information flow is uploaded to the corresponding cloud architecture, the system will immediately start the pathology identification and screening program. This program will use advanced algorithms and models to conduct an in-depth analysis of the guidance information flow, and compare and match it with the information in the medical resource database. In the end, the system will give an identification result containing the current pathological status. This result can not only help users understand their health status more accurately, but also provide strong support for subsequent medical services.

[0021] In a preferred embodiment, a diversion strategy is preset for dynamic classification, including: dynamically selecting to upload the guidance information flow to a public cloud architecture or a private cloud architecture based on the data volume, urgency and current network load of the guidance information flow; when the data volume, urgency and current network load are all lower than their respective thresholds, uploading the guidance information flow to the private cloud architecture for pathology identification and screening; when any one of the data volume, urgency and current network load is higher than its threshold, uploading the guidance information flow to the public cloud architecture for pathology identification and screening.

[0022] Specifically, the preset diversion strategy will comprehensively consider the data volume, urgency and current network load of the guidance information flow when dynamically classifying. Specifically, the system will evaluate these three factors in real time and dynamically determine the upload destination of the guidance information flow based on the comparison results between them and their respective preset thresholds. If the data volume of the guidance information flow is not large, the urgency is not high, and the current network load is also at a low level, that is, these three factors are all below their respective thresholds, then the system will choose to upload the guidance information flow to the private cloud architecture for pathology identification and screening. This is because the private cloud architecture can usually provide faster and more personalized services when dealing with small-scale, non-urgent and lightly loaded tasks, while ensuring data security and privacy. However, if the data volume of the guidance information flow is large, the urgency is high, or the current network load is heavy, that is, any of these three factors is higher than its threshold, then the system will choose to upload the guidance information flow to the public cloud architecture for pathology identification and screening. The public cloud architecture has powerful computing power and abundant resources, and can handle large-scale, high-urgency tasks and maintain stable performance when the network load is heavy, thereby ensuring the accuracy and efficiency of pathology identification and screening. For example, during a busy medical consultation period, if a large number of users upload the diagnosis information flow at the same time, causing a sharp increase in network load, the system will automatically upload part or all of the diagnosis information flow to the public cloud architecture for processing according to the preset diversion strategy to avoid overloading the private cloud architecture and affecting the service quality. Similarly, if a user uploads a diagnosis information flow that contains a large amount of data or describes an urgent condition, the system will also give priority to the public cloud architecture for rapid processing.

[0023] In a preferred embodiment, when uploading to a private cloud architecture for pathology identification and screening, it includes: combining the private cloud architecture, localizing the guidance information flow to obtain an identification result; wherein the private cloud architecture stores local user data, and verifying and analyzing the guidance information flow based on the local user data to obtain a pathology feature vector; inputting the pathology feature vector into a pre-trained local pathology identification model, wherein the local pathology identification model is trained based on local user pathology case data, and outputting a first pathology identification result; performing a quality assessment on the first pathology identification result, and if the assessment result meets the preset standard, the identification result is used as the final identification result; if the assessment result does not meet the preset standard, the guidance information flow is re-uploaded to the public cloud architecture for secondary identification and screening.

[0024] Preferably, when the guidance information flow is uploaded to the private cloud architecture for pathological identification and screening, the system will make full use of the local user data stored on the private cloud architecture to localize the guidance information flow. First, the system will verify and analyze the guidance information flow based on these local user data, and extract key pathological feature vectors, which can accurately reflect the characteristics of the disease, the cause of the disease, and possible examination items described in the guidance information flow. Next, the system will input the extracted pathological feature vectors into the pre-trained local pathological recognition model. This local pathological recognition model is trained based on a large amount of local user pathological case data, so it can accurately identify the specific pathological characteristics of local users. Specifically, the local pathological recognition model is built based on a deep learning neural network structure, which includes an input layer, a hidden layer, and an output layer. The input layer receives the pathological feature vector extracted from the processed guidance information flow; the hidden layer is composed of multiple neurons, and the input data is deeply extracted and nonlinearly transformed through complex weight connections; the output layer outputs the first pathological recognition result based on the processing result of the hidden layer, which contains the system's preliminary judgment of the disease, possible causes, and recommended examination items. In the specific training process, a large number of pathological cases of local users are collected, covering various types of diseases, symptoms, examination results, and diagnosis and treatment information. These data are cleaned and preprocessed to remove erroneous or incomplete data, and divided into training set, validation set, and test set according to a certain ratio. In the training stage, the pathological feature vectors in the training set are input into the model, and the model continuously adjusts the weights between the neurons in the hidden layer so that the output results of the model gradually approach the real diagnosis results. In the training process, the loss function is used to measure the difference between the model output and the real result, and the loss function is minimized with the help of the optimization algorithm, so as to continuously optimize the performance of the model. In the training process, the validation set is used to evaluate the model regularly, monitor the accuracy, recall rate and other indicators of the model, and prevent the model from overfitting. When the performance of the model on the validation set no longer improves, it is considered that the model training has reached a good state. Finally, the test set is used to comprehensively evaluate the trained model to ensure that the model has good generalization ability on unseen data and can accurately identify various pathological conditions.

[0025] However, in order to further ensure the accuracy and reliability of the recognition results, the system will also conduct a quality assessment on the first pathology recognition result. The assessment process will take into account multiple factors, such as the confidence of the recognition result, the degree of match with the local user data, etc. If the assessment result meets the preset standards, the system will use the recognition result as the final recognition result and provide the user with corresponding health care services based on it. However, if the assessment result does not meet the preset standards, it means that there may be errors or uncertainties in the first pathology recognition result. In this case, the system will re-upload the guidance information flow to the public cloud architecture for secondary recognition and screening. The public cloud architecture can conduct a deeper and more comprehensive analysis of the guidance information flow, thereby providing more accurate recognition results. For example, suppose that a user uploads a guidance information flow that describes a rare disease, and the local pathology recognition model on the private cloud architecture has limited recognition capabilities for this disease. During the initial recognition, the model may give a less accurate recognition result. However, through quality assessment, the system finds that the confidence of this result is low and the degree of match with the local user data is not high. Therefore, the system will re-upload the guidance information flow to the public cloud architecture for secondary recognition and screening. The more advanced pathology recognition model on the public cloud architecture can accurately identify this rare disease and give corresponding treatment recommendations, thereby providing users with more accurate health care services.

[0026] In a preferred embodiment, when uploading to a public cloud architecture for pathology identification screening, it includes: uploading the guidance information flow to the public cloud architecture, standardizing the guidance information flow to make it meet the input requirements of the universal pathology identification model on the public cloud architecture; inputting the standardized guidance information flow into the universal pathology identification model on the public cloud architecture, the universal pathology identification model is trained based on large-scale cross-regional and cross-population pathology data, and outputting a second pathology identification result; performing a security review on the second pathology identification result returned by the public cloud architecture, and using the identification result as the current pathology state after the review is passed.

[0027] In detail, when the guidance information flow is uploaded to the public cloud architecture for pathology identification and screening, it will first undergo a standardization process. Since the universal pathology identification model on the public cloud architecture needs to process pathology data from different regions and different populations, the guidance information flow must meet the input requirements of the model. The system will perform format conversion and data cleaning on the guidance information flow to ensure that it can be accurately identified and processed by the universal pathology identification model. Next, the standardized guidance information flow will be input into the universal pathology identification model on the public cloud architecture. This model is trained based on large-scale cross-regional and cross-population pathology data, so it has extremely high accuracy and generalization ability. It can quickly analyze the symptom description, cause of disease, and recommended examination items in the guidance information flow, and output a second pathology identification result. This result usually contains information such as detailed classification of the disease, possible cause analysis, and recommended treatment plans for the disease. However, before using the second pathology identification result as the current pathological state, the system will also conduct a security review on it. This is because although the public cloud architecture has powerful computing power and abundant resources, it also has security risks such as data leakage and tampering. The system will check the integrity of the recognition results, the reliability of the source, and whether there is abnormal data to ensure its authenticity and accuracy. Only after a security review and confirmation will the system use the recognition result as the current pathological state and provide further health care services to the user based on it. For example, suppose a user uploads a guidance information stream that describes a common cold, but the symptoms are more special. After receiving the standardized guidance information stream, the general pathology recognition model on the public cloud architecture will quickly analyze the symptom description and cause, and output a second pathology recognition result, indicating that the user may have a special type of cold, and give corresponding treatment suggestions. Before returning this result to the user, the system will conduct a security review on it. After confirming that the recognition result has not been tampered with or has abnormal data, it will use it as the current pathological state and inform the user of the specific symptom and treatment plan.

[0028] In a preferred embodiment, based on the current pathological state, the development of the pathological state is predicted at the historical load response level, and a continuous alarm is issued if the prediction result exceeds the warning threshold, including: collecting historical pathological data related to the current pathological state, including pathological index changes and treatment effect data of the same disease in different time periods; using the historical pathological data of the same disease in different time periods as the horizontal axis and the pathological state development trend as the vertical axis, curve fitting the pathological state development trend to obtain a prediction function graph; using the prediction function graph as a constraint to predict the result of the current pathological state, and obtain the prediction result of the pathological state development within a preset time period; setting a warning threshold, when the prediction result shows that the pathological state development exceeds the warning threshold, starting the continuous alarm mechanism, and sending an alarm signal to the user through the user terminal device in the form of sound, vibration, or pop-up window.

[0029] Exemplarily, based on the currently determined pathological state, the system will further predict the development of the pathological state to evaluate the possible direction of the disease and take timely countermeasures. First, the system will collect historical pathological data of users closely related to the current pathological state. These data cover the changes in pathological indicators of users with the same disease in different time periods and the corresponding treatment effects. For example, for the disease of diabetes, the system will collect data such as the user's past blood sugar level fluctuations, changes in glycosylated hemoglobin values, and blood sugar control effects under different treatment plans. Next, using these historical pathological data as the horizontal coordinates and the development trend of the pathological state as the vertical coordinates, mathematical algorithms are used to process these data, so as to curve fit the development trend of the pathological state and generate a prediction function curve graph that can reflect the law of disease development. This curve graph is like a prediction model. With it as a constraint condition, the current pathological state data is substituted into it, and then the prediction results of the development of the pathological state within a preset time period are obtained. These results refer to the possible trend of the current state under the curve of the original body mechanism response fluctuation change. For example, for patients with hypertension, the system collects the patient's blood pressure data for the past few months or even years, including systolic and diastolic blood pressure values ​​at different time points, as well as blood pressure changes under the use of different antihypertensive drugs or different lifestyle interventions, all of which belong to historical pathological data. Taking the blood pressure measurement values ​​and corresponding treatment conditions in chronological order as the horizontal axis, and the blood pressure control situation, blood pressure fluctuation trend and other pathological state development trends as the vertical axis, a curve reflecting the patient's blood pressure change law is drawn through a suitable curve fitting algorithm, that is, the prediction function curve graph. If the current patient's blood pressure is at a certain level, substituting this current state into the prediction function curve graph can predict how the user's blood pressure will change in the future, that is, its trend. In order to promptly discover and respond to possible changes in the condition, the system will set an early warning threshold, which is equivalent to virtualizing two early warning curves above and below the load fitting curve corresponding to the user's pathological state based on the early warning threshold. When the prediction result shows that the development of the change curve of the pathological state may exceed this early warning threshold, that is, when the curve intersection is generated, the system will immediately start the continuous alarm mechanism. This alarm mechanism will send alarm signals to users through user terminal devices in various ways such as sound, vibration, and pop-up windows to ensure that users can receive notifications in a timely manner and take appropriate actions. For example, if the prediction results show that the blood sugar level of a diabetic patient may rise sharply in the next few days and exceed the safe range, the system will send a loud alarm through the user's mobile phone and pop up an emergency notification window to remind the patient to seek medical treatment immediately or adjust the treatment plan to deal with possible situations.

[0030] In a preferred embodiment, nursing assistance is performed based on alarm signal linkage positioning, including: when the continuous alarm mechanism is activated, automatically obtaining the geographic location information of the user terminal, matching the geographic location information with a preset medical institution database, wherein the medical institution database stores the medical institution information around the geographic location information of the user terminal within a preset range; based on the matching result, combined with the current pathological state development prediction result, recommending a first medical institution, and generating navigation path information; pushing the navigation path information and the alarm signal to the user through the user terminal device, and at the same time, sending a nursing assistance request to the first medical institution; after receiving the nursing assistance request, the first medical institution decides whether to accept it according to its own real-time load situation, and generates a feedback processing result; if the first medical institution accepts the request, it is shared with the navigation path information; if the first medical institution cannot accept the request, a medical institution is recommended to the user again until a response is processed.

[0031] Specifically, when the continuous alarm mechanism is activated, the system will quickly and automatically obtain the geographical location information of the user terminal, which is crucial for subsequent nursing assistance. The system will accurately match this geographical location information with the preset medical institution database. This medical institution database stores in detail the information of various medical institutions around the geographical location information of the user terminal within the preset range, including hospitals, clinics, emergency centers, etc., as well as their service scope, professional characteristics, real-time load conditions and other key data. According to the matching results, the system will combine the current pathological state development prediction results to intelligently recommend the most suitable first medical institution. For example, if the user is predicted to have acute myocardial infarction and the disease progresses rapidly, the system will give priority to recommending nearby tertiary hospitals with cardiac interventional surgery capabilities. At the same time, the system will also generate detailed navigation path information to ensure that the user can quickly and accurately reach the recommended medical institution. Next, the system will push the navigation path information and alarm signals to the user through the user terminal device to ensure that the user can understand his or her condition and the medical institution to go to in the first time. At the same time, the system will also send a nursing assistance request to the first medical institution, which contains key information such as the user's pathological state, prediction results, and geographical location, so that the medical institution can prepare for the reception in advance. After receiving the nursing assistance request, the first medical institution will quickly decide whether to accept it based on its real-time load. If the request is accepted, the medical institution will share it with the navigation path information to ensure that the user can arrive quickly along the optimal route. If the first medical institution is unable to accept the request due to overload or other reasons, the system will re-recommend other suitable medical institutions for the user until a medical institution that can respond and handle the request is found.

[0032] An Internet-based health care service system provided by an embodiment of the present invention has at least the following technical effects: 1. By combining private semantic libraries for speech recognition, it is able to accurately understand the description of symptoms in the user's voice and generate a guidance information flow that includes a preliminary diagnosis of the symptoms, the cause of the disease, and recommended examination items. Furthermore, the hybrid cloud architecture network is used for pathology screening, and the processing architecture is dynamically selected according to the data volume, urgency, and network load of the guidance information flow, which not only ensures the efficiency of data processing, but also ensures the accuracy of recognition, effectively improving the accuracy and response speed of pathology recognition, and providing users with more timely and reliable health consultation services.

[0033] 2. Combine the current pathological state with the historical load response level to make a prediction and set the warning threshold. Once the prediction results show that the pathological state may exceed the warning threshold, the system will immediately start the continuous alarm mechanism and send alarm signals to users in various ways through user terminal devices. This intelligent prediction and alarm mechanism helps users to understand the development trend of their own disease in a timely manner, take necessary preventive measures or seek medical treatment, and effectively avoid the deterioration of their disease.

[0034] 3. When the continuous alarm mechanism is activated, it can automatically obtain the geographical location information of the user terminal, match it with the preset medical institution database, quickly recommend suitable medical institutions and generate navigation path information. At the same time, the system will also send a nursing assistance request to the medical institution to achieve rapid linkage between the user and the medical institution. After receiving the request, the medical institution will decide whether to accept it based on its own real-time load situation and generate feedback processing results. This nursing assistance linkage mechanism greatly shortens the time from the user's discovery of the disease to obtaining professional medical assistance, improves the efficiency and success rate of nursing assistance, and provides users with more convenient and efficient health care services.

[0035] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A health care service system based on the Internet, characterized in that: The implementation steps include: Obtain user voice data, combine it with the private semantic library for voice recognition, and generate a guidance information flow; Uploading the guidance information flow to a hybrid cloud architecture network for pathology screening to obtain a current pathology status, wherein the hybrid cloud architecture network includes a public cloud architecture and a private cloud architecture; Based on the current pathological state, the development of the pathological state is predicted at the historical load response level. If the predicted result exceeds the warning threshold, a continuous alarm is issued, and nursing assistance is provided based on the alarm signal linkage positioning.

2. The Internet-based health care service system according to claim 1, characterized in that: Obtain user voice data, perform voice recognition in combination with the private semantic library, and generate a guidance information flow. The execution steps include: Collecting voice data through user terminal equipment, preprocessing the collected user voice data, and obtaining voice data to be converted; Inputting the speech data to be converted into a semantic recognition model, combining with a pre-built private semantic library, performing semantic recognition on the speech data to be converted and converting it into text data, and obtaining semantic request information; A symptom list match is performed based on the semantic request information, and a guidance information flow is generated according to the matching result, wherein the guidance information flow includes a preliminary diagnosis of the symptom, the causes involved, and recommended examination items.

3. The Internet-based health care service system according to claim 1, characterized in that: The guidance information flow is uploaded to the hybrid cloud architecture network for pathological screening to obtain the current pathological status, wherein the hybrid cloud architecture network includes a public cloud architecture and a private cloud architecture, and the execution steps include: The guidance information flow is encrypted, dynamically classified according to a preset diversion strategy, and then uploaded to the hybrid cloud architecture network to complete pathology identification and screening to obtain identification results; The recognition result is used as the current pathological state.

4. The Internet-based health care service system according to claim 3, characterized in that: The preset diversion strategy is used for dynamic classification. The execution steps include: Dynamically select to upload the medical guidance information flow to the public cloud architecture or the private cloud architecture according to the data volume, urgency and current network load of the medical guidance information flow; When the data volume, urgency and current network load are all below their respective thresholds, the guidance information flow is uploaded to the private cloud architecture for pathology identification and screening; When any one of the data volume, urgency and current network load is higher than a threshold, the guidance information flow is uploaded to the public cloud architecture for pathology identification and screening.

5. The Internet-based health care service system according to claim 4, characterized in that: When uploaded to the private cloud architecture for pathology identification screening, the execution steps include: Combined with the private cloud architecture, the medical guidance information flow is processed locally to obtain the recognition result; Wherein, local user data is stored on the private cloud architecture, and verification and analysis of the diagnosis guidance information flow is performed based on the local user data to obtain a pathological feature vector; Inputting the pathological feature vector into a pre-trained local pathology recognition model, wherein the local pathology recognition model is trained based on local user pathology case data, and outputting a first pathology recognition result; Performing a quality assessment on the first pathology recognition result, and if the assessment result meets a preset standard, taking the recognition result as the final recognition result; If the evaluation result does not meet the preset standard, the medical guidance information flow is re-uploaded to the public cloud architecture for secondary identification and screening.

6. The Internet-based health care service system according to claim 4, characterized in that: When uploaded to the public cloud architecture for pathology identification screening, the execution steps include: The medical guidance information flow is uploaded to the public cloud architecture, and the medical guidance information flow is standardized so as to meet the input requirements of the general pathology recognition model on the public cloud architecture; The standardized diagnosis guidance information flow is input into a universal pathology recognition model on a public cloud architecture, where the universal pathology recognition model is trained based on large-scale cross-regional and cross-population pathology data, and outputs a second pathology recognition result; A security review is performed on the second pathology recognition result returned by the public cloud architecture, and after passing the review, the recognition result is used as the current pathology status.

7. The Internet-based health care service system according to claim 1, characterized in that: Based on the current pathological state, the pathological state development prediction is performed at the historical load response level. If the prediction result exceeds the warning threshold, a continuous alarm is issued. The execution steps include: Collect historical pathological data related to the current pathological status, including changes in pathological indicators and treatment effect data of the same disease in different time periods; Taking the historical pathological data of the same disease in different time periods as the horizontal coordinate and the pathological state development trend as the vertical coordinate, curve fitting is performed on the pathological state development trend to obtain a prediction function curve graph; Using the prediction function curve as a constraint, the current pathological state is predicted to obtain a prediction result of the pathological state development within a preset time period; Set a warning threshold. When the prediction results show that the pathological condition has developed beyond the warning threshold, activate the continuous alarm mechanism and send an alarm signal to the user through the user terminal device in the form of sound, vibration, and pop-up window.

8. The Internet-based health care service system according to claim 1, characterized in that: Nursing assistance is carried out based on alarm signal linkage positioning. The execution steps include: When the continuous alarm mechanism is activated, the geographical location information of the user terminal is automatically obtained, and the geographical location information is matched with a preset medical institution database, wherein the medical institution database stores the medical institution information around the geographical location information of the user terminal within a preset range; According to the matching results, combined with the current pathological state development prediction results, the first medical institution is recommended and navigation path information is generated; Pushing the navigation path information and the alarm signal to the user through the user terminal device, and at the same time, sending a nursing assistance request to the first medical institution; After receiving the nursing assistance request, the first medical institution decides whether to accept it according to its own real-time load situation and generates a feedback processing result; If the first medical institution accepts the request, it will be shared with the navigation path information; if the first medical institution cannot accept the request, a new medical institution will be recommended to the user until a response is processed.

Images