Intelligent medical platform based on big data

Abstract

The invention is suitable for the field of smart medical treatment, and provides a smart medical platform based on big data, and the platform comprises a data collection module, a data analysis module, a referral aid decision-making module, a comprehensive data sharing module, a health management monitoring module, an autonomous participation management module, and a library learning updating module. Data analysis and artificial intelligence technologies are utilized to optimize the referral process of the patient, it is ensured that the patient can be treated in the most suitable medical institution, and efficient and accurate referral decision support is achieved by combining the individual disease development characteristics of the patient, the current treatment effect and real-time data of medical resources in the area. The efficiency and the accuracy of the medical referral process are remarkably improved, meanwhile, the medical experience and the treatment result of the patient are improved, the problems that medical resources are distributed unevenly, the referral process is complex and the like are effectively solved, and the method has important significance on continuous optimization and development of a modern medical system.

Description

Technical Field

[0001] The present invention relates to the field of smart medical care, and in particular to a smart medical care platform based on big data. Background Art

[0002] Smart healthcare, also often referred to as digital healthcare or intelligent healthcare, refers to a series of activities that utilize modern information technology, such as digital technologies, big data analytics, artificial intelligence, and the internet, to optimize healthcare quality, improve efficiency, reduce costs, and enhance the healthcare experience. Key components of this field include electronic health records (EHRs), telemedicine services, patient health data analysis, personalized medical solutions, intelligent diagnostic systems, disease prediction models, medical robots, and wearable medical devices. The goal of smart healthcare is to achieve more personalized, precise, efficient, and convenient medical services through technological innovation.

[0003] Existing technologies suffer from inaccurate identification of personalized medical needs, difficulty integrating and utilizing medical resource information in real time, inefficient traditional referral decision-making, a lack of immediate updates and dynamic adjustments to patient conditions, uneven distribution of medical resources, and complex referral processes. These issues collectively impact the efficiency and quality of medical services, potentially preventing patients from receiving the most appropriate treatment in a timely manner and increasing the burden on the medical system. Summary of the Invention

[0004] The purpose of the present invention is to provide a smart medical platform based on big data, aiming to solve the technical problems existing in the existing technology identified in the background technology.

[0005] The present invention is implemented as follows: a smart medical platform based on big data, the platform comprising: The data acquisition module is used to connect with the information system of medical institutions to automatically obtain and manage relevant patient data, including medical records, physiological parameters, genomic data and medical records; The data analysis module is used to analyze the automatically acquired data, identify individual differences and characteristics in the patient group, and combine the relevant data of each patient to predict the patient's disease development trend and evaluate the current treatment effect; The referral decision-making support module is used to obtain the availability of medical resources at the current medical institution and medical institutions within the region. It then determines whether the availability of medical resources at the current medical institution can meet the patient's needs based on the patient's disease development trend forecast and current treatment effect. If not, it identifies all medical institutions within the region that can meet the patient's needs and adds a transfer mark for the patient. A comprehensive data sharing module is used to establish a regional data sharing platform, which establishes communication channels with all medical institutions in the region and stores patient-related data with transfer marks in the data sharing platform in an encrypted manner; The health management monitoring module is used to monitor the patient's physiological parameters and disease development trend data in real time, set physiological expected thresholds based on the relevant data of each patient, and issue warning messages when the monitored parameters and data exceed the physiological expected thresholds; The autonomous participation management module is used to receive the patient's application information and physiological parameters uploaded by the patient, and to feed back the doctor's feedback instructions to the patient; The library learning and updating module is used to establish a medical reserve library, add labels to each disease in the medical reserve library, and store the relevant data and final treatment plan of each patient in the medical reserve library.

[0006] As a further solution of the present invention, the data acquisition module includes: The institution docking unit is used to connect with the information system of each medical institution, establish a communication connection, and automatically obtain and transmit relevant patient data; The data management unit is used to store, integrate and manage the collected patient-related data and add a patient name tag to each patient's data; The data encryption unit encrypts the data of each patient.

[0007] As a further solution of the present invention, the data analysis module includes: A feature selection unit is used to extract individual differences and features from each patient's relevant data, including patient body shape data, genetic characteristics, clinical data, and current disease data; Individualized risk assessment unit, which is used to assess the patient's current disease transformation risk and predict the potential path and speed of disease progression based on the patient's clinical data and current disease data; The effectiveness evaluation unit is used to analyze all data of patients who have received treatment and evaluate the effectiveness of the treatment plan for the patient.

[0008] As a further solution of the present invention, the referral decision support module includes: The medical resource analysis unit is used to obtain the medical resource information of all medical institutions within the current medical institution and the patient's area, including the departments, medical equipment, vacant beds and medical level of each medical institution; The patient needs assessment unit is used to evaluate the medical level of the current medical institution based on the patient's disease development trend and current treatment effect, analyze the patient's treatment needs and the availability of medical resources in the medical institution, and determine whether the medical institution has the medical resources required to cure the corresponding patient; The regional institution screening unit is used to screen all medical institutions within the patient's area that can meet the patient's treatment needs based on the patient's treatment needs analysis results when the medical institution does not have the medical resources required to cure the patient; The referral marking unit is used to obtain the medical institution selected by the patient, establish communication with the medical institution, synchronize all the patient's data to the information database of the medical institution, and add a transfer mark for the patient.

[0009] As a further solution of the present invention, the comprehensive data sharing module includes: A communication building unit is used to establish a data sharing platform and, based on the coverage of the data sharing platform, to establish communication channels with all medical institutions within the coverage area; A data sharing permission unit is used to set the access permission of the data sharing platform, wherein the access permission is set to be accessible only to doctors; The data upload synchronization unit is used to upload the patient information that needs to be referred to the data sharing platform and encrypt the uploaded data. The encryption key will only be synchronized to the medical institution selected by the patient.

[0010] As a further solution of the present invention, the health management monitoring module includes: A real-time monitoring unit is used to connect to the sensor equipment worn by the patient to collect the patient's physiological parameter data and collect the patient's disease development trend data; A standard setting unit is used to set personalized physiological expected thresholds based on each patient's physiological parameter data and current disease status; The anomaly detection unit is used to compare and analyze the monitoring data in real time, and compare each monitoring data obtained with the set physiological expectation threshold. When any monitoring data exceeds the physiological expectation threshold, a warning message is issued.

[0011] As a further solution of the present invention, the autonomous participation management module includes: The application information receiving unit is used to interact with the patient, receive the patient's needs, consultation and appointment information, and receive the doctor's response to the interactive information and feed it back to the patient; The parameter autonomous uploading unit is used to receive physiological parameter data actively uploaded by patients.

[0012] As a further solution of the present invention, the library learning and updating module includes: The library establishment unit is used to build a medical storage library to store disease-related data and add a label index for each disease in the medical storage library; The patient data storage unit is used to store the patient's information in a tag under the same index based on the patient's disease data, and obtain the treatment plan and treatment effect adopted by the patient this time, and bind and store them in the tag.

[0013] The beneficial effects of the present invention are: Utilize data analysis and artificial intelligence technologies to optimize the patient referral process, ensuring that patients can receive treatment at the most suitable medical institution. By combining the patient's individual disease development characteristics, current treatment effects, and real-time data on regional medical resources, efficient and accurate referral decision support can be achieved.

[0014] By deeply analyzing each patient's individual data and regional medical resources, we provide personalized medical decision support. This personalized analysis ensures that patients are referred to the medical institution that best meets their specific needs, significantly improving medical efficiency and patient satisfaction. Furthermore, with real-time access to resource information across various medical institutions, referral decisions can be made based on the latest medical resource status. This real-time nature ensures the accuracy and timeliness of referral decisions, effectively alleviating the imbalance in the distribution of medical resources.

[0015] At the same time, it can also dynamically evaluate the patient's treatment needs and the service capabilities of current medical institutions. This dynamic evaluation mechanism allows the system to adjust referral decisions in real time when the patient's condition changes, ensuring that the patient is always on the best treatment path.

[0016] This invention significantly improves the efficiency and accuracy of the medical referral process, while improving patients' medical experience and treatment outcomes, effectively solving problems such as uneven distribution of medical resources and complex referral processes, and is of great significance to the continuous optimization and development of the modern medical system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A block diagram of the big data-based smart medical platform provided by an embodiment of the present invention; Figure 2 A structural block diagram of a data acquisition module provided in an embodiment of the present invention; Figure 3 A structural block diagram of a data analysis module provided in an embodiment of the present invention; Figure 4 A structural block diagram of a referral decision support module provided in an embodiment of the present invention; Figure 5A structural block diagram of the comprehensive data sharing module provided in an embodiment of the present invention; Figure 6 A structural block diagram of the health management and monitoring module provided in an embodiment of the present invention; Figure 7 A structural block diagram of the autonomous participation management module provided in an embodiment of the present invention; Figure 8 This is a structural block diagram of the library learning and updating module provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0019] It is understood that the terms "first," "second," etc., used herein may be used to describe various elements, but unless otherwise specified, these elements are not limited by these terms. These terms are only used to distinguish a first element from another element. For example, a first xx script may be referred to as a second xx script, and similarly, a second xx script may be referred to as a first xx script without departing from the scope of this application.

[0020] Figure 1 The block diagram of the smart medical platform based on big data provided by the embodiment of the present invention is as follows: Figure 1 As shown, a smart medical platform based on big data includes: The data acquisition module 100 is used to connect with the information system of the medical institution to automatically obtain and manage the patient's relevant data, including medical records, physiological parameters, genomic data and medical records; In big data-based smart healthcare platforms, the data acquisition module plays a crucial role, enabling efficient, accurate, and secure collection of patient data. By employing advanced interface technologies and communication protocols, it automatically identifies and adapts to the technical requirements of different medical institution information systems, enabling seamless integration with multiple medical information systems. This significantly improves data collection efficiency and coverage, while reducing error rates.

[0021] Utilizing the latest data management technologies, we not only store and integrate the collected data, but also clean, standardize and classify the data to establish a comprehensive and structured data archive for each patient, ensuring the high quality and easy analysis of the data, and providing an accurate and reliable data foundation for subsequent data analysis and decision support.

[0022] At the same time, the latest encryption technology provides strong protection for each patient's data, ensuring the security of data during transmission and storage, providing a solid guarantee for patient privacy and enhancing the trust of patients and medical institutions in the platform. In addition, the design of the data acquisition module takes into account changes in future technologies and medical needs, with good scalability and flexibility, enabling the platform to adapt to the rapid development needs of the medical industry.

[0023] This series of design advancements and technological advantages not only provides a solid data foundation for the smart medical platform, but also makes significant contributions to improving the efficiency, accuracy and safety of the entire medical system.

[0024] The data analysis module 200 is used to analyze the automatically acquired data, identify individual differences and characteristics in the patient group, and combine the relevant data of each patient to predict the disease development trend and evaluate the current treatment effect of the patient; The main task is to deeply explore and analyze medical data to provide personalized disease prediction and treatment effect evaluation.

[0025] Leveraging advanced data mining techniques, the unit extracts individual differences and characteristics from massive amounts of patient data that are crucial for disease prediction and treatment. These characteristics include, but are not limited to, the patient's physical measurements, genetic profile, clinical data, and current disease status. By identifying these key characteristics, the unit provides precise foundational data for subsequent risk assessment and treatment efficacy evaluation, significantly improving prediction accuracy and the potential for personalized treatment.

[0026] This characteristic data is then further analyzed, combined with the patient's clinical data and current disease status, using complex algorithmic models to assess the patient's current risk of disease progression and predict the path and speed of disease progression. This unit not only helps doctors and patients understand the potential development of the disease but also provides a scientific basis for developing more precise treatment plans.

[0027] This module also tracks and analyzes patients who have already received treatment, analyzing all pre- and post-treatment data to assess the effectiveness of the current treatment plan. This process is key to continuously improving treatment methods and increasing treatment efficiency. By evaluating treatment effectiveness in real time, the medical team can adjust treatment plans in a timely manner to improve treatment success rates.

[0028] Through efficient feature selection, the data analysis module can accurately identify patient characteristics that have a significant impact on disease prediction and treatment, providing strong support for personalized medicine. Secondly, individualized risk assessment enables more accurate medical decision-making, providing patients with more personalized treatment recommendations, significantly improving treatment outcomes and patient satisfaction. Finally, the implementation of the effectiveness evaluation unit ensures the continuous optimization and updating of treatment plans, driving the continuous improvement of medical service quality. These advances and advantages not only improve the efficiency and effectiveness of medical services, but also provide patients with safer, more effective, and personalized medical services, demonstrating the enormous potential of smart medical platforms in the modern medical field.

[0029] The referral decision support module 300 is used to obtain the medical resource availability of the current medical institution and medical institutions within the region, and, based on the patient's disease development trend prediction and current treatment effect, determine whether the medical resource availability of the current medical institution can meet the patient's cure needs. If not, it identifies all medical institutions within the region that can meet the patient's cure needs and adds a transfer mark for the patient; The referral decision support module is a core component of the big data-based smart healthcare platform. It leverages data analytics and artificial intelligence to optimize the patient referral process, ensuring patients receive treatment at the most appropriate medical institution. This module integrates individual patient disease progression characteristics, current treatment outcomes, and real-time data on regional medical resources to support efficient and accurate referral decisions.

[0030] The referral decision support module provides personalized medical decision support by deeply analyzing each patient's individual data (including disease development trends, treatment effect feedback, etc.) and regional medical resources. This personalized analysis ensures that patients are referred to the medical institution that best meets their specific needs, significantly improving medical efficiency and patient satisfaction.

[0031] The module can obtain real-time resource information from each medical institution (such as department capacity, equipment utilization, and number of vacant beds), enabling referral decisions to be made based on the latest medical resource status. This real-time nature ensures the accuracy and timeliness of referral decisions, effectively alleviating the imbalance in the distribution of medical resources.

[0032] It can also dynamically evaluate the patient's treatment needs and the service capabilities of current medical institutions. This dynamic evaluation mechanism allows the system to instantly adjust referral decisions when the patient's condition changes, ensuring that the patient is always on the best treatment path.

[0033] The modules then work together to optimize the referral process, from patient needs assessment to final selection of a healthcare provider, making it highly automated and efficient. Automatic synchronization of patient data reduces information loss and delays during referrals, while the addition of transfer tags to patients ensures smooth communication and data sharing between healthcare providers.

[0034] Furthermore, throughout the referral process, encryption is implemented during the transmission and storage of patients' personal and medical data, safeguarding patient privacy and data security. This data protection measure provides patients with a safer medical experience.

[0035] The referral decision support module significantly improves the efficiency and accuracy of the medical referral process through highly integrated data analysis and intelligent algorithms, while also enhancing patients' medical experience and treatment outcomes. The implementation of this module effectively addresses issues such as uneven distribution of medical resources and complex referral processes, and is of great significance to the continuous optimization and development of modern medical systems.

[0036] Comprehensive data sharing module 400, for establishing a regional data sharing platform, wherein the data sharing platform establishes communication channels with all medical institutions within the region, and stores patient-related data with transfer marks in the data sharing platform in an encrypted manner; The comprehensive data sharing module plays the role of a data hub in the smart medical platform. By building a secure and efficient data sharing platform, it promotes the transparency and circulation of medical information, thereby improving the response speed and treatment quality of the entire medical service system.

[0037] The data sharing module is responsible for establishing a stable communication channel between various medical institutions within the region. This communication framework must not only enable real-time data transmission but also ensure the security and stability of the transmission process. A significant advancement of this design lies in its ability to integrate multiple communication technologies and protocols, ensuring seamless connectivity between different medical institutions and laying a solid foundation for subsequent data sharing and access.

[0038] Protect patient privacy and data security by setting strict access rights. In this module, access rights are strictly limited to doctors only. This measure can effectively prevent the leakage of sensitive patient information, protect patient privacy, and prevent data abuse or unauthorized access while meeting medical needs.

[0039] This module ensures the secure synchronization of patient information required for referrals across medical institutions. Not only is patient data uploaded to the shared platform encrypted, but the encryption key is also synchronized only with the patient's selected medical institution. This approach ensures both secure data transmission and the continuity and efficiency of healthcare services, as only authorized medical institutions can decrypt and access this information, ensuring timely medical care for patients.

[0040] Overall, the integrated data sharing module is designed with patient safety and privacy at the forefront, while ensuring the efficiency and accuracy of medical information flows. This module improves the reliability and security of medical services, reduces information transmission errors, accelerates response times, and significantly enhances the patient experience and treatment outcomes. In this way, the smart healthcare platform optimizes the allocation of medical resources, promotes collaboration among regional medical institutions, and ultimately improves the efficiency of the entire healthcare system.

[0041] The health management monitoring module 500 is used to monitor the patient's physiological parameters and disease development trend data in real time, set physiological expected thresholds based on the relevant data of each patient, and issue a warning message when the monitored parameters and data exceed the physiological expected thresholds; The main function of this module is to monitor the patient's health status in real time, promptly identify potential health problems and prevent the deterioration of the disease, and enhance the early warning and response capabilities of the entire medical system.

[0042] The module connects to various medical sensor devices worn by patients, such as heart rate monitors, blood glucose meters, and blood pressure monitors. These devices continuously collect physiological parameters such as heart rate, blood glucose levels, and blood pressure, and transmit the data in real time to the smart healthcare platform. A significant advancement of this unit lies in its high degree of integration and scalability, enabling compatibility with a wide range of sensors and devices, enabling simultaneous monitoring of multiple physiological parameters.

[0043] Based on a patient's individual health history and combined with disease trend analysis from the big data analytics module, personalized physiological prediction thresholds are customized for each patient. These thresholds are not static but dynamically adjusted based on changes in the patient's condition, ensuring the accuracy and timeliness of monitoring results. The advantage of this unit lies in its ability to achieve refined management and tailor health management strategies to individual differences.

[0044] Finally, this module continuously compares the collected data against customized thresholds. If any physiological parameter exceeds a threshold, the unit immediately analyzes the severity of the exceedance and issues appropriate alerts based on the configured protocol. A significant advancement in this unit lies in its intelligent anomaly detection algorithm, which can identify abnormal changes that truly require attention based on the patient's disease type and clinical presentation.

[0045] The autonomous participation management module 600 is used to receive the patient's application information and physiological parameters uploaded by the patient, and to feed back the doctor's feedback instructions to the patient; This module not only enhances patients' active engagement but also improves the accessibility and personalization of healthcare services. It allows patients to conveniently submit their needs, inquiries, and appointment requests. Natural language processing technology is used to understand and categorize patient requests so they can be efficiently directed to the appropriate healthcare provider. Furthermore, it is used to receive physician responses and convey this feedback to patients in an easily understandable format.

[0046] It also allows patients to upload their physiological parameter data via smartphones, wearable devices, or home medical devices. It not only supports multiple data formats and upload methods, but also performs preliminary analysis on the uploaded data to quickly identify potential health problems.

[0047] The library learning and updating module 700 is used to establish a medical reserve library, add labels to each disease in the medical reserve library, and store the relevant data and final treatment plan of each patient in the medical reserve library.

[0048] Through precise data collection, storage, analysis, and updating mechanisms, it not only promotes the accumulation and utilization of medical knowledge, but also significantly improves the accuracy of medical decisions and the effectiveness of treatment plans. This module establishes and maintains a comprehensive medical repository for storing relevant data on various diseases. This repository not only includes standard disease information such as symptoms, treatments, and prognosis, but also includes different tag indexes such as disease stage, patient population (age, gender, etc.), and genetic information to facilitate more detailed data analysis and application.

[0049] In this module, each patient's disease data, treatment plan, and results are systematically stored under the corresponding disease tag. This includes not only basic diagnosis and treatment information, but also patient feedback, follow-up data, and long-term tracking results, providing a rich, multi-dimensional data resource for the medical reserve library.

[0050] Figure 2 The structural block diagram of the data acquisition module provided by the embodiment of the present invention is as follows: Figure 2 As shown, the data acquisition module includes: The institution docking unit 110 is used to connect with the information system of each medical institution, establish a communication connection, and automatically obtain and transmit relevant patient data; The data management unit 120 is used to store, integrate and manage the collected patient-related data and add a patient name tag to each patient's data; The data encryption unit 130 encrypts the data of each patient.

[0051] Figure 3 The structural block diagram of the data analysis module provided by the embodiment of the present invention is as follows: Figure 3 As shown, the data analysis module includes: A feature selection unit 210 is used to extract individual differences and features from the relevant data of each patient, including the patient's body shape data, genetic characteristics, clinical data and current disease data; Individualized risk assessment unit 220, for assessing the patient's current disease transformation risk and predicting the potential path and speed of disease progression based on the patient's clinical data and current disease data; The effect evaluation unit 230 is used to analyze all data of the patients who have received treatment and evaluate the effectiveness of the treatment plan for the patients.

[0052] As a further embodiment of the present invention, in the feature selection unit, extracting individual differences and features from the relevant data of each patient, including the patient's body shape data, genetic characteristics, clinical data, and current disease data, specifically comprises the following steps: Acquiring an original patient data set, the original patient data set including individual difference data and feature data, and performing standardization processing on the original patient data set to obtain a standardized patient data set, the standardized patient data set including standardized individual difference data and standardized feature data; ANOVA F value was used to analyze the F value scores between the standardized individual difference data and the standardized feature data, and the individual difference data with the highest F value scores and the corresponding feature data were selected in descending order to form the first data set; The mutual information method is used to analyze the nonlinear relationship between the standardized individual difference data and the standardized feature data to calculate the information gain value, and the individual difference data with the highest information gain value and the corresponding feature data are selected in descending order to form the second data set; The principal component analysis method was used to analyze the principal component sharing degree of the standardized individual difference data, and the individual difference data with the highest principal component sharing degree and the corresponding characteristic data were selected in descending order to form the third data set; The number of times the individual difference data and characteristic data are selected in the first data set, the second data set, and the third data set is counted, and the number of times selected is used as a measurement indicator to obtain the individual differences and characteristics in the patient-related data.

[0053] As a further embodiment of the present invention, in the personalized risk assessment unit, the step of assessing the patient's current disease transformation risk and predicting the potential path and speed of disease progression based on the patient's clinical data and current disease data specifically includes the following steps: The patient's clinical data includes several clinical characteristics, and a first interaction term is created based on the interaction relationship between different clinical characteristics; Based on the Cox model, a multivariate model was constructed with several clinical characteristics and the first interaction term as multiple variables; The multivariate model was incrementally fitted using stepwise regression to construct the optimal multivariate model; The coefficient of each variable was estimated by maximum likelihood estimation to fit the optimal multivariate model. Based on the fitting results, the contribution of each variable to the baseline risk was calculated to obtain the contribution of clinical characteristics to the overall risk. extracting genomic data from the genetic signatures, and determining genetic markers associated with the target disease from the genomic data; Perform a single-factor statistical test on each genetic marker, calculate the significance p-value of each genetic marker, evaluate the impact of the genetic marker on the disease, and obtain several original p-values; Bonferroni correction was used to adjust several original p values ​​to obtain several significant p values; Based on Fisher's method, several significant p-values ​​are combined to evaluate the genetic risk score of the overall genetic markers and obtain the overall contribution of genetic characteristics to disease risk; Obtain current and historical disease data; A time series analysis model is constructed using a dynamic Bayesian network, and the time series analysis model is trained using historical disease data to obtain a trained time series analysis model; Input current disease data into the trained time series analysis model to obtain future risk status; Based on the future risk status, the risk level at different time points is calculated using a weighted average method to obtain the comprehensive disease risk, and the potential path and speed of disease progression are generated based on the risk level at different time points; The contribution of clinical characteristics to the overall risk, the overall contribution of genetic characteristics to the disease risk, and the comprehensive disease risk are combined in a weighted combination to obtain a comprehensive risk assessment result and the current disease transformation risk.

[0054] As a further solution of the present invention, in the effect evaluation unit, analyzing all data of the treated patient and evaluating the effectiveness of the treatment plan for the patient specifically includes the following steps: All data of the patients who have received treatment include the patient's body shape data, genetic characteristics and clinical data, and the patient's body shape data, genetic characteristics and clinical data are standardized to obtain standardized patient body shape data, standardized genetic characteristics and standardized clinical data; The fully connected layer is used to perform nonlinear transformation on the standardized patient body shape data to obtain body shape feature representation; The fully connected layer is used to perform nonlinear transformation on the standardized genetic features to obtain the genetic feature representation; The standardized clinical data is input into the long short-term memory network layer, which is used to capture the dynamic change trend of the patient's status over time and obtain the clinical feature representation; The body shape feature representation, genetic feature representation and clinical feature representation are fused in a splicing manner to obtain a comprehensive feature vector; Apply self-attention technology to the comprehensive feature vector to obtain attention output; Pass the attention output through the output layer to get the prediction result; Obtain the known efficacy of the treatment plan, and perform weighted calculation on the known efficacy of the treatment plan and the predicted results according to the weight to obtain the comprehensive treatment effect index; A comparison threshold is set and the comprehensive treatment effect index is compared with the comparison threshold. If the comprehensive treatment effect index exceeds the comparison threshold, the treatment plan is considered effective; otherwise, the treatment plan is considered ineffective.

[0055] Figure 4 This is a structural block diagram of the referral decision support module provided by an embodiment of the present invention, such as Figure 4 As shown, the referral decision support module includes: The medical resource analysis unit 310 is used to obtain medical resource information of all medical institutions in the current medical institution and the area where the patient is located, including the departments, medical equipment, vacant beds and medical level of each medical institution; The patient needs assessment unit 320 is used to evaluate the medical level of the current medical institution based on the patient's disease development trend and current treatment effect, analyze the patient's treatment needs and the availability of medical resources of the medical institution, and determine whether the medical institution has the medical resources required to cure the corresponding patient; The regional institution screening unit 330 is used to screen all medical institutions in the patient's area that can meet the patient's treatment needs based on the patient's treatment needs analysis results when the medical institution does not have the medical resources required to treat the patient; The referral marking unit 340 is used to obtain the medical institution selected by the patient, establish communication with the medical institution, synchronize all the patient's data to the information database of the medical institution, and add a transfer mark for the patient.

[0056] Figure 5 The structural block diagram of the comprehensive data sharing module provided by the embodiment of the present invention is as follows: Figure 5 As shown, the comprehensive data sharing module includes: The communication establishment unit 410 is used to establish a data sharing platform and, based on the coverage of the data sharing platform, establish communication channels with all medical institutions within the coverage area; The data sharing permission unit 420 is used to set the access permission of the data sharing platform, wherein the access permission is set to be accessible only to doctors; The data upload synchronization unit 430 is used to upload the patient information that needs to be referred to the data sharing platform, and encrypt the uploaded data, and the encryption key will only be synchronized to the medical institution selected by the patient.

[0057] Figure 6 This is a structural block diagram of the health management monitoring module provided by an embodiment of the present invention, such as Figure 6 As shown, the health management monitoring module includes: The real-time monitoring unit 510 is used to connect with the sensor device worn by the patient to collect the patient's physiological parameter data and the patient's disease development trend data; The standard setting unit 520 is used to set a personalized physiological expected threshold value based on the physiological parameter data and current disease status of each patient; The abnormality detection unit 530 is used to compare and analyze the monitoring data in real time, and compare each acquired monitoring data with the set physiological expectation threshold. When any monitoring data exceeds the physiological expectation threshold, a warning message is issued.

[0058] Figure 7 The structural block diagram of the autonomous participation management module provided by the embodiment of the present invention is as follows: Figure 7 As shown, the autonomous participation management module includes: The application information receiving unit 610 is used to interact with the patient, receive the patient's needs, consultation and appointment information, and receive the doctor's response to the interactive information and feed it back to the patient; The parameter autonomous uploading unit 620 is used to receive physiological parameter data actively uploaded by the patient.

[0059] Figure 8 The structural block diagram of the library learning and updating module provided by the embodiment of the present invention is as follows: Figure 8 As shown, the library learning and updating module includes: A library building unit 710 is used to build a medical storage library to store disease-related data and add a label index to each disease in the medical storage library; The patient data storage unit 720 is used to store the patient's information in a tag under the same index based on the patient's disease data, and obtain the treatment plan and treatment effect adopted by the patient this time, and bind and store them in the tag.

[0060] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0061] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0062] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0063] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0064] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A smart medical platform based on big data, It is characterized in that The platform includes: The data acquisition module is used to connect with the information system of medical institutions to automatically obtain and manage relevant patient data, including medical records, physiological parameters, genomic data and medical records; The data analysis module is used to analyze the automatically acquired data, identify individual differences and characteristics in the patient group, and combine the relevant data of each patient to predict the disease development trend and evaluate the current treatment effect of the patient; The referral decision-making support module is used to obtain the availability of medical resources of the current medical institution and medical institutions within the region, and to determine whether the availability of medical resources of the current medical institution can meet the patient's cure needs in combination with the patient's disease development trend prediction and current treatment effect. If it cannot be met, all medical institutions within the region that can meet the patient's cure needs are identified, and a transfer mark is added for the patient; A comprehensive data sharing module is used to establish a data sharing platform within the region, and the data sharing platform establishes communication channels with all medical institutions within the region, and the patient-related data with transfer marks are placed in the data sharing platform for encrypted storage; The health management monitoring module is used to monitor the patient's physiological parameters and disease development trend data in real time, and set physiological expected thresholds based on the relevant data of each patient, and issue warning information when the monitored parameters and data exceed the physiological expected thresholds; The autonomous participation management module is used to receive the patient's application information and the physiological parameters actively uploaded by the patient, and to feed back the doctor's feedback instructions to the patient; The library learning and updating module is used to establish a medical reserve library and add labels to each disease in the medical reserve library, storing the relevant data and final treatment plan of each patient in the medical reserve library.

2. According to the big data-based smart medical platform of claim 1, It is characterized in that The data acquisition module comprises: The institution docking unit is used to connect with the information system of each medical institution, establish a communication connection, and automatically obtain and transmit relevant patient data; A data management unit is used to store, integrate and manage the collected patient-related data and add a patient name tag to each patient's data; The data encryption unit encrypts the data of each patient.

3. According to the big data-based smart medical platform of claim 2, It is characterized in that The data analysis module includes: A feature selection unit is used to extract individual differences and features from each patient's relevant data, including patient body shape data, genetic characteristics, clinical data, and current disease data; Individualized risk assessment unit, which is used to assess the patient's current disease transformation risk and predict the potential path and speed of disease progression based on the patient's clinical data and current disease data; The effectiveness evaluation unit is used to analyze all data of patients who have received treatment and evaluate the effectiveness of the treatment plan for the patient.

4. According to the big data-based smart medical platform of claim 3, It is characterized in that In the feature selection unit, the extraction of individual differences and features in the relevant data of each patient, including the patient's body shape data, genetic characteristics, clinical data and current disease data, specifically includes the following steps: Acquire an original patient data set, the original patient data set includes individual difference data and feature data, perform standardization on the original patient data set to obtain a standardized patient data set, the standardized patient data set includes standardized individual difference data and standardized feature data; ANOVA F value was used to analyze the F value scores between the standardized individual difference data and the standardized feature data, and the individual difference data with the first several high F value scores and the corresponding feature data were selected in descending order to form the first data set; The mutual information method is used to analyze the nonlinear relationship between the standardized individual difference data and the standardized feature data to calculate the information gain value, and the individual difference data with the highest information gain value and the corresponding feature data are selected in descending order to form the second data set; The principal component analysis method is used to analyze the principal component sharing degree of the standardized individual difference data, and the first several individual difference data with high principal component sharing degree and the corresponding characteristic data are selected in descending order to form the third data set; The number of times the individual difference data and characteristic data are selected in the first data set, the second data set, and the third data set is counted, and the number of times selected is used as a measurement indicator to obtain individual differences and characteristics in the patient-related data.

5. According to the big data-based smart medical platform of claim 4, It is characterized in that In the individualized risk assessment unit, the step of assessing the patient's current disease transformation risk and predicting the potential path and speed of disease progression through the patient's clinical data and current disease data specifically includes the following steps: The clinical data of the patient includes several clinical characteristics, and a first interaction term is created according to the interaction relationship between different clinical characteristics; Based on the Cox model, a multivariate model was constructed with several clinical characteristics and the first interaction term as multiple variables; The multivariate model was incrementally fitted using stepwise regression operations to construct the optimal multivariate model; The coefficient of each variable was estimated by the maximum likelihood estimation method to fit the optimal multivariate model. Based on the fitting results, the contribution of each variable to the baseline risk was calculated to obtain the contribution of clinical characteristics to the overall risk. extracting genomic data from the genetic features, and determining genetic markers associated with the target disease from the genomic data; Perform a single-factor statistical test on each genetic marker, calculate the significant p-value of each genetic marker, evaluate the impact of the genetic marker on the disease, and obtain several original p-values; Bonferroni correction was used to adjust several original p values ​​to obtain several significant p values; Based on Fisher's method, several significant p values ​​were combined to evaluate the genetic risk score of the overall genetic markers and obtain the overall contribution of genetic characteristics to disease risk; Obtain current and historical disease data; A time series analysis model is constructed using a dynamic Bayesian network, and the time series analysis model is trained using historical disease data to obtain a trained time series analysis model; Input current disease data into the trained time series analysis model to obtain future risk status; Based on the future risk status, the risk level at different time points is calculated by weighted average to obtain the comprehensive disease risk, and the potential path and speed of disease progression are generated based on the risk level at different time points; The contribution of clinical characteristics to the overall risk, the overall contribution of genetic characteristics to disease risk, and the comprehensive disease risk are combined in a weighted combination to obtain a comprehensive risk assessment result and the current disease change risk.

6. According to the big data-based smart medical platform of claim 5, It is characterized in that In the effect evaluation unit, analyzing all the data of the patients who have received treatment and evaluating the effectiveness of the treatment plan for the patients specifically includes the following steps: All data of the patients who have received treatment include patient body shape data, genetic characteristics and clinical data, and the patient body shape data, genetic characteristics and clinical data are standardized to obtain standardized patient body shape data, standardized genetic characteristics and standardized clinical data; The standardized patient body shape data is transformed nonlinearly using the fully connected layer to obtain body shape feature representation; The fully connected layer is used to perform nonlinear transformation on the standardized genetic features to obtain the genetic feature representation; The standardized clinical data is input into the long short-term memory network layer, and the long short-term memory network layer is used to capture the dynamic change trend of the patient's status over time to obtain the clinical feature representation; The body shape feature representation, genetic feature representation and clinical feature representation are fused by splicing to obtain a comprehensive feature vector; The comprehensive feature vector is subjected to self-attention technique to obtain the attention output; Pass the attention output through the output layer to get the prediction result; Obtain the known efficacy of the treatment plan, and perform weighted calculation on the known efficacy of the treatment plan and the predicted results according to the weights to obtain a comprehensive treatment effect index; A comparison threshold is set and the comprehensive treatment effect index is compared with the comparison threshold. If the comprehensive treatment effect index exceeds the comparison threshold, the treatment plan is considered effective; otherwise, the treatment plan is considered ineffective.

7. According to the big data-based smart medical platform of claim 6, It is characterized in that The referral auxiliary decision module includes: The medical resource analysis unit is used to obtain the medical resource information of all medical institutions in the current medical institution and the area where the patient is located, including the departments, medical equipment, vacant beds and medical level of each medical institution; The patient needs assessment unit is used to evaluate the medical level of the current medical institution based on the patient's disease development trend and current treatment effect, analyze the patient's treatment needs and the availability of medical resources of the medical institution, and determine whether the medical institution has the medical resources required to cure the corresponding patient; The regional institution screening unit is used to screen out all medical institutions that can meet the patient's treatment needs from other medical institutions within the patient's area according to the patient's treatment needs analysis results when the medical institution does not have the medical resources required to cure the corresponding patient; The referral marking unit is used to obtain the medical institution selected by the patient, establish communication with the medical institution, synchronize all the patient's data to the information database of the medical institution, and add a transfer mark for the patient.

8. According to the big data-based smart medical platform of claim 7, It is characterized in that The comprehensive data sharing module includes: A communication building unit is used to establish a data sharing platform and, based on the coverage of the data sharing platform, establish communication channels with all medical institutions within the coverage area; A data sharing permission unit, used to set the access permission of the data sharing platform, wherein the access permission is set to be accessible only to doctors; The data upload synchronization unit is used to upload the patient information that needs to be referred to the data sharing platform and encrypt the uploaded data. The encryption key will only be synchronized to the medical institution selected by the patient.

9. According to the big data-based smart medical platform of claim 8, It is characterized in that The health management monitoring module includes: A real-time monitoring unit is used to connect with the sensor equipment worn by the patient to collect the patient's physiological parameter data and the patient's disease development trend data; A standard setting unit, used to set personalized physiological expected thresholds based on each patient's physiological parameter data and current disease status; The anomaly detection unit is used to compare and analyze the monitoring data in real time, and compare each acquired monitoring data with the set physiological expected threshold. When any monitoring data exceeds the physiological expected threshold, a warning message is issued.

10. According to the big data-based smart medical platform of claim 9, It is characterized in that The autonomous participation management module includes: The application information receiving unit is used to interact with the patient, receive the patient's needs, consultation and appointment information, and receive the doctor's reply to the interactive information and feed it back to the patient; Parameter autonomous uploading unit, used to receive physiological parameter data actively uploaded by patients; The library learning update module includes: The library establishment unit is used to build a medical library to store disease-related data and add a label index to each disease in the medical library; The patient data storage unit is used to store the patient's information in a tag under the same index according to the patient's disease data, and obtain the treatment plan and treatment effect adopted by the patient this time, and bind and store them in the tag.

Images