Big data artificial intelligence monitoring system

Abstract

The invention discloses a big data artificial intelligence monitoring system, and relates to the technical field of medical monitoring. Comprising a medical data acquisition module and a central processing module, the medical data acquisition module is used for transmitting information data of a patient acquired in real time to the central processing module; the central processing module obtains a plurality of alternative treatment schemes corresponding to the information data acquired by the medical data acquisition module according to the disease-treatment scheme mapping model, and obtains medical resource demands of the plurality of alternative treatment schemes according to the medical resource demand matrix; taking the minimum difference between the medical resource demand of the alternative disposal scheme and the available medical resource as an optimization target, taking a cross-department cooperation rule as a constraint condition to construct an optimization model, and solving the optimization; and taking the alternative disposal scheme which enables the optimization target to be optimal as priority recommendation.

Description

Technical Field

[0001] The present invention relates to the technical field of medical monitoring, and particularly relates to a big data artificial intelligence monitoring system. Background Art

[0002] In the modern medical system, first aid treatment is an important means to save lives, prevent the deterioration of the condition, and promote the recovery of patients. The scientific nature and efficiency of the first aid treatment process are directly related to the survival rate and recovery quality of patients. With the wide application of 5G and Internet of Things technologies, and the continuous popularization of portable instruments such as in-vehicle electrocardiogram monitoring devices, medical real-time monitoring has gradually become a reality.

[0003] A certain hospital used technologies such as 5G data collection, transmission, and integration platform construction to build a pre-hospital and in-hospital intelligent first aid center. For pre-hospital first aid information, such as the patient's name, age, past medical history, clinical symptoms, etc., the information is directly entered, photos are taken or videos are recorded in the first aid passage. After receiving the patient's information in real time, emergency doctors conduct remote consultations to achieve seamless docking between pre-hospital first aid and in-hospital treatment. During the in-hospital treatment process, medical staff use the emergency green channel information management system to collect data at important nodes during the patient's treatment process and notify relevant departments to automatically feedback the examination results.

[0004] Although the design of this first aid system has great potential in improving medical efficiency, it still faces some challenges in actual applications. The smooth progress of the first aid process is closely related to the allocation of medical resources. The medical resources required for different first aid treatment processes vary greatly, and the large differences in the demand for medical resources in these first aid processes make it difficult to efficiently allocate resources when medical resources are scarce. Summary of the Invention

[0005] Based on this, in order to solve the technical problems existing in the prior art, the present invention provides a big data artificial intelligence monitoring system.

[0006] The present invention provides a big data artificial intelligence monitoring system, including:

[0007] including a medical data collection module and a central processing module;

[0008] The medical data collection module is used to collect the physical condition information of the patient and the basic judgment information of the medical staff on the patient's symptoms in real time, and send the collected information data to the central processing module;

[0009] The central processing module includes:

[0010] a database unit, storing a disease-treatment plan mapping model, a medical resource demand matrix, cross-department cooperation rules, and available medical resources;

[0011] A data analysis unit, configured to obtain a number of alternative treatment plans corresponding to the information data collected by the medical data collection module according to the disease-treatment plan mapping model, and obtain the medical resource requirements of the number of alternative treatment plans according to the medical resource demand matrix;

[0012] A process recommendation unit, configured to construct an optimization model with the minimum gap between the medical resource requirements of the alternative treatment plans and the available medical resources as the optimization objective and the cross-department collaboration rules as the constraint conditions, solve the optimization, and recommend the alternative treatment plan that makes the optimization objective optimal as the priority treatment plan.

[0013] Further, the disease-treatment plan mapping model specifically includes:

[0014] A multi-modal feature fusion layer, configured to perform multi-modal fusion on the information data collected by the medical data collection module to obtain fusion features;

[0015] A probability reasoning layer, configured to calculate the similarity weight between the fusion features and the pre-stored treatment plan vectors through an attention mechanism.

[0016] Further, the number of alternative treatment plans are the top K treatment plans with the highest similarity weights output.

[0017] Further, the medical resource demand matrix is saved in the form of a dictionary, where the keys in the dictionary are treatment plans, and the values in the dictionary are the corresponding medical resource requirements.

[0018] Further, the cross-department collaboration rules specifically include:

[0019] Department resource capacity constraint:

[0020]

[0021] Among them, x i is the selection variable of treatment plan i, and the value range is {0, 1}; is the resource requirement of treatment plan i for department k; B k is the maximum resource capacity of department k;

[0022] Collaboration mutual exclusion constraint:

[0023]

[0024] Among them, C is the set of department pairs with complementary relationships; means that only one of department k1 and department k2 participates in treatment plan i;

[0025] Task concurrency constraint:

[0026]

[0027] Among them, J is the set of treatment plans to be executed, is the resource requirement of the treatment plan j to be executed for the department k; I is the set of executed treatment plans, is the resource occupancy of the executed treatment plan i for the department k;

[0028] Priority resource reservation constraint:

[0029]

[0030] Among them, w k is the priority weight of the department k, and P k is the reserved resource threshold.

[0031] Furthermore, the medical data collection module includes:

[0032] A multi-modal biosensor array unit configured to collect the vital sign data of the patient in real time, including but not limited to electrocardiogram signals, blood oxygen saturation, respiratory rate, and body surface temperature;

[0033] An interactive medical information collection unit equipped with a structured symptom entry interface for receiving clinical observation indicators, preliminary diagnosis opinions, and imaging examination identification codes input by medical staff;

[0034] A data encryption and transmission unit that classifies and codes the data obtained by the multi-modal biosensor array unit and the interactive medical information collection unit using an encryption algorithm and transmits it in real time;

[0035] A voice and video broadcast unit for feeding back the alternative treatment plan that optimizes the optimization target to medical staff through voice or video broadcast.

[0036] Furthermore, the medical resources include drug medical resources, medical staff human resources, medical device resources, and bed resources.

[0037] The above at least one technical solution adopted by the present invention can achieve the following beneficial effects:

[0038] In the big data artificial intelligence monitoring system provided by the present invention, by collecting the patient's vital sign and medical resource status data in real time and combining the disease-treatment plan mapping model, the system dynamically calculates the current resource gap to ensure that the recommended plan is accurately matched with the real-time resource supply; at the same time, with the resource gap minimization as the objective function and the cross-department collaboration rules embedded, the dual optimization of resource allocation and clinical treatment efficiency is achieved, helping hospitals or medical institutions to reasonably schedule and allocate medical resources when resources are scarce, and avoiding over-reliance on single resources or processing flows. Brief Description of the Drawings

[0039] The accompanying drawings described herein are used to provide a further understanding of the present invention and form a part of the present invention. The illustrative embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0040] Figure 1 It is a schematic diagram of a big data artificial intelligence monitoring system framework provided by the present invention. Specific embodiments

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below in conjunction with specific embodiments of the present invention and the corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the specification, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present invention.

[0042] Embodiment 1

[0043] Figure 1 Shows the big data artificial intelligence monitoring system of this embodiment. The following specifically combines Figure 1 Describe the system in detail, specifically including the following modules:

[0044] A medical data acquisition module, configured to collect the physical condition information of patients and the basic judgment information of medical staff on the symptoms of patients in real time, and send the collected information data to the central processing module, specifically including:

[0045] A multimodal biosensor array unit, configured to collect the vital sign data of patients in real time, including but not limited to electrocardiogram signals, blood oxygen saturation, respiratory rate, and body surface temperature. The multimodal biosensor array unit is a system integrating multiple sensors, which can non-invasively monitor the key vital signs of patients. These sensors have the characteristics of high sensitivity, high precision, and low latency, ensuring the real-time and reliability of the data

[0046] An interactive medical information acquisition unit, equipped with a structured symptom entry interface, for receiving the clinical observation indicators, preliminary diagnosis opinions, and imaging examination identification codes input by medical staff. The interactive medical information acquisition unit provides a user-friendly interface, allowing medical staff to quickly enter the clinical symptoms, preliminary diagnosis, and necessary examination information of patients. The interface supports structured data input. For example, in the preliminary diagnosis column, the doctor enters the imaging examination identification code, such as "CT12345", to indicate the subsequent cardiac CT scan that needs to be performed.

[0047] The data encryption and transmission unit classifies and encodes the data obtained by the multi-modal biosensor array unit and the interactive medical information acquisition unit using an encryption algorithm, and performs real-time transmission.

[0048] The voice and video broadcast unit is used to feedback the alternative treatment plan that optimizes the optimization objective to the medical staff by voice or video broadcast.

[0049] The central processing module includes a database unit, a data analysis unit, and a process recommendation unit. The central processor receives and analyzes the multi-source data collected by the medical data acquisition module, and outputs an emergency treatment plan corresponding to the patient's data. Specifically:

[0050] The database unit stores a disease-treatment plan mapping model, a medical resource demand matrix, cross-department collaboration rules, and available medical resources.

[0051] The disease-treatment plan mapping model includes: a multi-modal feature fusion layer, which inputs the information data collected by the medical data acquisition module into a multi-modal model constructed based on a deep learning framework for multi-modal fusion, and then splices their outputs to obtain fusion features; a probability inference layer, which calculates the similarity weights between the fusion features and the pre-stored treatment plan vectors through an attention mechanism, and outputs the top K treatment plans with higher similarity weights. The essence of this model is to correspond the disease information contained in the multi-modal data with the treatment plans, as shown in Table 1:

[0052] Table 1 Disease information and corresponding treatment plans

[0053]

[0054]

[0055] The medical resource demand matrix is stored in dictionary form. The keys in the dictionary are treatment plans, and the values are the corresponding medical resource demands. For example, the key in the dictionary is {drug treatment}, and the value is {drugs = ["aspirin", "heparin"]; doctor = 1; observation room bed = 1}. The medical resource demands of several alternative treatment plans can be obtained according to the medical resource demand matrix.

[0056] Available medical resources can track the real-time status of all medical resources in the hospital, including medical resources, medical staff human resources, medical device resources, bed resources, etc. Compare the resource requirement list of the treatment plan with the available resources monitored in real time. For each resource requirement, check whether there are sufficient available resources to meet the requirement. If the available quantity of a certain resource is lower than the required quantity, it is marked as a resource gap, and the type, quantity, and urgency level of the gap resource are recorded. According to the resource gap, take measures such as emergency procurement, dispatching resources from other departments, and seeking help from external institutions to obtain the probability and time to fill the resource gap, and promptly notify the medical staff of the information related to the resource gap so that they can make corresponding treatment adjustments. Suppose a heart disease patient needs to undergo a PCI operation immediately. The emergency treatment process output by the process matching module requires the following resources: 1 cardiac catheterization laboratory, 1 set of PCI operation instruments, 1 cardiologist, 2 nurses, and a number of relevant drugs. Query the usage status of the current hospital's cardiac catheterization laboratory, the inventory of surgical instruments, the work schedules of cardiologists and nurses, and the drug inventory through the resource monitoring system. After comparison, it is found that the cardiac catheterization laboratory is in use, the surgical instruments are complete, but the drug inventory is insufficient. It is identified that the cardiac catheterization laboratory being in use is a resource gap, and the insufficient drug inventory is another resource gap. Notify the medical staff to adjust the treatment plan, and at the same time, urgently dispatch drugs and prepare other operating rooms as alternatives.

[0057] The process recommendation unit is used to construct an optimization model with the minimum gap between the medical resource requirements of the alternative treatment plan and the available medical resources as the optimization goal and the cross-departmental collaboration rules as the constraint conditions, and solve the optimization. The alternative treatment plan that makes the optimization goal optimal is recommended as the priority treatment plan. The optimization goal is to minimize the gap between the medical resource requirements of the alternative treatment plan and the currently available medical resources in the hospital; the constraint conditions are used to constrain the resource capabilities: the resource requirements of each treatment plan cannot exceed the available quantity of the corresponding resources in the hospital, because the treatment plan must comply with the rules of cross-departmental collaboration within the hospital. For example, the resources of a specific department can only be used by other departments under specific conditions. For example, the catheterization laboratory can only be used by the cardiology department, and the interventional nurses can only cooperate with cardiologists. Use linear programming or integer programming algorithms (such as CPLEX, Gurobi, etc.) to solve the optimization model, find the treatment plan that minimizes the resource gap, and according to the optimization results, select the treatment plan with the smallest gap as the priority treatment plan. In this way, the hospital can ensure that under limited resources, the most appropriate treatment plan is selected while complying with the cross-departmental collaboration rules, improving the efficiency and effectiveness of medical services.

[0058] Among them, the cross-departmental collaboration rules include:

[0059] Department resource capacity constraint:

[0060]

[0061] Among them, x i is the selection variable of disposal plan i, and its value range is {0, 1}; is the resource requirement of disposal plan i for department k; B k is the maximum resource capacity of department k.

[0062] Cooperation mutual exclusion constraint:

[0063]

[0064] Among them, C is the set of department pairs with complementary relationships; means that only one of department k1 and department k2 participates in disposal plan i.

[0065] Task concurrency constraint:

[0066]

[0067] Among them, J is the set of disposal plans to be executed, is the resource requirement of disposal plan j to be executed for department k; I is the set of executed disposal plans, is the resource occupancy of executed disposal plan i for department k.

[0068] Priority resource reservation constraint:

[0069]

[0070] Among them, w k is the priority weight of department k, P k is the reserved resource threshold.

[0071] The system has the following benefits:

[0072] The entire system assists medical staff in making more accurate and rapid diagnoses, which is crucial for the timely treatment of patients. By recommending the treatment process with the smallest gap in medical resources, the hospital can allocate resources more efficiently and reduce resource waste. In emergency situations, the system can quickly identify and recommend the optimal treatment process, buying precious treatment time for patients. The assistance of artificial intelligence reduces misdiagnosis and missed diagnosis caused by human factors, thus improving the overall quality of medical services. Data-based decision-making helps hospital management better understand the usage of medical resources and provides strong data support for hospital management. The process recommendation of the system promotes the standardization of medical behavior and reduces the inconsistency of medical behavior caused by individual habits or experience differences. By analyzing the gaps in medical resources of different first-aid treatment processes, the system can recommend the optimal first-aid process, thereby achieving the rational allocation of medical resources and reducing resource waste. The recommended first-aid treatment process can ensure that in the case of limited resources, the most critical patients or conditions are given priority treatment, improving the treatment efficiency. Providing immediate process recommendations for medical staff reduces the decision-making time required in emergency situations, thus accelerating the treatment speed.

[0073] Those of ordinary skill in the art can understand that all or part of the processes in the above-described embodiment systems can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-described system embodiments. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided by the present invention can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.

[0074] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the various technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded by the present invention.

Claims

1. A big data artificial intelligence monitoring system, characterized in that, It includes a medical data collection module and a central processing module; The medical data collection module is used to collect the physical condition information of patients and the basic judgment information of medical staff on patients' symptoms in real time, and send the collected information data to the central processing module; The central processing module includes: A database unit that stores a disease-treatment plan mapping model, a medical resource demand matrix, cross-department collaboration rules, and available medical resources; A data analysis unit that is used to obtain several alternative treatment plans corresponding to the information data collected by the medical data collection module according to the disease-treatment plan mapping model, and obtain the medical resource requirements of several alternative treatment plans according to the medical resource demand matrix; A process recommendation unit that is used to construct an optimization model with the minimum gap between the medical resource requirements of alternative treatment plans and available medical resources as the optimization goal and cross-department collaboration rules as the constraint conditions, solve the optimization, and recommend the alternative treatment plan that makes the optimization goal optimal as the priority treatment plan.

2. The big data artificial intelligence monitoring system according to claim 1, wherein, The disease-treatment plan mapping model specifically includes: A multi-modal feature fusion layer that is used to perform multi-modal fusion on the information data collected by the medical data collection module to obtain fusion features; A probability reasoning layer that is used to calculate the similarity weight between the fusion features and the pre-stored treatment plan vectors through an attention mechanism.

3. The big data artificial intelligence monitoring system according to claim 2, wherein The several alternative treatment plans are to output the top K treatment plans with higher similarity weights.

4. The big data artificial intelligence monitoring system according to claim 1, wherein The medical resource demand matrix is saved in the form of a dictionary, where the keys in the dictionary are treatment plans, and the values in the dictionary are the corresponding medical resource requirements.

5. The big data artificial intelligence monitoring system according to claim 1, wherein The cross-department collaboration rules specifically include: Department resource capacity constraint: ∑x i ·R i k <<B k where x i is the selection variable of disposal plan i, and its value range is {0, 1}; is the resource requirement of disposal plan i for department k; B k is the maximum resource capacity of department k; Collaboration mutual exclusion constraint: Among them, C is a set of pairs of departments with complementary relationships; It means that only one of department k1 and department k2 participates in the treatment plan i; Task concurrency constraint: Among them, J is the set of treatment plans to be executed, is the resource requirement of the treatment plan j to be executed for department k; I is the set of executed treatment plans, is the resource occupancy of the executed treatment plan i for department k; Priority resource reservation constraint: Among them, w k is the priority weight of department k, and P k is the reserved resource threshold.

6. The big data artificial intelligence monitoring system according to claim 1, characterized in that, The medical data collection module includes: A multi-modal biosensor array unit configured to collect patients' vital sign data in real time, including but not limited to electrocardiogram signals, blood oxygen saturation, respiratory rate, and body surface temperature; An interactive medical information collection unit equipped with a structured symptom entry interface for receiving clinical observation indicators, preliminary diagnosis opinions, and imaging examination identification codes input by medical staff; A data encryption and transmission unit that uses an encryption algorithm to classify and encode the data obtained by the multi-modal biosensor array unit and the interactive medical information collection unit, and transmits it in real time; A voice and video broadcast unit that is used to feedback the alternative treatment plan that makes the optimization goal optimal to medical staff in the form of voice or video broadcast.

7. The big data artificial intelligence monitoring system according to claim 1, wherein The medical resources include drug medical resources, medical staff human resources, medical device resources, and bed resources.

Images