A medical equipment management method and system based on big data
Through big data and neural network prediction models, fine-grained management of medical equipment is realized, which solves the shortcomings of periodic management and problem-driven management in the existing technology, improves the efficiency and accuracy of equipment management, and ensures the normal operation of equipment and safe.
Patent Information
- Application Number
- CN202310188501.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-03-02
- Publication Date
- 2024-12-17
- Estimated Expiration
- 2043-03-02
AI Technical Summary
The existing medical equipment management methods have problems such as periodic management that leads to excessive operation of equipment to reduce service life and increase labor costs, as well as problems that lead to abnormal operation of equipment that cannot be solved in time.
Using a neural network prediction model based on big data, a triple is constructed by obtaining maintenance records of medical equipment, divided into training sets and test sets, using the BP neural network prediction model to predict the target maintenance time point, and correct it in advance according to the device status or Reset maintenance time points to achieve fine-grained management.
It improves the efficiency and accuracy of medical equipment management, avoids excessive frequency of periodic management and lag in problem-driven management, and ensures the normal operation and safety of equipment.
Smart Images
Figure CN116189868B_ABST
Abstract
Description
Technical Field
[0001] The present application belongs to the field of big data processing, and in particular, relates to a medical equipment management method and system based on big data. Background Art
[0002] Medical equipment is the basic infrastructure of medical institutions and is widely used in diagnosis, treatment, health care, rehabilitation, scientific research, clinical trials, etc. In large medical institutions, there are many types and quantities of medical equipment. The management of medical equipment is an important task. Good medical equipment management can maintain or even enhance the value of the medical equipment itself, reduce the cost of medical institutions, and improve medical efficiency, thereby fundamentally improving the medical service capabilities of medical institutions. In the prior art, the management methods for medical equipment include inspection, repair, maintenance, care, recording, testing, parameter correction and other operations. When performing management operations, it is usually for the same type of medical equipment (1) based on a specific periodic time to manage the medical equipment, (2) when the equipment has an error alarm, the equipment cannot operate normally, or the user finds that the equipment has an operating problem, the professional personnel will manage the equipment. For method (1), although medical equipment management can be completed and medical equipment can operate normally as much as possible, the types and quantities of medical equipment are numerous, and the management cycles and requirements of different types of medical equipment or different medical equipment of the same type are uneven. Fixed periodic management operations are likely to lead to excessive operation of medical equipment, reduce service life, and increase labor costs. For method (2), problem-driven medical equipment management operations, although they reduce the management frequency of medical equipment and labor costs to a certain extent, the normal operation of medical equipment cannot be guaranteed. Although medical equipment can be operated through multiple parallel backups to solve the problem of abnormal operation of a small number of equipment, for high-precision, high-value or high-load equipment, even the abnormal operation of a few medical equipment is unbearable and easily leads to adverse consequences. Summary of the invention
[0003] In view of the problems existing in the prior art, the present invention provides a medical equipment management method and system based on big data.
[0004] Specifically, the medical device management method based on big data claimed in this application includes:
[0005] S110, obtaining an initial medical device maintenance record; the maintenance record includes a medical device identifier, a maintenance time point, and a maintenance content, and each maintenance record constitutes a triple (ID, T, Content) for storage;
[0006] S120, establishing a neural network prediction model based on the maintenance record triples; dividing the triples in the storage space into a training set and a test set and inputting them into the neural network prediction model for training, and predicting the maintenance time point of the target maintenance content of the target medical device based on the trained model;
[0007] S130, obtaining the status of the medical device at the maintenance time point predicted most recently, if the status is normal, forming a new maintenance record triplet with the target medical device flag, target maintenance content, and predicted maintenance time point for storage; if the status is an imminent status requiring urgent maintenance, correcting the predicted maintenance time point in advance to obtain the corrected maintenance time point, and combining the target medical device and the target maintenance content to form a new maintenance record triplet for storage; if the status is a problem status of equipment failure, resetting the maintenance time point based on the failure content to obtain the reset maintenance time point, and combining the target medical device and the target maintenance content to form a new maintenance record triplet for storage;
[0008] S140, performing medical equipment maintenance based on the medical equipment mark, maintenance content, and maintenance time point in the new maintenance record triplet, and predicting the next maintenance time point based on the historical records and the new maintenance record triplet.
[0009] In the S110, when the initial maintenance time point is a periodic time, a maintenance time point detection operation is performed, and the maintenance time point detection operation is specifically: setting an initial random weight w, and performing a modified detection on each maintenance time point of the week based on the weight; the weight w is preferably in the range of (-0.1, 0.1);
[0010] In the S110, each medical device corresponds to a medical device identifier, each medical device corresponds to multiple maintenance contents, and one maintenance operation corresponds to a maintenance time point and a maintenance content;
[0011] In the step S120, the neural network prediction model is preferably a BP neural network prediction model. In the step S120, the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training. The triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, specifically: (1) the triplets are preprocessed and normalized, divided into a training set and a test set in a ratio of 8:2, the regression length is set to an integer multiple of the average maintenance cycle, and the training set data is input into the BP neural network with the regression length; (2) the connection weights and moving thresholds are initialized, and the input and output values of the input layer, hidden layer neurons, and output layer are calculated; (3) it is determined whether the error between the predicted value and the actual value meets the prediction accuracy and whether the number of training times is reached. If any of the conditions is met, the training is terminated; otherwise, the feedback error is obtained and the connection weights and moving thresholds are corrected based on the feedback error; (5) the hidden layer and output layer outputs are recalculated and the process goes to step (3).
[0012] The advance correction in step S130 is specifically: subtract the predicted maintenance time point from the last maintenance time point, correct the difference proportionally, and obtain the corrected maintenance time point. The reset maintenance time point in step 130 is specifically: based on the fault content, set the next maintenance time according to the medical device product requirements, and the maintenance time is the initial value specified by the factory.
[0013] This application also seeks to protect a medical equipment management system based on big data, including:
[0014] A big data preprocessing module is used to obtain initial medical equipment maintenance records; the maintenance records include medical equipment identification, maintenance time point, maintenance content, and each maintenance record constitutes a triple (ID, T, Content) for storage;
[0015] A prediction model training module is used to establish a neural network prediction model based on maintenance record triplets; the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, and the maintenance time point of the target maintenance content of the target medical device is predicted based on the trained model;
[0016] A big data correction module is used to obtain the status of the medical equipment at the maintenance time point predicted most recently. If the status is normal, the target medical equipment mark, target maintenance content, and predicted maintenance time point are formed into a new maintenance record triple for storage; if the status is an imminent status that urgently requires maintenance, the predicted maintenance time point is corrected in advance to obtain the corrected maintenance time point, and a new maintenance record triple is formed in combination with the target medical equipment and the target maintenance content for storage; if the status is a problem status of equipment failure, the maintenance time point is reset based on the failure content to obtain a reset maintenance time point, and a new maintenance record triple is formed in combination with the target medical equipment and the target maintenance content for storage;
[0017] The big data update module is used to perform medical equipment maintenance based on the medical equipment logo, maintenance content, and maintenance time point in the new maintenance record triplet, and to predict the next maintenance time point based on historical records and the new maintenance record triplet.
[0018] Among them, it also includes a detection submodule, which is used to perform a maintenance time point detection operation when the initial maintenance time point is a periodic time. The maintenance time point detection operation is specifically: setting an initial random weight w, and performing a modified detection on each maintenance time point of the week based on the weight; the weight w is preferably in the range of (-0.1, 0.1);
[0019] Among them, each medical device corresponds to a unique medical device mark, each medical device corresponds to multiple maintenance contents, and one maintenance operation corresponds to a unique maintenance time point and unique maintenance content;
[0020] Among them, in the prediction model training module, the neural network prediction model is preferably a BP neural network prediction model. In the prediction model training module, the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training. The triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, specifically: (1) the triplets are preprocessed and normalized, divided into a training set and a test set in a ratio of 8:2, the regression length is set to an integer multiple of the average maintenance cycle, and the training set data is input into the BP neural network with the regression length; (2) the connection weights and moving thresholds are initialized, and the input and output values of the input layer, hidden layer neurons, and output layer are calculated; (3) it is determined whether the error between the predicted value and the actual value meets the prediction accuracy and whether the training times are reached. If any of the conditions are met, the training is terminated; otherwise, the feedback error is obtained and the connection weights and moving thresholds are corrected based on the feedback error; (5) the hidden layer and output layer outputs are recalculated, and the process goes to step (3).
[0021] Among them, the early correction in the big data correction module is specifically: the predicted maintenance time point is subtracted from the last maintenance time point, the difference is corrected proportionally, and the corrected maintenance time point is obtained. Among them, the reset maintenance time point in the big data correction module is specifically: based on the fault content, the next maintenance time is set according to the requirements of the medical device product, and the maintenance time is the initial value specified by the factory.
[0022] In the present invention, through the above scheme, (1) a triple maintenance record is set, the content of medical equipment management is used as the management basis, and the maintenance time is predicted based on both the equipment and the content as parameters, thereby realizing fine-grained management of medical equipment and improving the efficiency of medical equipment management; (2) based on the big data processing of the detection operation of the maintenance time point, a more reasonable maintenance time is predicted, and each prediction result is matched with the medical equipment status in the BP neural network prediction model, and the prediction result processing and correction methods of three situations are respectively adopted, so as to reasonably improve the prediction accuracy. In terms of the prediction results, the scheme of the present invention can accurately predict the next medical equipment management time point for each medical equipment, retain the advantages of problem-driven management, improve the differentiation of medical equipment, and further solve the hysteresis problem existing in the prior art due to the excessive frequency of periodic management and problem-driven management. BRIEF DESCRIPTION OF THE DRAWINGS
[0023] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art description. Obviously, the drawings described below are only used to illustrate some embodiments of the present application. For ordinary technicians in this field, without paying creative work, they can also obtain other technical features, connection relationships and even method steps not mentioned in the drawings based on these drawings.
[0024] Figure 1 This is a flow chart of a medical equipment management method based on big data provided by an embodiment of the present invention;
[0025] Figure 2 The present invention provides a structural diagram of the composition of a medical equipment management system based on big data. Specific embodiments
[0026] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.
[0027] The terms used in the embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit the present application. The singular forms of "a", "said", and "the" used in the embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two, but does not exclude the inclusion of at least one.
[0028] Depending on the context, the words "if" and "if" as used herein may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to monitoring." Similarly, depending on the context, the phrases "if it is determined" or "if monitoring (stated condition or event)" may be interpreted as "when determining" or "in response to determining" or "when monitoring (stated condition or event)" or "in response to monitoring (stated condition or event)."
[0029] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.
[0030] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0031] Embodiment 1: Figure 1 As shown, the present invention provides a medical equipment management method based on big data, comprising:
[0032] S110, obtain the initial medical equipment maintenance record; the maintenance record includes the medical equipment mark, maintenance time point, maintenance content, and each maintenance record constitutes a triple (ID, T, Content) storage; each medical device corresponds to a medical device mark, each medical device corresponds to multiple maintenance contents, and one maintenance operation corresponds to a maintenance time point and a maintenance content. Considering the difference in the management content of medical equipment, it is beneficial to improve the granularity of medical equipment management. For example, for nuclear magnetic resonance equipment, the nuclear magnetic resonance detection effect is at least closely related to the magnetic field strength and the irradiation position. The magnetic field strength usually fluctuates within a small range of 0.5T-2T, and the irradiation position is related to the angle of the robot arm. After a period of use, the fluctuation of the magnetic field strength is not large. Relatively speaking, the position and angle of the robot arm are greatly affected by the operator and the user, which will lead to a large difference in the management cycle of the maintenance of the magnetic field strength and the robot arm angle. Therefore, differentiated management is conducive to improving management efficiency.
[0033] Since the periodic maintenance operation time will cause the maintenance time predicted by the neural network to tend to a single time point, it is impossible to perform big data correction on the maintenance time. Therefore, it is considered to perform a maintenance time point detection operation when the initial maintenance time point is periodic. The maintenance time point detection operation is specifically: setting an initial random weight w, and performing a correction detection on each maintenance time point of the week based on the weight; the weight w preferably has a value range of (-0.1, 0.1). In the case of detecting and correcting the management time point, the medical equipment management personnel conduct a small range of detection around the recorded management time, and based on the feedback loop correction of the detection results, the management time differentiation of each medical device is improved. For example, equipment A and equipment B belong to the same type of medical equipment, and their initial management time points are the same. After the maintenance time detection operation, the management time point of equipment A is advanced by 10%, while the management time point of equipment B is moved back by 5%, thereby improving the management time differentiation of each medical device.
[0034] S120, establishing a neural network prediction model based on maintenance record triplets; dividing the triplets in the storage space into a training set and a test set and inputting them into the neural network prediction model for training, and predicting the maintenance time point of the target maintenance content of the target medical device based on the trained model; preferably, the neural network prediction model is preferably a BP neural network prediction model, and in the step S120, dividing the triplets in the storage space into a training set and a test set and inputting them into the neural network prediction model for training, specifically: (1) Perform preprocessing and normalization, divide the data into training set and test set in a ratio of 8:2, set the regression length to an integer multiple of the average maintenance period, and input the training set data into the BP neural network with the regression length; (2) initialize the connection weights and moving thresholds, calculate the input and output values of the input layer, hidden layer neurons, and output layer; (3) determine whether the error between the predicted value and the actual value meets the prediction accuracy and whether the training times have been reached. If any of the conditions are met, end the training; otherwise, obtain the feedback error and correct the connection weights and moving thresholds based on the feedback error; (5) recalculate the hidden layer and output layer outputs and enter step (3).
[0035] S130, obtaining the status of the medical device at the maintenance time point predicted most recently, if the status is normal, forming a new maintenance record triplet with the target medical device flag, target maintenance content, and predicted maintenance time point for storage; if the status is an imminent status requiring urgent maintenance, correcting the predicted maintenance time point in advance to obtain the corrected maintenance time point, and combining the target medical device and the target maintenance content to form a new maintenance record triplet for storage; if the status is a problem status of equipment failure, resetting the maintenance time point based on the failure content to obtain the reset maintenance time point, and combining the target medical device and the target maintenance content to form a new maintenance record triplet for storage;
[0036] Specifically, the above-mentioned advance correction is specifically: subtract the predicted maintenance time point from the last maintenance time point, correct the difference proportionally, and obtain the corrected maintenance time point. In the case that the equipment status urgently needs maintenance, the predicted maintenance time point is appropriately advanced to ensure the safety of the equipment. At the same time, the advanced maintenance time point has the effect of affecting the next predicted time point in the prediction model, thereby preventing the equipment from having more serious problems due to untimely maintenance.
[0037] The maintenance time point is reset in step 130 as follows: based on the fault content, the next maintenance time is set according to the product requirements of the medical device, and the maintenance time is the initial value specified by the factory. In the case of a more serious problem with the product, the predicted time point is set based on the specified value of the product requirements to ensure the safety of the equipment, and it also has the effect of affecting the next predicted time point.
[0038] More preferably, in order to reduce the maintenance frequency of medical equipment under the premise of normal operation of the medical equipment, when the medical equipment state is normal in step 130, the predicted maintenance time point is slightly offset backward, and the target medical equipment mark, target maintenance content, and the maintenance time point with the backward offset correction are formed into a new maintenance record triple for storage. The slight offset can be specifically to calculate the difference between the predicted maintenance time point and the last maintenance time point, and to amplify and correct the difference in proportion (such as 5%-10%) to obtain the maintenance time point with the slight offset correction.
[0039] S140, performing medical equipment maintenance based on the medical equipment mark, maintenance content, and maintenance time point in the new maintenance record triplet, and predicting the next maintenance time point based on the historical records and the new maintenance record triplet.
[0040] Example 2: See Figure 2 , the present application discloses a medical equipment management system based on big data, including:
[0041] A big data preprocessing module is used to obtain initial medical equipment maintenance records; the maintenance records include medical equipment identification, maintenance time point, maintenance content, and each maintenance record constitutes a triple (ID, T, Content) for storage;
[0042] A prediction model training module is used to establish a neural network prediction model based on maintenance record triplets; the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, and the maintenance time point of the target maintenance content of the target medical device is predicted based on the trained model;
[0043] A big data correction module is used to obtain the status of the medical equipment at the maintenance time point predicted most recently. If the status is normal, the target medical equipment mark, target maintenance content, and predicted maintenance time point are formed into a new maintenance record triple for storage; if the status is an imminent status that urgently requires maintenance, the predicted maintenance time point is corrected in advance to obtain the corrected maintenance time point, and a new maintenance record triple is formed in combination with the target medical equipment and the target maintenance content for storage; if the status is a problem status of equipment failure, the maintenance time point is reset based on the failure content to obtain a reset maintenance time point, and a new maintenance record triple is formed in combination with the target medical equipment and the target maintenance content for storage;
[0044] The big data update module is used to perform medical equipment maintenance based on the medical equipment logo, maintenance content, and maintenance time point in the new maintenance record triplet, and to predict the next maintenance time point based on historical records and the new maintenance record triplet.
[0045] Among them, it also includes a detection submodule, which is used to perform a maintenance time point detection operation when the initial maintenance time point is a periodic time. The maintenance time point detection operation is specifically: setting an initial random weight w, and performing a modified detection on each maintenance time point of the week based on the weight; the weight w is preferably in the range of (-0.1, 0.1);
[0046] Among them, each medical device corresponds to a unique medical device mark, each medical device corresponds to multiple maintenance contents, and one maintenance operation corresponds to a unique maintenance time point and unique maintenance content;
[0047] Among them, in the prediction model training module, the neural network prediction model is preferably a BP neural network prediction model. In the prediction model training module, the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training. The triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, specifically: (1) the triplets are preprocessed and normalized, divided into a training set and a test set in a ratio of 8:2, the regression length is set to an integer multiple of the average maintenance cycle, and the training set data is input into the BP neural network with the regression length; (2) the connection weights and moving thresholds are initialized, and the input and output values of the input layer, hidden layer neurons, and output layer are calculated; (3) it is determined whether the error between the predicted value and the actual value meets the prediction accuracy and whether the training times are reached. If any of the conditions are met, the training is terminated; otherwise, the feedback error is obtained and the connection weights and moving thresholds are corrected based on the feedback error; (5) the hidden layer and output layer outputs are recalculated, and the process goes to step (3).
[0048] Among them, the early correction in the big data correction module is specifically: the predicted maintenance time point is subtracted from the last maintenance time point, the difference is corrected proportionally, and the corrected maintenance time point is obtained. Among them, the reset maintenance time point in the big data correction module is specifically: based on the fault content, the next maintenance time is set according to the requirements of the medical device product, and the maintenance time is the initial value specified by the factory.
[0049] In order to reduce the maintenance frequency of medical equipment under the premise of normal operation of medical equipment, when the medical equipment is in normal state, the predicted maintenance time point is slightly offset backward, and the target medical equipment mark, target maintenance content, and the maintenance time point with the backward offset correction are formed into a new maintenance record triple for storage. The small amplitude can be specifically to calculate the difference between the predicted maintenance time point and the last maintenance time point, and to amplify and correct the difference in proportion (such as 5%-10%) to obtain the maintenance time point with the small amplitude backward offset correction.
[0050] In the present invention, through the above scheme, (1) a triple maintenance record is set, the content of medical equipment management is used as the management basis, and the maintenance time is predicted based on both the equipment and the content as parameters, thereby realizing fine-grained management of medical equipment and improving the efficiency of medical equipment management; (2) based on the big data processing of the detection operation of the maintenance time point, a more reasonable maintenance time is predicted, thereby realizing fine-grained management of medical equipment, and in the BP neural network prediction model, each prediction result is matched with the medical equipment status, and the prediction result processing and correction methods of three situations are respectively adopted, thereby reasonably improving the prediction accuracy. In terms of the prediction results, the scheme of the present invention can accurately predict the next medical equipment management time point, retaining the advantages of problem-driven management, and further solving the hysteresis problem existing in the prior art due to the excessive frequency of periodic management and problem-driven management.
[0051] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a product or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such a product or system. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the product or system including the elements.
[0052] The above description shows and describes several preferred embodiments of the present invention, but as mentioned above, it should be understood that the present invention is not limited to the form disclosed herein, and should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications and environments, and can be modified within the scope of the invention concept described herein through the above teachings or the technology or knowledge of the relevant field. Changes and variations made by those skilled in the art do not depart from the spirit and scope of the present invention, and should be within the scope of protection of the claims attached to the present invention.
Claims
1. A medical equipment management method based on big data, comprising: S110, obtaining an initial medical device maintenance record; the maintenance record includes a medical device identifier, a maintenance time point, and a maintenance content, and each maintenance record constitutes a triple (ID, T, Content) for storage; S120, establishing a neural network prediction model based on the maintenance record triples; dividing the triples in the storage space into a training set and a test set and inputting them into the neural network prediction model for training, and predicting the maintenance time point of the target maintenance content of the target medical device based on the trained model; S130, obtaining the medical device status at the most recent predicted maintenance time point, and if the status is normal, storing a new maintenance record triplet of the target medical device flag, target maintenance content, and predicted maintenance time point; If the status is an imminent status that urgently requires maintenance, the predicted maintenance time point is corrected in advance to obtain the corrected maintenance time point, and a new maintenance record triple is formed and stored in combination with the target medical equipment and the target maintenance content; If the status is a problem status of equipment failure, the maintenance time point is reset based on the fault content, and the reset maintenance time point is obtained, and a new maintenance record triple is formed and stored in combination with the target medical device and the target maintenance content; S140, performing medical equipment maintenance based on the medical equipment mark, maintenance content, and maintenance time point in the new maintenance record triplet, and predicting the next maintenance time point based on the historical records and the new maintenance record triplet; In the S110, when the initial maintenance time point is periodic, a maintenance time point detection operation is performed, and the maintenance time point detection operation is specifically: setting an initial random weight w, and performing a modified detection based on the weight for each maintenance time point of the week; the weight w has a value range of (-0.1, 0.1); The advance correction in step S130 is specifically as follows: subtract the predicted maintenance time point from the last maintenance time point, correct the difference proportionally, and obtain the corrected maintenance time point; the resetting of the maintenance time point in step 130 is specifically as follows: based on the fault content, set the next maintenance time in accordance with the medical device product requirements, and the maintenance time is the initial set value specified by the factory.
2. A medical equipment management method based on big data as described in claim 1, in S110, each medical equipment corresponds to a medical equipment mark, each medical equipment corresponds to multiple maintenance contents, and one maintenance operation corresponds to a maintenance time point and a maintenance content.
3. A medical equipment management method based on big data as described in claim 1, wherein in S120, the neural network prediction model is a BP neural network prediction model, and in step S120, the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training. The triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, specifically: (1) the triplets are preprocessed and normalized, divided into a training set and a test set in a ratio of 8:2, the regression length is set to an integer multiple of the average maintenance cycle, and the training set data is input into the BP neural network with the regression length; (2) the connection weights and moving thresholds are initialized, and the input and output values of the input layer, hidden layer neurons, and output layer are calculated; (3) it is determined whether the error between the predicted value and the actual value meets the prediction accuracy and whether the number of training times is reached. If any of the conditions is met, the training is terminated; otherwise, the feedback error is obtained and the connection weights and moving thresholds are corrected based on the feedback error; (5) the hidden layer and output layer outputs are recalculated and the process proceeds to step (3).
4. A medical equipment management system based on big data, comprising: A big data preprocessing module is used to obtain initial medical equipment maintenance records; the maintenance records include medical equipment identification, maintenance time point, maintenance content, and each maintenance record constitutes a triple (ID, T, Content) for storage; A prediction model training module is used to establish a neural network prediction model based on maintenance record triplets; the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, and the maintenance time point of the target maintenance content of the target medical device is predicted based on the trained model; A big data correction module is used to obtain the medical device status at the most recent predicted maintenance time point. If the status is normal, a new maintenance record triple is formed by the target medical device flag, target maintenance content, and predicted maintenance time point for storage; If the status is an imminent status that urgently requires maintenance, the predicted maintenance time point is corrected in advance to obtain the corrected maintenance time point, and a new maintenance record triple is formed and stored in combination with the target medical equipment and the target maintenance content; If the status is a problem status of equipment failure, the maintenance time point is reset based on the fault content, and the reset maintenance time point is obtained, and a new maintenance record triple is formed and stored in combination with the target medical device and the target maintenance content; Big data update module, used to perform medical equipment maintenance based on the medical equipment mark, maintenance content, and maintenance time point in the new maintenance record triplet, and to predict the next maintenance time point based on the historical records and the new maintenance record triplet; Also includes a detection submodule for performing a maintenance time point detection operation when the initial maintenance time point is periodic, the maintenance time point detection operation is specifically: setting an initial random weight w, and performing a weight-based correction detection on each maintenance time point of the week; the weight w value range is (-0.1, 0.1); The advance correction in the big data correction module is specifically as follows: subtracting the predicted maintenance time point from the last maintenance time point, correcting the difference proportionally, and obtaining a corrected maintenance time point; wherein, resetting the maintenance time point in the big data correction module is specifically as follows: based on the fault content, setting the next maintenance time in accordance with the medical device product requirements, and the maintenance time is the initial set value specified by the factory.
5. A medical equipment management system based on big data as described in claim 4, each medical equipment corresponds to a unique medical equipment mark, each medical equipment corresponds to multiple maintenance contents, and one maintenance operation corresponds to a unique maintenance time point and unique maintenance content.
6. A medical equipment management system based on big data as described in claim 4, wherein in the prediction model training module, the neural network prediction model is a BP neural network prediction model, and in the prediction model training module, the triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training. The triplets in the storage space are divided into a training set and a test set and input into the neural network prediction model for training, specifically: (1) the triplets are preprocessed and normalized, divided into a training set and a test set in a ratio of 8:2, the regression length is set to an integer multiple of the average maintenance cycle, and the training set data is input into the BP neural network with the regression length; (2) the connection weights and moving thresholds are initialized, and the input and output values of the input layer, hidden layer neurons, and output layer are calculated; (3) it is determined whether the error between the predicted value and the actual value meets the prediction accuracy and whether the training times are reached. If any of the conditions are met, the training is terminated; otherwise, the feedback error is obtained and the connection weights and moving thresholds are corrected based on the feedback error; (5) the hidden layer and output layer outputs are recalculated and the process proceeds to step (3).