Medical equipment operation monitoring system and method based on Internet of Things

Abstract

The invention discloses a medical equipment operation monitoring system and method based on the Internet of Things, and relates to the technical field of medical equipment monitoring, and the method comprises the steps: obtaining the historical operation fault record and feature operation range data of medical equipment, and evaluating the operation index in the feature operation range data, and the equipment matching degree between the operation index and the medical equipment; obtaining a target operation index of the medical equipment from the target operation data, and evaluating data relevance between an environment index in the historical operation environment record and the target operation index to obtain environment associated data; acquiring environment associated data, acquiring fault time data of the medical equipment, acquiring historical operation environment records and historical operation records of the medical equipment, and establishing an equipment fault prediction model of the medical equipment; and performing equipment fault prediction on the medical equipment by using the equipment fault prediction model in the current period, and monitoring the medical equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of medical equipment monitoring, and in particular to a medical equipment operation monitoring system and method based on the Internet of Things. Background Art

[0002] With the continuous development of science and technology, the application field of Internet of Things technology has become more and more extensive. In the operation monitoring of medical equipment, the use of Internet of Things technology includes but is not limited to the following benefits: 1. Real-time, Internet of Things technology can collect and transmit the operation data of medical equipment in real time, so as to promptly detect equipment failures or abnormalities and ensure the normal operation of the equipment; 2. Remote monitoring, through Internet of Things technology, medical institutions can remotely monitor medical equipment without frequent on-site inspections; 3. Data integration and analysis, Internet of Things technology can integrate data from different medical devices, providing a basis for comprehensive data analysis, making the monitoring of medical equipment more accurate.

[0003] Medical equipment is different from other types of equipment. Once a medical equipment fails, it will not only cause losses, but may also put the patient's life in danger. The current conventional method of monitoring medical equipment includes regular maintenance of medical equipment according to the equipment maintenance manual. However, this method not only requires a lot of manpower and time costs, but also fails to monitor the equipment in time and discover failures during equipment operation. Another method is to use Internet of Things technology to monitor medical equipment in real time, and determine the equipment operation status of the medical equipment based on the equipment operation parameters and preset ranges obtained in real time. However, in the actual use of medical equipment, different medical equipment will be affected by factors such as the equipment operating environment. Simply using the preset range to determine the equipment status is prone to inaccuracies, and there are currently few methods to predict the equipment status of medical equipment. This means that once a medical device fails during operation, it is very likely to cause great harm to the patient and seriously affect the patient's life safety. Summary of the invention

[0004] The purpose of the present invention is to provide a medical equipment operation monitoring system and method based on the Internet of Things to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solution: a method for monitoring the operation of medical equipment based on the Internet of Things, the method comprising: Step S100: Obtain historical operation fault records and characteristic operation range data of the medical device, evaluate the operation index in the characteristic operation range data and the degree of equipment compatibility between the medical device and obtain target operation data; Step S200: Obtain historical operating environment records and historical operating records of the medical device, obtain target operating indicators of the medical device from the target operating data, evaluate the data correlation between the environmental indicators in the historical operating environment records and the target operating indicators, and obtain environmental correlation data; Step S300: Acquire environment-related data, acquire failure time data of medical equipment, acquire historical operating environment records and historical operating records of medical equipment, and establish a device failure prediction model for medical equipment; Step S400: using the equipment failure prediction model in the current cycle to predict equipment failures of medical equipment and monitor the medical equipment.

[0006] Furthermore, step S100 includes: Step S101: Acquire historical operation fault records of medical equipment, and extract data corresponding to various operation indicators in the medical equipment from the historical operation fault records; Step S102: Acquire characteristic operating range data of the medical device, where the characteristic operating range data includes preset operating ranges of various operating indicators in the medical device; Step S103: when the data of a certain operation indicator in a certain historical operation fault record of the medical device are all within the operation range of the certain operation indicator, the certain historical operation fault record is recorded as a marked historical operation fault record of the certain operation indicator; Step S104: Evaluate the compatibility between each operating indicator in the characteristic operating range data and the medical device, wherein the operation indicator item a in the characteristic operating range data is evaluated for the compatibility between the medical device and the medical device. The specific process is as follows: Set a unit time, obtain the value of the a-th operation indicator from the historical operation fault record every unit time and collect them to obtain a data set of the a-th operation indicator in the marked historical operation fault record; Calculate the mark value B of the a-th operation indicator a =(B´ a,max +B´ a,min ) / 2, where B´ a,max It is expressed as the a-th operating index, the maximum value of the operating range in the characteristic operating range data, B´ a,min It is expressed as the minimum value of the operating range of the a-th operating index in the characteristic operating range data; Step S105: Calculate the label distance ratio of each element in the data set of the a-th operation index in the labeling history operation fault record, where the label distance ratio L of the c-th element in the data set of the a-th operation index in the labeling history operation fault record is a,c : , Among them, D a,c is the value of the a-th running indicator corresponding to the c-th element; When the marker distance ratio L a,c When the marking distance ratio of each element in the data set of the marking historical operation fault record is the maximum value, D a,c Recorded as the characteristic value of the a-th operation index in the marked historical operation fault record; Step S106: Calculate the ath operation index within the characteristic operation range data and the degree of compatibility between the medical device and the medical device a : , Among them, D i a is the characteristic value of the a-th operation indicator in the corresponding i-th marked historical operation fault record; n is the total number of marked historical operation fault records of the a-th operation indicator; When the equipment fits F a If the value is less than the preset equipment match threshold, the a-th operating indicator in the characteristic operating range data is determined to be inconsistent with the medical device, and the a-th operating indicator is marked as the target operating indicator of the medical device. The target operating indicators of each item in the medical device are obtained and aggregated to obtain the target operating data.

[0007] Further, step S200 includes: Step S201: monitoring the operating environment of the medical device in a historical period, obtaining a historical operating environment record of the medical device, and obtaining an average value of various environmental indicators of the medical device from the historical operating environment record; Step S202: Obtain various target operation indicators of the medical device from the target operation data, monitor the device operation status of the medical device in the historical period, and obtain the historical operation records of the medical device, wherein the total number of historical operation records in the medical device is the same as the total number of historical operation environment records, and extract the average value of each target operation indicator from the historical operation records; Step S203: obtaining the average values ​​of the target operating indicators in each historical operating record of the medical device, and sorting them from small to large, and recording the order corresponding to the target operating indicators in each historical operating record as the characteristic rank of the target operating indicators in each historical operating record; Step S204: Obtain the feature ranks of the target operation indicators in each historical operation record and aggregate them to obtain a feature rank set G = {g1, g2, ..., g m}, where g1, g2, ..., g mThey are the characteristic ranks of the target operation indicators in the 1st, 2nd, ..., mth historical operation records of the medical device respectively; Step S205: Evaluate the data correlation between each environmental indicator in the historical operation environment record and each target operation indicator of the medical device. Among them, when evaluating the data correlation between the hth environmental indicator in the historical operation environment record and the kth target operation indicator of the medical device, the specific process is as follows: Obtain the set of characteristic ranks S of the hth environmental indicator h , obtain the set of characteristic ranks G of the kth target operation indicator k , calculate the data correlation value R between the hth environmental indicator and the kth target operation indicator h,k : , where, g k,z is the characteristic rank corresponding to the zth element in the set of characteristic ranks G k ; s h,z is the characteristic rank corresponding to the zth element in the set of characteristic ranks S h ; g' k is the average value of the characteristic ranks corresponding to each element in the set of characteristic ranks G k ; s' h is the average value of the characteristic ranks corresponding to each element in the set of characteristic ranks S h ; m is the total number of each element in the set of characteristic ranks G k ; Set the first correlation coefficient R'1 and the second correlation coefficient R'2, where R'1 > 0 > R'2. When the data correlation value R h,k > R'1, it is determined that there is a positive correlation between the hth environmental indicator and the kth target operation indicator. When R h,k < R'2, it is determined that there is a negative correlation between the hth environmental indicator and the kth target operation indicator. When R'2 ≤ R h,k ≤ R'1, it is determined that there is no correlation between the hth environmental indicator and the kth target operation indicator; Step S206: Obtain several environmental indicators that are correlated with the target operation indicators of the medical device from each environmental indicator, and mark and collect them to obtain the environmental correlation data of the medical device.

[0008] Furthermore, step S300 includes: Step S301: Obtain the fault time data of the medical device. The fault time data includes the historical period to which the fault of the medical device belongs, and obtain the historical operation environment record and historical operation record of the medical device in each historical period; Step S302: Obtain environment-related data, and obtain several marked environmental indicators from the environment-related data. Take the average value of the operation indicators in the historical operation records within the historical period and the average value of the several marked environmental indicators in the historical operation environment records as input data, and take the probability of the medical device malfunctioning as output data; Step S303: Establish a device failure prediction model for the medical device. The specific establishment process is as follows: According to a preset ratio, divide the input data and output data of the medical device in each historical period into a training set and a test set. Obtain the average value of the operation indicators in the historical operation records within the samples in the training set and the average value of the several marked environmental indicators in the historical operation environment records, perform normalization processing and aggregation to obtain the feature input vector X of the samples in the training set; Obtain the preset weight vector W and bias term b in the device failure prediction model; Step S304: Obtain the objective function f(W,b) in the device failure prediction model, set the penalty parameter β in the objective function, and use the training set to minimize the objective function. The specific formula is: , where the constraint condition for minimizing the objective function f(W,b) is: y α ·(W·X α +b)≥1-ξ α ,ξ α ≥0; α is the total number of training samples in the training set; X α is the feature input vector of the α-th sample in the training set; ξ α is the slack variable of the α-th sample in the objective function; y α is the true category of the α-th sample. When the medical device in the α-th sample malfunctions, y α is +k, otherwise, y α is -k; Step S305: Use the test set to calculate the accuracy rate P of the device failure prediction model accuracy : , where TP is the total number of samples in the test set where the medical device is correctly predicted to malfunction; TN is the total number of samples in the test set where the medical device is correctly predicted not to malfunction; FP is the total number of samples in the test set where the medical device is incorrectly predicted to malfunction; FN is the total number of samples in the test set where the medical device is incorrectly predicted not to malfunction; When the accuracy rate P accuracyIf it is greater than the preset accuracy threshold, it is determined that the device fault prediction model is established. Otherwise, the device fault prediction model continues to be trained until the accuracy P accuracy is greater than the preset accuracy threshold.

[0009] Further, step S400 includes: Step S401: Collect the operating state of the medical device within the first unit time period of the current cycle, and input the obtained data into the device fault prediction model. Use the device fault prediction model to predict the medical device; Step S402: When the probability of the medical device failing output by the device fault prediction model is greater than the preset probability threshold, it is determined that the medical device has a device fault risk within the current cycle. Dispatch relevant personnel to repair the medical device and perform intelligent monitoring on the medical device; The above steps collect the operating state of the medical device within the first unit time period of the current cycle, input the obtained data into the device fault prediction model, and perform device fault prediction on the medical device, so as to judge when the medical device will fail. It is to predict the device fault situation within the current cycle based on the operating state of the medical device within the unit time period, and preparations can be made in advance. Because once the medical device fails during the treatment process, it will cause serious consequences. Knowing the device state of the medical device in advance not only helps to process the device, greatly reduces the damage caused by device faults to the device, but also ensures the life safety of patients.

[0010] To better implement the above method, a medical device operation monitoring system based on the Internet of Things is also proposed. The system includes a device fitness evaluation module, an environment-related data module, a model establishment module, and a device monitoring module; The device fitness evaluation module is used to evaluate the device fitness degree between the operating indicators within the characteristic operation range data and the medical device to obtain target operation data; The environment-related data module is used to evaluate the data correlation between the environment indicators in the historical operation environment record and the target operation indicators to obtain environment-related data; The model establishment module is used to obtain the historical operation environment record and historical operation record of the medical device and establish a device fault prediction model for the medical device; The device monitoring module is used to perform device fault prediction on the medical device and monitor the medical device.

[0011] Further, the device fitness evaluation module includes a device fitness degree unit and a target operation data unit; The device fitness degree unit is used to calculate the device fitness degree between each operation index within the feature operation range data and the medical device; The target operation data unit is used to evaluate the device fitness degree between the operation index within the feature operation range data and the medical device according to the device fitness degree, so as to obtain the target operation data.

[0012] Furthermore, the environment association data module includes a data association value unit and an association evaluation unit; The data association value unit is used to calculate the data association value between the environment index in the historical operation environment record and the target operation index in the medical device; The association evaluation unit is used to evaluate the data correlation between the environment index in the historical operation environment record and the target operation index according to the data association value, so as to obtain the environment association data.

[0013] Furthermore, the model establishment module includes a data acquisition unit and a model establishment unit; The data acquisition unit is used to acquire the fault time data of the medical device, and acquire the historical operation environment record and historical operation record in the medical device; The model establishment unit is used to establish the device fault prediction model of the medical device.

[0014] Furthermore, the device monitoring module includes a device monitoring unit; The device monitoring unit is used to use the device fault prediction model in the medical device to predict the medical device. When the probability of the medical device having a fault output by the device fault prediction model is greater than the preset probability threshold, it is determined that the medical device has a device fault risk during the current period, and relevant personnel are dispatched to repair the medical device, and the medical device is intelligently monitored.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention realizes the intelligent monitoring of the operation of medical devices. By evaluating the device fitness degree between the operation index within the feature operation range data and the medical device, the operation index that often appears inaccurate in the device fault determination is found and recorded as the target operation index. Then, the environment index associated with the target operation index is found, and the device fault prediction model of the medical device is established, making the determination of the device state of the medical device more accurate and reliable. It also avoids affecting the treatment of patients due to misjudging the device fault, greatly improving the treatment success rate of patients and enabling the medical device to be repaired in a timely manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is the method flow chart of a method for monitoring the operation of a medical device based on the Internet of Things according to the present invention; Figure 2It is a schematic diagram of the modules of a medical device operation monitoring system based on the Internet of Things according to the present invention. Specific embodiments

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] Embodiment: As Figure 1-Figure 2 shown, the present invention provides a technical solution, a method for monitoring the operation of medical devices based on the Internet of Things, and the method includes: Step S100: Obtain the historical operation fault records and characteristic operation range data of the medical device, evaluate the degree of fit between the operation indicators in the characteristic operation range data and the medical device, and obtain the target operation data; Among them, step S100 includes: Step S101: Obtain the historical operation fault records of the medical device, and extract the data corresponding to each operation indicator in the medical device from the historical operation fault records; For example, each operation indicator includes operating voltage, operating current, etc.; Step S102: Obtain the characteristic operation range data of the medical device, and the characteristic operation range data includes the preset operation ranges of each operation indicator in the medical device; Step S103: When the data of a certain operation indicator in a certain historical operation fault record of the medical device are all within the operation range of a certain operation indicator, record the certain historical operation fault record as the marked historical operation fault record of the certain operation indicator; Step S104: Evaluate the degree of fit between each operation indicator in the characteristic operation range data and the medical device. Among them, the specific process of evaluating the degree of fit between the a-th operation indicator in the characteristic operation range data and the medical device is: Set the unit time period, and obtain the value of the a-th operation indicator from the historical operation fault records every unit time period and gather them to obtain the data set of the a-th operation indicator in the marked historical operation fault record; Calculate the marked value B of the a-th operation indicator a =(B´ a,max +B´ a,min ) / 2, where B´ a,max represents the maximum value of the operation range of the a-th operation indicator in the characteristic operation range data, and B´ a,minIt is expressed as the minimum value of the operating range of the a-th operating index in the characteristic operating range data; For example, B´ 2,max Expressed as 100, B´ 2,min Expressed as 20, calculate the mark value of the second operating indicator B2=(20+100) / 2=60; Step S105: Calculate the label distance ratio of each element in the data set of the a-th operation index in the labeling history operation fault record, where the label distance ratio L of the c-th element in the data set of the a-th operation index in the labeling history operation fault record is a,c : , Among them, D a,c is the value of the a-th running indicator corresponding to the c-th element; When the marker distance ratio L a,c When the marking distance ratio of each element in the data set of the marking historical operation fault record is the maximum value, D a,c Recorded as the characteristic value of the a-th operation index in the marked historical operation fault record; Step S106: Calculate the ath operation index within the characteristic operation range data and the degree of compatibility between the medical device and the medical device a : , Among them, D i a is the characteristic value of the a-th operation indicator in the corresponding i-th marked historical operation fault record; n is the total number of marked historical operation fault records of the a-th operation indicator; When the equipment fits F a is less than a preset device matching threshold, determining that the ath operation indicator in the characteristic operation range data is not matching with the medical device, marking the ath operation indicator as the target operation indicator of the medical device, obtaining and aggregating the target operation indicators of the medical device, and obtaining the target operation data; Step S200: Obtain historical operating environment records and historical operating records of the medical device, obtain target operating indicators of the medical device from the target operating data, evaluate the data correlation between the environmental indicators in the historical operating environment records and the target operating indicators, and obtain environmental correlation data; Wherein, step S200 includes: Step S201: monitoring the operating environment of the medical device in a historical period, obtaining a historical operating environment record of the medical device, and obtaining an average value of various environmental indicators of the medical device from the historical operating environment record; For example, various environmental indicators include equipment operating temperature, equipment operating humidity, etc.; Step S202: Obtain various target operation indicators of the medical device from the target operation data, monitor the device operation status of the medical device in the historical period, and obtain the historical operation records of the medical device, wherein the total number of historical operation records in the medical device is the same as the total number of historical operation environment records, and extract the average value of each target operation indicator from the historical operation records; Step S203: obtaining the average values ​​of the target operating indicators in each historical operating record of the medical device, and sorting them from small to large, and recording the order corresponding to the target operating indicators in each historical operating record as the characteristic rank of the target operating indicators in each historical operating record; Step S204: Obtain the feature ranks of the target operation indicators in each historical operation record and aggregate them to obtain a feature rank set G = {g1, g2, ..., g m}, where g1, g2, ..., g m are the characteristic ranks of the target operation indicators in the 1st, 2nd, ..., mth historical operation records of the medical equipment respectively; Step S205: Evaluate the data correlation between each environmental indicator in the historical operating environment record and each target operating indicator of the medical device, wherein the data correlation between the hth environmental indicator in the historical operating environment record and the kth target operating indicator of the medical device is evaluated, and the specific process is as follows: Get the characteristic rank set S of the hth environmental indicator h , obtain the characteristic rank set G of the kth target operation index k , calculate the data correlation value R of the hth environmental indicator and the kth target operation indicator h,k : , Among them, g k,z is the feature rank set G k The feature rank corresponding to the zth element in s h,z is the feature rank set S h The feature rank corresponding to the zth element in g´ k is the feature rank set G k The average value of the feature rank corresponding to each element in ; s´ h is the feature rank set S h The average value of the feature rank corresponding to each element in; m is the feature rank set G k The total number of elements in ; Set the first correlation coefficient R'1 and the second correlation coefficient R'2, where R'1>0>R'2. h,k>R´1, it is determined that there is a positive correlation between the h-th environmental indicator and the k-th target operation indicator. When R h,k <R´2, it is determined that there is a negative correlation between the h-th environmental indicator and the k-th target operation indicator. When R´2 ≤ R h,k ≤ R´1, it is determined that there is no correlation between the h-th environmental indicator and the k-th target operation indicator; Step S206: From each environmental indicator, obtain several environmental indicators that are correlated with the target operation indicators of the medical device, and perform marking and aggregation to obtain the environmental correlation data of the medical device; Step S300: Obtain the environmental correlation data, obtain the fault time data of the medical device, obtain the historical operating environment record and historical operating record of the medical device, and establish the device fault prediction model of the medical device; Among them, step S300 includes: Step S301: Obtain the fault time data of the medical device. The fault time data includes the historical cycle to which the fault of the medical device belongs, and obtain the historical operating environment record and historical operating record of the medical device in each historical cycle; Step S302: Obtain the environmental correlation data, obtain several marked environmental indicators from the environmental correlation data, use the average value of the operation indicators in the historical operating record within the historical cycle and the average value of the several marked environmental indicators in the historical operating environment record as input data, and use the probability of the medical device having a fault as output data; For example, when the medical device has a fault within the historical cycle, the probability is 1, and when the medical device does not have a fault within the historical cycle, the probability is 0; Step S303: Establish the device fault prediction model of the medical device. The specific establishment process is as follows: According to a preset ratio, divide the input data and output data of the medical device in each historical cycle into a training set and a test set. Obtain the average value of the operation indicators in the historical operating record within the samples in the training set and the average value of the several marked environmental indicators in the historical operating environment record, perform normalization processing and aggregation to obtain the feature input vector X of the samples in the training set; Obtain the preset weight vector W and bias term b in the device fault prediction model; Step S304: Obtain the objective function f(W, b) in the device fault prediction model, set the penalty parameter β in the objective function, and use the training set to minimize the objective function. The specific formula is: , Among them, the constraint condition for minimizing the objective function f(W, b) is: y α ·(W·Xα +b)≥1-ξ α ,ξ α ≥0; α is the total number of training samples in the training set; X α is the feature input vector of the αth sample in the training set; ξ α is the slack variable of the αth sample in the objective function; y α is the true category of the αth sample. When the medical device in the αth sample fails, y α is +k, otherwise, y α is -k; For example, k is usually set to 1; Step S305: Using the test set, calculate the accuracy P in the equipment failure prediction model accuracy : , Among them, TP is the total number of samples in the test set that are correctly predicted to be faulty; TN is the total number of samples in the test set that are correctly predicted to be non-faulty; FP is the total number of samples in the test set that are incorrectly predicted to be faulty; FN is the total number of samples in the test set that are incorrectly predicted to be non-faulty; For example, TP is 30; TN is 40; FP is 10; FN is 20; Calculate the accuracy P in the equipment failure prediction model accuracy : , When the accuracy P accuracy If the accuracy is greater than the preset threshold, the equipment fault prediction model is considered to be established. Otherwise, the equipment fault prediction model continues to be trained until the accuracy is P accuracy Greater than the preset accuracy threshold; Step S400: using the equipment failure prediction model in the current cycle to predict equipment failures of medical equipment and monitor the medical equipment; Wherein, step S400 includes: Step S401: collecting the equipment operation status of the medical equipment within the first unit time of the current cycle, inputting the data into the equipment failure prediction model, and using the equipment failure prediction model to predict the medical equipment; Step S402: When the probability of medical equipment failure output by the equipment failure prediction model is greater than a preset probability threshold, it is determined that the medical equipment has a risk of equipment failure in the current cycle, and relevant personnel are dispatched to inspect and repair the medical equipment, and the medical equipment is intelligently monitored; In order to better implement the above method, a medical device operation monitoring system based on the Internet of Things is also proposed, the system includes a device fit evaluation module, an environment-related data module, a model building module, and a device monitoring module; The equipment compatibility assessment module is used to assess the compatibility between the operating indicators within the characteristic operating range data and the medical equipment to obtain the target operating data; The environment-related data module is used to evaluate the data correlation between the environmental indicators in the historical operating environment records and the target operating indicators to obtain the environment-related data; A model building module is used to obtain historical operating environment records and historical operating records of medical equipment and establish an equipment failure prediction model for medical equipment; Equipment monitoring module, used to predict equipment failures of medical equipment and monitor medical equipment; Among them, the equipment fit assessment module includes an equipment fit degree unit and a target operation data unit; The device compatibility unit is used to calculate the compatibility between various operating indicators within the characteristic operating range data and the medical device; The target operation data unit is used to evaluate the operation index within the characteristic operation range data and the degree of equipment fit between the medical equipment according to the equipment fit, and obtain the target operation data; Among them, the environment-related data module includes a data-related value unit and a related evaluation unit; A data association value unit, used to calculate the data association value of the environmental index in the historical operating environment record and the target operating index in the medical device; A correlation evaluation unit, used to evaluate the data correlation between the environmental indicators in the historical operating environment records and the target operating indicators according to the data correlation value, and obtain environmental correlation data; Wherein, the model building module includes a data acquisition unit and a model building unit; A data acquisition unit, used to acquire the failure time data of the medical device, and acquire the historical operating environment records and historical operating records in the medical device; A model building unit, used for building a device failure prediction model for medical equipment; Wherein, the equipment monitoring module includes an equipment monitoring unit; The equipment monitoring unit is used to use the equipment failure prediction model in the medical equipment to predict the medical equipment. When the probability of medical equipment failure output by the equipment failure prediction model is greater than the preset probability threshold, it is determined that the medical equipment has the risk of equipment failure in the current cycle, and relevant personnel are dispatched to inspect and repair the medical equipment, and the medical equipment is intelligently monitored.

[0019] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for monitoring the operation of medical equipment based on the Internet of Things, characterized in that: The method comprises: Step S100: obtaining historical operation fault records and characteristic operation range data of the medical device, evaluating the operation index in the characteristic operation range data and the degree of compatibility between the medical device and the medical device, and obtaining target operation data; Step S200: obtaining historical operating environment records and historical operating records of the medical device, obtaining target operating indicators of the medical device from the target operating data, evaluating the data correlation between the environmental indicators in the historical operating environment records and the target operating indicators, and obtaining environment correlation data; Step S300: Acquire the environment-related data, acquire the failure time data of the medical device, acquire the historical operating environment records and historical operating records of the medical device, and establish a device failure prediction model for the medical device; Step S400: using the equipment failure prediction model in the current cycle to predict equipment failure of the medical equipment and monitor the medical equipment.

2. According to the method for monitoring the operation of medical equipment based on the Internet of Things in claim 1, it is characterized in that: The step S100 includes: Step S101: Acquire historical operation fault records of the medical device, and extract data corresponding to various operation indicators of the medical device from the historical operation fault records; Step S102: Acquire characteristic operating range data of the medical device, wherein the characteristic operating range data includes a preset operating range of each operating indicator in the medical device; Step S103: when the data of a certain operation indicator in a certain historical operation fault record of the medical device are all within the operation range of the certain operation indicator, the certain historical operation fault record is recorded as a marked historical operation fault record of the certain operation indicator; Step S104: evaluating the compatibility of each operating indicator in the characteristic operating range data with the medical device, wherein the specific process of evaluating the compatibility of the item a of the operating indicator in the characteristic operating range data with the medical device is as follows: Set a unit time, obtain the value of the a-th operation indicator from the historical operation fault record every unit time and collect them to obtain a data set of the a-th operation indicator in the marked historical operation fault record; Calculate the mark value B of the a-th operation indicator a =(B´ a,max +B´ a,min ) / 2, where B´ a,max It is represented as the a-th operation index, the maximum value of the operation range in the characteristic operation range data, B´ a,min The a-th operation index is represented by the minimum value of the operation range in the characteristic operation range data; Step S105: Calculate the label distance ratio of each element in the data set of the labeled historical operation fault record of the a-th operation index, wherein the label distance ratio L of the c-th element in the data set of the labeled historical operation fault record of the a-th operation index is a,c : ， Among them, D a,c is the value of the a-th running indicator corresponding to the c-th element; When the marker distance ratio L a,c When the marking distance ratio of each element in the data set in the marking historical operation fault record is the maximum value, the D a,c Recorded as the characteristic value of the item a of the operation index in the marked historical operation fault record; Step S106: Calculate the ath operation index within the characteristic operation range data and the degree of compatibility between the medical device and the medical device. a : ， Among them, D i a is the characteristic value of the a-th operation indicator in the corresponding i-th marked historical operation fault record; n is the total number of marked historical operation fault records of the a-th operation indicator; When the equipment meets the degree F a If the value is less than a preset device match threshold, it is determined that the a-th operating indicator in the characteristic operating range data is not consistent with the medical device, and the a-th operating indicator is marked as the target operating indicator of the medical device. The target operating indicators of each item in the medical device are obtained and collected to obtain the target operating data.

3. The method for monitoring the operation of medical equipment based on the Internet of Things according to claim 2, characterized in that: The step S200 includes: Step S201: monitoring the operating environment of the medical device in a historical period, obtaining a historical operating environment record of the medical device, and obtaining an average value of various environmental indicators of the medical device from the historical operating environment record; Step S202: acquiring various target operation indicators of the medical device from the target operation data, monitoring the device operation status of the medical device in a historical period, and obtaining historical operation records of the medical device, wherein the total number of historical operation records in the medical device is the same as the total number of historical operation environment records, and extracting the average value of each target operation indicator from the historical operation records; Step S203: obtaining the average values ​​of the target operating indicators in each historical operating record of the medical device, and sorting them from small to large, and recording the order corresponding to the target operating indicators in each historical operating record as the characteristic rank of the target operating indicators in each historical operating record; Step S204: Obtain and aggregate the feature ranks of the target operation indicators in each of the historical operation records to obtain a feature rank set G = {g1, g2, ..., g m }, where g1, g2, ..., g m are the characteristic ranks of the target operating indicators in the 1st, 2nd, ..., mth historical operating records of the medical device, respectively; Step S205: Evaluate the data correlation between each environmental indicator in the historical operating environment record and each target operating indicator of the medical device, wherein the data correlation between the hth environmental indicator in the historical operating environment record and the kth target operating indicator of the medical device is evaluated, and the specific process is as follows: Get the characteristic rank set S of the h-th environmental indicator h , obtain the characteristic rank set G of the k-th target operation index k , calculate the data correlation value R of the hth environmental indicator and the kth target operation indicator h,k : ， Among them, g k,z is the feature rank set G k The feature rank corresponding to the zth element in s h,z is the feature rank set S h The feature rank corresponding to the zth element in g´ k is the feature rank set G k The average value of the feature rank corresponding to each element in ; s´ h is the feature rank set S h The average value of the feature rank corresponding to each element in; m is the feature rank set G k The total number of elements in ; Set the first correlation coefficient \(R'_1\) and the second correlation coefficient \(R'_2\), where \(R'_1>0>R'_2\). When the data correlation value \(R\) h,k \(>R'_1\), it is determined that there is a positive correlation between the \(h\)-th environmental indicator and the \(k\)-th target operation indicator. When \(R\) h,k \(<R'_2\), it is determined that there is a negative correlation between the \(h\)-th environmental indicator and the \(k\)-th target operation indicator. When \(R'_2\leq R\) h,k \(\leq R'_1\), it is determined that there is no correlation between the \(h\)-th environmental indicator and the \(k\)-th target operation indicator; Step S206: From the various environmental indicators, a number of environmental indicators that are associated with the target operating indicators of the medical device are obtained, and they are marked and collected to obtain environmental related data of the medical device.

4. The method for monitoring the operation of medical equipment based on the Internet of Things according to claim 3, characterized in that: The step S300 includes: Step S301: Acquire the failure time data of the medical device, the failure time data including the historical period to which the failure of the medical device occurs, and acquire the historical operating environment record and historical operating record of the medical device in each historical period; Step S302: Obtain the environment-related data, obtain a number of marked environmental indicators from the environment-related data, use the average value of the operating indicators in the historical operating records within the historical period and the average value of the number of marked environmental indicators in the historical operating environment records as input data, and use the probability of failure of the medical device as output data; Step S303: Establishing a device failure prediction model for the medical device, the specific establishment process is as follows: According to a preset ratio, the input data and output data of the medical device in each historical period are divided into a training set and a test set, and the average values ​​of the operation indicators in the historical operation records of the samples in the training set and the average values ​​of several marked environmental indicators in the historical operation environment records are obtained, and normalized and aggregated to obtain the feature input vector X of the samples in the training set; Obtaining a weight vector W and a bias term b preset in the equipment fault prediction model; Step S304: Obtain the objective function f(W, b) in the equipment fault prediction model, set the penalty parameter β in the objective function, and use the training set to minimize the objective function. The specific formula is: ， Among them, the constraint condition for minimizing the objective function f(W,b) is: α ·(W·X α +b)≥1-ξ α ,ξ α ≥0; α is the total number of training samples in the training set; X α is the feature input vector of the αth sample in the training set; ξ α is the slack variable of the αth sample in the objective function; y α is the true category of the αth sample. When the medical device in the αth sample fails, y α is +k, otherwise, y α is -k; Step S305: Using the test set, calculate the accuracy P in the equipment failure prediction model accuracy : ， Wherein, TP is the total number of samples in the test set that are correctly predicted to be faulty; TN is the total number of samples in the test set that are correctly predicted to be non-faulty; FP is the total number of samples in the test set that are incorrectly predicted to be faulty; FN is the total number of samples in the test set that are incorrectly predicted to be non-faulty; When the accuracy P accuracy If the accuracy rate is greater than a preset threshold, it is determined that the equipment fault prediction model is established. Otherwise, the equipment fault prediction model continues to be trained until the accuracy rate P accuracy Greater than the preset accuracy threshold.

5. The method for monitoring the operation of medical equipment based on the Internet of Things according to claim 4, characterized in that: The step S400 includes: Step S401: collecting the equipment operation status of the medical equipment within the first unit time of the current cycle, inputting the data into the equipment failure prediction model, and using the equipment failure prediction model to predict the medical equipment; Step S402: When the probability of failure of the medical device output by the device failure prediction model is greater than a preset probability threshold, it is determined that the medical device has a risk of equipment failure in the current cycle, and relevant personnel are dispatched to inspect and repair the medical device, and the medical device is intelligently monitored.

6. A medical device operation monitoring system based on the Internet of Things, used to execute a medical device operation monitoring method based on the Internet of Things as described in any one of claims 1 to 5, characterized in that: The system includes an equipment fit assessment module, an environment-related data module, a model building module, and an equipment monitoring module; The device fit assessment module is used to assess the degree of fit between the operation index within the characteristic operation range data and the medical device to obtain target operation data; The environment-related data module is used to evaluate the data correlation between the environment indicators in the historical operating environment records and the target operating indicators to obtain environment-related data; The model building module is used to obtain the historical operating environment records and historical operating records of the medical device and establish an equipment failure prediction model for the medical device; The equipment monitoring module is used to predict equipment failure of the medical equipment and monitor the medical equipment.

7. The medical equipment operation monitoring system based on the Internet of Things according to claim 6 is characterized in that: The device fit assessment module includes a device fit degree unit and a target operation data unit; The device compatibility unit is used to calculate the device compatibility between each operating index within the characteristic operating range data and the medical device; The target operation data unit is used to evaluate the operation index within the characteristic operation range data and the degree of equipment compatibility between the medical device and obtain target operation data according to the degree of equipment compatibility.

8. The medical equipment operation monitoring system based on the Internet of Things according to claim 6 is characterized in that: The environment-related data module includes a data-related value unit and a related evaluation unit; The data association value unit is used to calculate the data association value of the environmental index in the historical operating environment record and the target operating index in the medical device; The association evaluation unit is used to evaluate the data association between the environmental indicators in the historical operating environment records and the target operating indicators according to the data association value to obtain environmental association data.

9. The medical equipment operation monitoring system based on the Internet of Things according to claim 6 is characterized in that: The model building module includes a data acquisition unit and a model building unit; The data acquisition unit is used to acquire the failure time data of the medical device, and acquire the historical operating environment records and historical operating records of the medical device; The model building unit is used to build a device failure prediction model for the medical device.

10. The medical equipment operation monitoring system based on the Internet of Things according to claim 6, characterized in that: The device monitoring module includes a device monitoring unit; The equipment monitoring unit is used to use the equipment failure prediction model in the medical equipment to predict the medical equipment. When the probability of failure of the medical equipment output by the equipment failure prediction model is greater than a preset probability threshold, it is determined that the medical equipment has a risk of equipment failure in the current cycle, and relevant personnel are dispatched to inspect and repair the medical equipment, thereby performing intelligent monitoring of the medical equipment.

Images