A medical device operation and maintenance management system based on information mining and analysis

Abstract

The present invention relates to the field of operation and maintenance management technology, and specifically to a medical equipment operation and maintenance management system based on information mining and analysis. In the present invention, by introducing accurate screening of data abnormality features, the evaluation of the equipment operating status is made more refined. Based on the mean offset of the blood oxygen sensor measurement data, abnormal data can be effectively identified, and combined with the distribution trend of the measurement error, the accuracy of the equipment status evaluation can be ensured. The impact of environmental factors on equipment performance has been deeply explored. By using the correlation calculation of ambient temperature changes and power consumption offsets, the dynamic change trend of equipment performance under different environmental conditions can be identified, and the sensitivity to the influence of external factors can be improved. The evaluation method of the operating status is more comprehensive. By measuring the collaborative analysis of the frequency fluctuation amplitude and the signal response time, the subtle change trend of the equipment operating status can be accurately portrayed, and the accuracy of fault prediction can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of operation and maintenance management, and particularly to a medical device operation and maintenance management system based on information mining and analysis. Background Art

[0002] The technical field of operation and maintenance management includes the maintenance, monitoring, and optimization management of devices, systems, and infrastructure to ensure their stable operation throughout their life cycle. This technical field involves aspects such as device status monitoring, fault diagnosis, maintenance strategy optimization, and maintenance plan formulation. With the development of information technology, automation means have been gradually introduced into operation and maintenance management to achieve real-time monitoring and remote control of device status through data collection, analysis, and intelligent prediction.

[0003] Among them, a medical device operation and maintenance management system refers to a system for data monitoring, status evaluation, and maintenance management of the operation and maintenance of medical devices. The system obtains the operation parameters of medical devices through data collection devices, including working status, usage duration, temperature, humidity, and performance indicators of key components, and evaluates the device status in combination with preset fault judgment criteria. The system adopts an analysis method based on device operation data to identify possible fault hidden dangers and formulates a maintenance plan in combination with historical data and maintenance strategies.

[0004] The evaluation of device operation status mainly relies on a single operation parameter, which cannot accurately depict the dynamic changes of the device status, resulting in some potential faults being difficult to detect in a timely manner. The screening method for abnormal data is relatively rough, only judging based on fixed thresholds, lacking in-depth analysis of the data error distribution, and may ignore some key abnormal points. The influence of environmental factors is not fully considered, and the performance changes of devices in different working environments are difficult to quantify, resulting in devices being likely to malfunction under specific environmental conditions and being difficult to give early warnings. The evaluation of the operation status lacks the correlation analysis of multi-dimensional parameters, and indicators such as signal response time and measurement frequency do not form effective coordination, unable to capture the subtle changes in the device status, reducing the detection ability of device operation anomalies. The identification method for device degradation status is relatively single, mainly relying on the changes of a single performance indicator, ignoring the cumulative trend of measurement errors and the degradation of signal response time, resulting in insufficient accuracy in evaluating the device degradation status. The operation and maintenance strategy lacks an intelligent optimization mechanism, with a fixed maintenance cycle and failure to adjust in combination with the actual operation situation of the device, which may lead to waste of maintenance resources or insufficient maintenance, thus affecting the stable operation and service life of the device. Summary of the Invention

[0005] The purpose of the present invention is to solve the deficiencies existing in the prior art and propose a medical device operation and maintenance management system based on information mining and analysis.

[0006] To achieve the above object, the present invention adopts the following technical solutions: A medical device operation and maintenance management system based on information mining and analysis includes:

[0007] The blood oxygen data evaluation module obtains the blood oxygen saturation measurement values of the blood oxygen sensor on the blood oxygen meter and calculates the corresponding mean offset, selects abnormal measurement data according to the offset, analyzes the distribution trend of the abnormal measurement data, and generates a blood oxygen data abnormal feature set;

[0008] The environmental impact analysis module, based on the blood oxygen data abnormal feature set, obtains the power consumption of the blood oxygen sensor and all environmental temperature data, calculates the correlation coefficient between the environmental temperature change and the power consumption offset, selects the environmental impact time period, and obtains the load environmental impact analysis result;

[0009] The operating state evaluation module, based on the load environmental impact analysis result, obtains the measurement frequency and signal response time of the blood oxygen sensor at the load environmental impact time point, evaluates the offset degree of the sensor operating state, and generates a sensor operating state analysis result;

[0010] The decline stage identification module, based on the sensor operating state analysis result, analyzes the cumulative decline change trend of the blood oxygen sensor measurement error and generates blood oxygen sensor decline information;

[0011] The operation and maintenance strategy optimization module, based on the blood oxygen sensor decline information, optimizes and adjusts according to the current operation and maintenance cycle, and generates a blood oxygen sensor operation and maintenance management result.

[0012] As a further solution of the present invention, the blood oxygen data abnormal feature set includes measurement error variance distribution, abnormal measurement data ratio, and abnormal measurement data trend. The load environmental impact analysis result includes blood oxygen sensor power consumption change, environmental temperature gradient impact, and power consumption offset related parameters. The sensor operating state analysis result includes measurement frequency fluctuation amplitude, signal response time offset, and operating state offset degree. The blood oxygen sensor decline information includes measurement error cumulative change trend, data loss rate, and signal response time change amplitude. The blood oxygen sensor operation and maintenance management result includes maintenance adjustment requirements, operation and maintenance cycle optimization, and failure prevention rate analysis.

[0013] As a further solution of the present invention, the blood oxygen data evaluation module includes:

[0014] The blood oxygen offset screening sub-module obtains the blood oxygen saturation measurement values of the blood oxygen sensor on the desktop blood oxygen meter, calculates the mean offset of the blood oxygen saturation measurement values within a specified time, screens the time periods with the offset exceeding the preset offset threshold, calculates the offset rate within the screened time periods, and obtains an offset time period set;

[0015] The measurement error analysis sub-module, based on the offset time period set, uses the formula:

[0016] ;

[0017] Calculate the variance of the measurement error , filter the error variances exceeding the stability threshold to form a first abnormal interval;

[0018] where N is the number of measurement values within the abnormal interval, is the i-th measurement value within the abnormal interval, M is the number of measurement points used to calculate the variance of the measurement error within the abnormal interval, is the j-th measurement value used to calculate the variance of the measurement error within the abnormal interval;

[0019] The abnormal feature extraction sub-module obtains the abnormal proportion of the measurement data within the first abnormal interval, analyzes the distribution trend of the abnormal measurement data, calculates the aggregation degree and distribution pattern of the abnormal values, obtains the distribution trend of the abnormal data, and establishes an abnormal feature set of the blood oxygen data.

[0020] As a further solution of the present invention, the environmental impact analysis module includes:

[0021] The power consumption mean change analysis sub-module obtains the power consumption data of the blood oxygen sensor and all environmental temperature data, filters the time period corresponding to the first abnormal interval in the abnormal feature set of the blood oxygen data, and extracts the power consumption measurement values of the blood oxygen sensor within the time period to obtain the power consumption mean change;

[0022] The load and environment correlation calculation sub-module, based on the power consumption mean change, filters the time periods with the mean change exceeding the load change threshold to form a second abnormal interval, and uses the formula:

[0023] ;

[0024] Calculate the correlation coefficient r between the environmental temperature change and the power consumption offset;

[0025] where, represents the environmental temperature value at the -th time point within the second abnormal interval, represents the power consumption value of the blood oxygen sensor at the -th time point within the second abnormal interval, represents the mean value of all environmental temperature values within the time period, represents the mean value of all power consumption values within the time period, represents the number of data points within the second abnormal interval;

[0026] The load environment impact analysis sub-module, based on the correlation coefficient between the environmental temperature change and the power consumption offset, filters the parameters with the correlation coefficient exceeding the preset correlation threshold and marks the corresponding time periods to obtain the load environment impact analysis result.

[0027] As a further aspect of the present invention, the operating state evaluation module includes:

[0028] Based on the results of the load environment impact analysis, the measurement parameter extraction sub-module obtains the measurement frequency and signal response time of the blood oxygen sensor at the time points affected by the load environment, and records the time stamps corresponding to the data to form a measurement parameter set;

[0029] Based on the measurement parameter set, the measurement frequency fluctuation calculation sub-module uses the formula:

[0030] ;

[0031] Calculate the measurement frequency fluctuation amplitude ;

[0032] Wherein, represents the measurement frequency at the th measurement time point, represents the highest measurement frequency value within the measurement time period, represents the lowest measurement frequency value within the measurement time period;

[0033] Based on the measurement frequency fluctuation amplitude, the offset trend analysis sub-module analyzes the offset trend between the change in the measurement frequency fluctuation amplitude and the change in the signal response time, using the formula:

[0034] ;

[0035] Calculate the offset parameter B of the operating state of the blood oxygen sensor, evaluate the offset degree of the sensor operating state, and generate the analysis result of the sensor operating state;

[0036] Wherein, represents the measurement frequency fluctuation amplitude at the th measurement time point, represents the average value of all measurement frequency fluctuation amplitudes within the time period, represents the signal response time at the th measurement time point, represents the average value of all signal response times within the time period, represents the number of data points within the time period.

[0037] As a further aspect of the present invention, the decline stage identification module includes:

[0038] Based on the analysis result of the sensor operating state, the measurement error evaluation sub-module obtains the cumulative amount of measurement errors at the corresponding time points of the blood oxygen sensor in the first abnormal interval and the second abnormal interval, calculates the total cumulative amount of measurement errors at each time point, and obtains the measurement error value of the data loss rate. By statistically analyzing the distribution characteristics of the measurement errors at each time point, the cumulative amount of measurement errors is obtained.

[0039] The error cumulative analysis sub-module analyzes the cumulative change trend of the measurement error cumulative amount, calculates the change amplitude of the signal response time of the blood oxygen sensor, combines the error change trend and the signal response change to evaluate the overall decline state, and generates the decline information of the blood oxygen sensor.

[0040] As a further solution of the present invention, the operation and maintenance strategy optimization module includes:

[0041] The maintenance record acquisition sub-module obtains the maintenance records of the blood oxygen sensor at the decline characteristic time points based on the decline information of the blood oxygen sensor, determines the maintenance methods at the corresponding time points, extracts the maintenance execution time, calculates the maintenance interval time, and obtains the current operation and maintenance cycle by calculating multiple maintenance time point interval sequences.

[0042] The operation and maintenance cycle adjustment sub-module calls the current operation and maintenance cycle, analyzes the impact of multiple cycle adjustments on the operation and maintenance stability with reference to the failure prevention rate of the maintenance method, evaluates the necessity of the current cycle adjustment and optimizes the operation and maintenance management, and generates the operation and maintenance management result of the blood oxygen sensor.

[0043] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0044] In the present invention, through the precise screening of data anomaly features, the evaluation of the device operating state is made more refined. Based on the mean offset of the measurement data of the blood oxygen sensor, abnormal data can be effectively identified, and combined with the distribution trend of the measurement errors, the accuracy of the device state evaluation is ensured. The influence of environmental factors on the device performance is deeply explored. By calculating the correlation between the environmental temperature change and the power consumption offset, the dynamic change trend of the device performance under different environmental conditions can be identified, and the sensitivity to the influence of external factors is improved. The evaluation method of the operating state is more comprehensive. Through the collaborative analysis of the measurement frequency fluctuation amplitude and the signal response time, the subtle change trend of the device operating state can be accurately depicted, and the accuracy of fault prediction is improved. The identification of the device decline stage no longer depends solely on a single index, but combines the cumulative change of the measurement error and the fluctuation of the signal response time to form multi-dimensional decline information, accurately depicting the evolution process of the device health state. Based on the intelligent optimization strategy of the operation and maintenance cycle, the maintenance plan is made more scientific and reasonable. By calculating the maintenance time interval and comparing and analyzing the failure prevention rate, the maintenance rhythm can be dynamically adjusted to achieve the rational allocation of resources and the maximization of the device life. Brief Description of the Drawings

[0045] Figure 1 is the system flow chart of the present invention;

[0046] Figure 2 is the flow chart of the blood oxygen data evaluation module of the present invention;

[0047] Figure 3 is the flow chart of the environmental impact analysis module of the present invention;

[0048] Figure 4 is the flow chart of the operation status evaluation module of the present invention;

[0049] Figure 5 is the flow chart of the decline stage identification module of the present invention;

[0050] Figure 6 is the flow chart of the operation and maintenance strategy optimization module of the present invention. Detailed implementation manners

[0051] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0052] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. In addition, in the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0053] Please refer to Figure 1 , a medical device operation and maintenance management system based on information mining and analysis includes:

[0054] The blood oxygen data evaluation module obtains the blood oxygen saturation measurement value of the blood oxygen sensor on the desktop blood oxygen meter, calculates the mean offset of the blood oxygen saturation measurement value within a specified time, screens the time periods with the offset exceeding the preset offset threshold, calculates the measurement error variance at the selected time points, screens the first abnormal interval with the measurement error variance exceeding the stability threshold and obtains the abnormal proportion of the measurement data within the first abnormal interval, analyzes the distribution trend of the abnormal measurement data, and generates a blood oxygen data abnormal feature set;

[0055] The environmental impact analysis module obtains the power consumption of the blood oxygen sensor and all ambient temperature data, calculates the mean change of the power consumption of the blood oxygen sensor during the time period corresponding to the first abnormal interval in the abnormal blood oxygen data feature set, iteratively screens the second abnormal interval where the mean change exceeds the load change threshold, calculates the correlation coefficient between the ambient temperature change and the power consumption offset according to the ambient temperature gradient within the second abnormal interval, screens the parameters whose correlation coefficient exceeds the preset correlation threshold and marks the corresponding time periods to obtain the load environmental impact analysis result;

[0056] The operating state evaluation module, based on the load environmental impact analysis result, obtains the measurement frequency and signal response time of the blood oxygen sensor at the load environmental impact time point, calculates the fluctuation amplitude of the measurement frequency, analyzes the offset trend between the change of the measurement frequency fluctuation amplitude and the change of the signal response time, evaluates the offset degree of the sensor operating state, and generates the sensor operating state analysis result;

[0057] The decline stage identification module, based on the sensor operating state analysis result, obtains the measurement errors of the measurement error cumulative amount and data loss rate of the blood oxygen sensor at the corresponding time points in the first abnormal interval and the second abnormal interval, analyzes the cumulative change trend of the measurement error and the change amplitude of the signal response time of the blood oxygen sensor, and generates the blood oxygen sensor decline information;

[0058] The operation and maintenance strategy optimization module, based on the blood oxygen sensor decline information, obtains the maintenance records and the current operation and maintenance cycle of the blood oxygen sensor at the decline characteristic time point, refers to the failure prevention rate of the maintenance method, analyzes the adjustment requirements of the current operation and maintenance cycle and makes optimized adjustments to generate the blood oxygen sensor operation and maintenance management result;

[0059] The abnormal blood oxygen data feature set includes the measurement error variance distribution, the proportion of abnormal measurement data, and the abnormal measurement data trend. The load environmental impact analysis result includes the power consumption change of the blood oxygen sensor, the influence of the ambient temperature gradient, and the power consumption offset related parameters. The sensor operating state analysis result includes the measurement frequency fluctuation amplitude, the signal response time offset, and the operating state offset degree. The blood oxygen sensor decline information includes the cumulative change trend of the measurement error, the data loss rate, and the change amplitude of the signal response time. The blood oxygen sensor operation and maintenance management result includes the maintenance adjustment requirements, the operation and maintenance cycle optimization, and the failure prevention rate analysis.

[0060] Please refer to Figure 2 for the blood oxygen data evaluation module, which includes:

[0061] The blood oxygen offset screening sub-module obtains the blood oxygen saturation measurement values of the blood oxygen sensor on the desktop blood oxygen meter, calculates the mean offset of the blood oxygen saturation measurement values within the specified time, screens the time periods where the offset exceeds the preset offset threshold, and calculates the offset rate within the screened time periods to obtain the offset time period set;

[0062] To obtain the blood oxygen saturation measurement value of the blood oxygen sensor on a desktop pulse oximeter, first collect the blood oxygen saturation data within a continuous time period through the blood oxygen sensor of the desktop pulse oximeter. These data are recorded in seconds and stored in the device's data storage unit. For example, within a 10-second monitoring period, the sequence of blood oxygen saturation values collected by the device may be: , and then calculate the mean offset of these data. The method for calculating the mean offset is to first find the mean of this time period , for example: , and then calculate the absolute deviation value between each measurement value and the mean: . Next, set an offset threshold, such as 0.3%, and filter out the time periods with an offset greater than this threshold. In this example, the time points with an offset exceeding 0.3% are: 97.4% and 98.2%. Therefore, the corresponding time periods of these time points will be marked and finally form a set of offset time periods.

[0063] Based on the set of offset time periods, the measurement error analysis sub-module uses the formula: ;

[0064] Calculate the measurement error variance , filter out the error variances exceeding the stability threshold to form the first abnormal interval;

[0065] where N is the number of measurement values within the abnormal interval, is the i-th measurement value within the abnormal interval, M is the number of measurement points used to calculate the measurement error variance within the abnormal interval, is the j-th measurement value used to calculate the measurement error variance within the abnormal interval.

[0066] Set N = 3, M = 3, the measurement values within the abnormal interval are 97.5%, 98.1%, 97.8%, and the measurement values for calculating the mean are 97.4%, 97.9%, 98.0%. First, calculate the mean: . Then calculate the variance components:

[0067] , ;

[0068] Then calculate the error variance: . Set the stability threshold, such as 0.05. When the calculated error variance is greater than this threshold, this time point is considered abnormal and forms the first abnormal interval.

[0069] The abnormal feature extraction sub-module obtains the abnormal proportion of the measurement data within the first abnormal interval, analyzes the distribution trend of the abnormal measurement data, calculates the aggregation degree and distribution pattern of the abnormal values, obtains the distribution trend of the abnormal data, and establishes an abnormal feature set of the blood oxygen data;

[0070] Count the total number of all measured values within this interval, denoted as Among them, the number of abnormal measured values is denoted as , and the abnormal proportion The calculation formula is as follows: , assuming that the first abnormal interval contains 5 measured values, and 2 of them exceed the set threshold, then the abnormal proportion is calculated as follows: . Then, analyze the distribution trend of abnormal measurement data, and focus on whether the abnormal values have concentration or randomness in time. The distribution trend of abnormal values can be judged by statistically analyzing the relative positions of abnormal data within the entire abnormal interval. For example, if the abnormal data is concentrated in the first 50% of the time period, there may be a trend of abnormal early signals; if the abnormal data is evenly distributed, it may be a random fluctuation anomaly. The sliding window method can be used to calculate the time interval between adjacent abnormal values to obtain the time distribution trend of abnormal values. If abnormal data appears continuously in a short period of time, it indicates that the abnormal measured values show aggregation; if the interval between abnormal data is large, it indicates that the abnormal values are randomly distributed. Finally, based on the distribution trend of abnormal data, an abnormal feature set of blood oxygen data is established for subsequent further analysis.

[0071] Please refer to Figure 3 , the environmental impact analysis module includes:

[0072] The power consumption mean change analysis sub-module obtains the power consumption data of the blood oxygen sensor and all environmental temperature data, filters the time period corresponding to the first abnormal interval in the abnormal feature set of blood oxygen data, and extracts the power consumption measurement values of the blood oxygen sensor within the time period to obtain the power consumption mean change;

[0073] Obtain the power consumption data of the blood oxygen sensor and all environmental temperature data, and extract the power consumption and temperature data at all measurement time points from the stored data set. When the blood oxygen sensor works continuously in time, its power consumption will be affected by temperature and change. Assume that the sensor collects data once per second and records and stores it in the device. Within a 30-second time period, the power consumption measurement data may be as follows . Subsequently, filter the time period corresponding to the first abnormal interval in the abnormal feature set of blood oxygen data from these data, and extract the power consumption measurement values within this time period. For example, if the first abnormal interval occurs in the time period of 10s to 20s, then take the power consumption data of this time period . Calculate the power consumption mean within this time period, and the mean is calculated using the arithmetic mean of all data points , and its calculation is as follows: . Then, calculate the mean change amount of this time period. The change amount is calculated as the difference between the mean of this time period and the mean of the previous time period. Assume that the mean of the previous time period from 0s to 10s is 2.2W, then the mean change amount is calculated as follows: Finally, the variation of the average power consumption is obtained.

[0074] Based on the variation of the average power consumption, the load and environment correlation calculation sub-module filters out the time periods with the average variation exceeding the load change threshold to form a second abnormal interval, using the formula: ;

[0075] Calculate the correlation coefficient r between the environmental temperature change and the power consumption offset;

[0076] Among them, represents the environmental temperature value at the th time point within the second abnormal interval, represents the power consumption value of the blood oxygen sensor at the th time point within the second abnormal interval, represents the average value of all environmental temperature values within the time period, represents the average value of all power consumption values within the time period, represents the number of data points within the second abnormal interval.

[0077] Set the load change threshold to 0.15W. If the average variation of a certain time period is greater than this threshold, then this time period is marked as the second abnormal interval. In this example, , therefore, this time period is identified as the second abnormal interval. Then, for the environmental temperature data within the second abnormal interval, calculate the temperature gradient. The temperature gradient calculation method is the rate of change of temperature over time within this time period. Assume that the temperature data changes as follows within the second abnormal interval time period:

[0078] . Subsequently, calculate the correlation coefficient between the environmental temperature change and the power consumption offset, where , the temperature data is , the power consumption data is , the average value of the temperature data is: , the average value of the power consumption data is: 2.4W.

[0079] Calculate the numerator part:

[0080] ;

[0081] Calculate the denominator part:

[0082] ;

[0083] Finally, calculate the correlation coefficient: .

[0084] The load environment impact analysis sub-module filters the parameters with correlation coefficients exceeding the preset correlation threshold based on the correlation coefficient between environmental temperature change and power consumption offset, marks the corresponding time periods, and obtains the load environment impact analysis result;

[0085] Filter the parameters with correlation coefficients exceeding the preset correlation threshold and mark the corresponding time periods. Set the preset correlation threshold to 0.3, and the value range of the correlation coefficient r is , where: Indicates a positive correlation. As the environmental temperature increases, the power consumption also increases; Indicates a negative correlation. As the environmental temperature increases, the power consumption decreases; r = 0 indicates no correlation. Generally speaking, the threshold setting of the correlation coefficient usually refers to the significance level in statistics. , indicating a weak correlation between variables, represents a medium correlation, represents a strong correlation. Therefore, in practical applications, a reasonable threshold can be set according to experience, such as 0.3 as the lowest significant correlation limit. Since the power consumption of the blood oxygen sensor is affected by the environmental temperature, but this effect may not be linear, it is necessary to adjust it in combination with experimental data. For example, measure the power consumption fluctuations of multiple sensors at different environmental temperatures in the experiment, and after statistically analyzing the correlation coefficients of each time period, analyze at which threshold the correlation performance above can stably reflect the power consumption offset trend. If the experimental data shows that when , the power consumption change trend is consistent with the temperature change trend, and when , the power consumption change trend is relatively random, then 0.3 can be set as the effective correlation threshold for this system. If , it is considered that the environmental temperature has a correlation with the power consumption change, and this time period is marked. Since the calculated correlation coefficient r = 0.36 is greater than 0.3, this time period is marked and the timestamp information is recorded. Finally, the load environment impact analysis result is obtained.

[0086] Please refer to Figure 4 , the operating state evaluation module includes:

[0087] Based on the load environment impact analysis result, the measurement parameter extraction sub-module obtains the measurement frequency and signal response time of the blood oxygen sensor at the load environment impact time points, and records the timestamps corresponding to the data, forming a measurement parameter set;

[0088] Extract the measurement frequency and signal response time of the blood oxygen sensor at the load environment impact time points. First, filter out all time periods affected by the load environment and sort them according to the timestamps to ensure data integrity. For example, at the time point ;

[0089] The corresponding measurement frequency data may be: ;

[0090] The signal response time data may be ;

[0091] Then, the measurement parameters at all time points are summarized to form a measurement parameter set for subsequent calculations.

[0092] The measurement frequency fluctuation calculation sub-module is based on the measurement parameter set and uses the formula:

[0093] ;

[0094] Calculate the measurement frequency fluctuation amplitude ;

[0095] where represents the measurement frequency at the th measurement time point, represents the highest measurement frequency value within the measurement time period, represents the lowest measurement frequency value within the measurement time period.

[0096] The measurement frequency data is , calculate the highest measurement frequency value and the lowest measurement frequency value:

[0097] . Substitute into the formula to calculate the measurement frequency fluctuation amplitude: .

[0098] The offset trend analysis sub-module analyzes the offset trend between the change in the measurement frequency fluctuation amplitude and the change in the signal response time based on the measurement frequency fluctuation amplitude, and uses the formula: ;

[0099] Calculate the offset parameter B of the operating state of the blood oxygen sensor, evaluate the offset degree of the sensor operating state, and generate the analysis result of the sensor operating state;

[0100] where represents the measurement frequency fluctuation amplitude at the th measurement time point, represents the mean value of all measurement frequency fluctuation amplitudes within the time period, represents the signal response time at the th measurement time point, represents the mean value of all signal response times within the time period, represents the number of data points within the time period, represents the summation operation on all data points within this time period.

[0101] Set , the measurement frequency fluctuation amplitude data ;

[0102] Signal response time data Calculate the mean value:

[0103] Calculate the mean value of the fluctuation amplitude:

[0104] ;

[0105] Calculate the mean value of the response time:

[0106] ;

[0107] Calculate the numerator part:

[0108] .

[0109] The offset parameter B is used to measure the correlation between the fluctuation amplitude of the measurement frequency and the change of the signal response time. When the value of B is small, it indicates that the change trends of the two are relatively stable and the offset degree is low, that is, the operating state of the sensor is relatively stable. When the value of B increases, it indicates that there is a strong offset trend between the change of the fluctuation amplitude of the measurement frequency and the change of the signal response time, which may mean that the sensor appears abnormal or unstable under the influence of the load environment. In this example, B = 0.00018 indicates that the relationship between the fluctuation amplitude of the measurement frequency and the change of the signal response time is small, and the offset degree of the sensor operating state is low. If it is found that the value of B continues to increase during the subsequent monitoring process, it can be used as an early warning signal for the abnormal operating state of the sensor.

[0110] Please refer to Figure 5 , the decline stage identification module includes:

[0111] Based on the analysis results of the sensor operating state, the measurement error evaluation sub-module obtains the cumulative amount of measurement errors at the corresponding time points of the blood oxygen sensor in the first abnormal interval and the second abnormal interval, calculates the total cumulative amount of measurement errors at each time point, and obtains the measurement error value of the data loss rate. By statistically analyzing the distribution characteristics of the measurement errors at each time point, the cumulative amount of measurement errors is obtained;

[0112] First, select the data at specific time points in the first abnormal interval and the second abnormal interval, and perform cumulative calculation on the measurement errors of the blood oxygen sensor at these time points. For example, assume that the measured value of the sensor at a certain time point is , and the actual reference value is , then the measurement error at this time point . Calculate the error values at the time points in all abnormal intervals in turn, and then obtain the overall cumulative amount of measurement errors. The calculation of the data loss rate is based on the set sampling period. For example, if the set sampling period is 1 s, a total of 10 groups of data should be obtained within 10 s. If 8 groups of data are actually obtained, the data loss rate L is calculated as follows: When calculating the cumulative amount of measurement errors, it is necessary to cumulatively sum all the error values. For example, in the first abnormal interval, assuming the error values at 5 time points are respectively , then the cumulative amount of measurement errors in this interval is calculated as follows: , and the cumulative amount of measurement errors can be obtained through the above calculation.

[0113] The error cumulative analysis sub-module analyzes the cumulative change trend of the cumulative amount of measurement errors, calculates the change amplitude of the signal response time of the blood oxygen sensor, evaluates the overall decline state by combining the error change trend and the signal response change, and generates the decline information of the blood oxygen sensor;

[0114] Call the cumulative amount of measurement errors, analyze the cumulative change trend of the errors over time, calculate the error growth rate, assuming the time point interval is 1 s;

[0115] The corresponding error cumulative data is ;

[0116] Then calculate the error change rate , where is the cumulative amount of measurement errors at time points, in %, representing the cumulative measurement error of the sensor at the current moment. The cumulative amount of measurement errors at the time point, is compared with to calculate the error change rate. is the time stamp of the time point, in s (seconds), representing the acquisition time of the measurement data. The time stamp of the time point, is compared with to calculate the time interval. For the above data, calculate the error change rate between the first two points: , and calculate the rate of each point in turn to obtain the error change rate sequence:

[0117] , at the same time, calculate the change amplitude of the signal response time, set the measured value sequence of the signal response time as , calculate the signal response change range , where S is the signal response time data sequence, which contains signal response time measurement values at multiple time points. max(S) represents the maximum value of the signal response time, indicating the longest response time of the sensor at a certain time point. min(S) represents the minimum value of the signal response time, indicating the shortest response time of the sensor at a certain time point. Through the error change rate and the signal response change amplitude, the decline trend of the blood oxygen sensor can be identified. If the cumulative measurement error continues to increase and the error growth rate remains in a high-value range (for example, close to 2% / s), it indicates that the measurement stability of the sensor has decreased. At the same time, if the fluctuation range of the signal response time increases (for example, exceeds 0.3 s), it means that the reaction delay of the sensor to the input signal increases, further exacerbating the unpredictability of the measurement error. In the above cases, it can be determined that the blood oxygen sensor has entered the decline stage, its working state is no longer stable, and it may affect the measurement reliability. Therefore, by combining the cumulative error change trend and the signal response change, the decline information of the blood oxygen sensor is finally generated.

[0118] Please refer to Figure 6 , the operation and maintenance strategy optimization module includes:

[0119] The maintenance record acquisition sub-module, based on the decline information of the blood oxygen sensor, acquires the maintenance record of the blood oxygen sensor at the decline characteristic time point, determines the corresponding maintenance method at this time point, extracts the maintenance execution time, calculates the maintenance interval time, and obtains the current operation and maintenance cycle by calculating multiple maintenance time point interval sequences;

[0120] First, it is necessary to screen out the decline characteristic time points, which are usually significantly related to the sensor performance degradation, the increase in measurement error, or the change in signal response time. Assume that the normal measurement error range of the blood oxygen sensor is within . If the error value at a certain time point exceeds 2.5%, this time point can be determined as a potential decline characteristic point. For example, if the measurement error at a certain time point is 3.1%, which is higher than the threshold, further analysis of the maintenance record is required. Then, call the maintenance database to obtain the maintenance method corresponding to the time point. The maintenance method may include sensor calibration, component replacement, firmware upgrade, etc. The maintenance record stores information such as the maintenance type, maintenance time, and fault description. For example, a sensor calibration was performed at 08:00 on January 15, 2024, and the measurement error returned to 1.2% after the maintenance. Extract the execution time of all maintenance operations and establish a time series, such as . Calculate the interval between adjacent maintenance times to evaluate the current operation and maintenance cycle. For example, assume that the time intervals between two maintenances are 14 days, 21 days, and 10 days respectively. Then the current operation and maintenance cycle can be calculated as the average value 15 days, and finally obtain the current operation and maintenance cycle.

[0121] The operation and maintenance cycle adjustment sub-module calls the current operation and maintenance cycle, analyzes the impact of multiple cycle adjustments on operation and maintenance stability with reference to the failure prevention rate of the maintenance method, evaluates the necessity of the current cycle adjustment, optimizes operation and maintenance management, and generates the operation and maintenance management result of the blood oxygen sensor;

[0122] Call the current operation and maintenance cycle, evaluate the failure prevention capabilities of different maintenance strategies. The failure prevention rate measures the effectiveness of the maintenance method, and calculate the prevention capabilities under different operation and maintenance cycles. For example, for the relationship between the maintenance interval T and the failure incidence rate F, the following formula can be used: represents the failure prevention rate under the current maintenance cycle, is the number of failures in the current cycle, is the number of failures in the reference operation and maintenance cycle. If 10 failures occur within 15 days in the reference cycle and 14 failures occur in 20 days in the current cycle, then calculate: , a negative value indicates an increase in failures in the current cycle, and the maintenance interval needs to be shortened. Analyze the impact of different adjustment strategies on operation and maintenance stability in this way, and finally generate the operation and maintenance management result of the blood oxygen sensor.

[0123] The above is only the preferred embodiment of the present invention, and does not limit the present invention in other forms. Any person skilled in the art may use the technical content disclosed above to make changes or modifications into equivalent embodiments with equivalent changes and apply them to other fields. However, as long as it does not depart from the technical solution content of the present invention, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A medical equipment operation and maintenance management system based on information mining and analysis, characterized in that: The system comprises: The blood oxygen data evaluation module obtains the blood oxygen saturation measurement value of the blood oxygen sensor on the oximeter and calculates the corresponding mean offset, selects abnormal measurement data according to the offset, analyzes the distribution trend of the abnormal measurement data, and generates an abnormal feature set of blood oxygen data; The environmental impact analysis module obtains the power consumption and all ambient temperature data of the blood oxygen sensor based on the abnormal feature set of the blood oxygen data, calculates the correlation coefficient between the ambient temperature change and the power consumption offset, selects the environmental impact time period, and obtains the load environmental impact analysis result; The operation status evaluation module obtains the measurement frequency and signal response time of the blood oxygen sensor at the load environment impact time point based on the load environment impact analysis result, evaluates the deviation degree of the sensor operation status, and generates the sensor operation status analysis result; The decay stage identification module analyzes the cumulative decay change trend of the blood oxygen sensor measurement error based on the sensor operation status analysis result, and generates blood oxygen sensor decay information; The operation and maintenance strategy optimization module performs optimization and adjustment according to the blood oxygen sensor degradation information and the current operation and maintenance cycle to generate the blood oxygen sensor operation and maintenance management results.

2. The medical equipment operation and maintenance management system based on information mining analysis according to claim 1 is characterized in that: The abnormal feature set of blood oxygen data includes measurement error variance distribution, abnormal measurement data proportion, and abnormal measurement data trend. The load environment impact analysis result includes blood oxygen sensor power consumption change, ambient temperature gradient impact, and power consumption offset related parameters. The sensor operation status analysis result includes measurement frequency fluctuation amplitude, signal response time offset, and operation status offset degree. The blood oxygen sensor decay information includes measurement error cumulative change trend, data loss rate, and signal response time change amplitude. The blood oxygen sensor operation and maintenance management result includes maintenance adjustment requirements, operation and maintenance cycle optimization, and fault prevention rate analysis.

3. The medical equipment operation and maintenance management system based on information mining analysis according to claim 2 is characterized in that: The blood oxygen data evaluation module includes: The blood oxygen offset screening submodule obtains the blood oxygen saturation measurement value of the blood oxygen sensor on the desktop oximeter, calculates the mean offset of the blood oxygen saturation measurement value within a specified time, screens the time period in which the offset exceeds a preset offset threshold, calculates the offset rate within the screened time period, and obtains an offset time period set; The measurement error analysis submodule adopts the formula based on the offset time period set: ； Calculate measurement error variance , screen the error variance that exceeds the stability threshold to form the first abnormal interval; Where N is the number of measurements within the abnormal interval, is the ith measurement value in the abnormal interval, M is the number of measurement points used to calculate the variance of the measurement error in the abnormal interval, are the j measurements used to calculate the variance of the measurement error within the anomaly interval; The abnormal feature extraction submodule obtains the abnormal proportion of the measurement data in the first abnormal interval, analyzes the distribution trend of the abnormal measurement data, calculates the aggregation degree and distribution pattern of the abnormal values, obtains the distribution trend of the abnormal data, and establishes an abnormal feature set of the blood oxygen data.

4. The medical equipment operation and maintenance management system based on information mining analysis according to claim 3 is characterized in that: The environmental impact analysis module includes: The power consumption mean change analysis submodule obtains the power consumption data and all ambient temperature data of the blood oxygen sensor, selects the time period corresponding to the first abnormal interval in the blood oxygen data abnormal feature set, and extracts the power consumption measurement value of the blood oxygen sensor in the time period to obtain the power consumption mean change; The load and environment association calculation submodule selects the time period in which the mean change exceeds the load change threshold based on the mean change of power consumption to form a second abnormal interval, using the formula: ； Calculate the correlation coefficient r between ambient temperature change and power consumption offset; in, Represents the second abnormal interval The ambient temperature value at a time point, Represents the second abnormal interval The power consumption value of the blood oxygen sensor at a time point, Represents the average value of all ambient temperature values ​​within the time period, Represents the average of all power consumption values ​​within the time period, Represents the number of data points in the second anomaly interval; The load environment impact analysis submodule selects parameters whose correlation coefficients exceed a preset correlation threshold based on the correlation coefficient between the ambient temperature change and the power consumption offset, and marks the corresponding time period to obtain the load environment impact analysis result.

5. The medical equipment operation and maintenance management system based on information mining analysis according to claim 4 is characterized in that: The operating status assessment module comprises: The measurement parameter extraction submodule obtains the measurement frequency and signal response time of the blood oxygen sensor at the load environment impact time point based on the load environment impact analysis result, and records the timestamp corresponding to the data to form a measurement parameter set; The measurement frequency fluctuation calculation submodule adopts the formula based on the measurement parameter set: ； Calculate the measurement frequency fluctuation amplitude ; in, Representative The measurement frequency at each measurement time point, Represents the highest measurement frequency value within the measurement period. Represents the lowest measurement frequency value within the measurement period; The offset trend analysis submodule analyzes the offset trend between the change of the measured frequency fluctuation amplitude and the change of the signal response time based on the measured frequency fluctuation amplitude, using the formula: ； Calculate the offset parameter B of the blood oxygen sensor operating state, evaluate the offset degree of the sensor operating state, and generate the sensor operating state analysis result; in, Representative The measured frequency fluctuation amplitude at each measurement time point, Represents the average value of all measured frequency fluctuations within the time period, Representative The signal response time at each measurement time point, Represents the mean of all signal response times within a time period, Represents the number of data points in the time period.

6. The medical equipment operation and maintenance management system based on information mining analysis according to claim 5 is characterized in that: The decay stage identification module comprises: The measurement error evaluation submodule obtains the measurement error accumulation of the blood oxygen sensor at the time points corresponding to the first abnormal interval and the second abnormal interval based on the analysis result of the sensor operation state, calculates the total measurement error accumulation at each time point, obtains the measurement error value of the data loss rate, and obtains the measurement error accumulation by statistically analyzing the distribution characteristics of the measurement error at each time point; The error accumulation analysis submodule analyzes the cumulative change trend of the measurement error accumulation amount, calculates the change amplitude of the blood oxygen sensor signal response time, evaluates the overall decay state by combining the error change trend and the signal response change, and generates blood oxygen sensor decay information.

7. The medical equipment operation and maintenance management system based on information mining analysis according to claim 6 is characterized in that: The operation and maintenance strategy optimization module includes: The maintenance record acquisition submodule acquires the maintenance record of the blood oxygen sensor at the decay characteristic time point based on the decay information of the blood oxygen sensor, determines the maintenance method at the corresponding time point, extracts the maintenance execution time, calculates the maintenance interval time, and acquires the current operation and maintenance cycle by calculating the interval sequence of multiple maintenance time points; The operation and maintenance cycle adjustment submodule calls the current operation and maintenance cycle, refers to the fault prevention rate of the maintenance method, analyzes the impact of multiple cycle adjustments on the operation and maintenance stability, evaluates the necessity of the current cycle adjustment, optimizes the operation and maintenance management, and generates the blood oxygen sensor operation and maintenance management results.

Images