Method for Identifying Equipment Status, Fault and Handling Abnormality in Central Sterile Supply Department

Abstract

The present invention relates to the technical field of medical device management, and particularly to a method for identifying the status, faults and handling abnormalities of devices used in a disinfection supply center. The method includes: Step 1, obtaining a sterilization task, and collecting operation status data of different target devices at a preset frequency based on the sterilization task; Step 2, constructing an abnormal response framework, analyzing the obtained operation status data, and implementing different response strategies according to the analysis results; Step 3, based on the sterilization task, obtaining biological monitoring data and chemical monitoring data of the corresponding target devices, and judging whether the sterilization effect is qualified based on the chemical monitoring data and the biological monitoring data. If so, return to Step 1; otherwise, enter Step 4; Step 4, constructing a data matrix using the device operation status data, the biological monitoring data and the chemical monitoring data, and obtaining the reasons for the unqualified sterilization effect based on the analysis of the data matrix.

Description

Technical Field

[0001] The present invention relates to the technical field of medical device management, and particularly to a method for identifying the status, faults and handling abnormalities of devices used in a disinfection supply center. Background Art

[0002] The disinfection supply center is a department in a medical institution for cleaning, disinfecting, sterilizing and managing reusable medical devices, instruments, etc.

[0003] Currently, the management of sterilization equipment in the disinfection supply center mainly relies on manual monitoring. There are various types of equipment models in the disinfection supply center, resulting in diverse system data and a lack of unified data standards, making it impossible to achieve real-time alarms; manual monitoring is difficult to ensure the accuracy and timeliness of monitoring, which may thus affect the normal operation of sterilization equipment and the disinfection effect.

[0004] To solve the above problems, the present invention combines the Internet of Things and big data analysis, and uses artificial intelligence computing to conduct full-cycle monitoring and intelligent alarm for sterilization equipment, improving the timeliness of monitoring and exception handling. Summary of the Invention

[0005] The present invention analyzes the operation status data of multiple devices in real time, and through an exception response framework, analyzes the operation status data and uses different alarm methods for hierarchical alarms according to the alarm priority.

[0006] The technical solution proposed by the present invention is: a method for identifying the status, faults and handling abnormalities of devices used in a disinfection supply center, the method comprising:

[0007] Step 1: Obtain a sterilization task, and collect the operation status data of different target devices at a preset frequency based on the sterilization task;

[0008] Step 2: Construct an exception response framework, analyze the obtained operation status data, and implement different response strategies according to the analysis results;

[0009] Step 3: Based on the sterilization task, obtain the biological monitoring data and chemical monitoring data of the corresponding target device, and judge whether the sterilization effect is qualified based on the chemical monitoring data and biological monitoring data. If so, return to Step 1; otherwise, enter Step 4;

[0010] Step 4: Use the device operation status data, biological monitoring data and chemical monitoring data to construct a data matrix, and obtain the reasons for the unqualified sterilization effect based on the analysis of the data matrix.

[0011] Preferably, the operation status data includes the pressure, temperature, sterilization time, humidity and gas concentration inside the device;

[0012] The chemical monitoring data includes the image data of chemical indicators before and after sterilization, and the status characteristics of chemical indicators are obtained through the analysis of the image data;

[0013] The biological monitoring data includes the status characteristics of biological test packages obtained from biological monitoring instruments, and the status characteristics of biological test packages include that the biological indicator is negative and the biological indicator is positive.

[0014] Preferably, the abnormal response framework is constructed, the operation status data obtained is analyzed, and different response strategies are implemented according to the analysis results, including:

[0015] Obtain the operation status data, perform normalization processing, and form an operation status feature sequence: , where, respectively represent the pressure, temperature, sterilization duration, humidity and gas concentration of the equipment at moment;

[0016] The analysis of the obtained operation status data and the implementation of different response strategies according to the analysis results include:

[0017] Input the operation status feature sequence into the abnormal alarm framework for alarm feature extraction and coding;

[0018] Calculate the alarm priority to determine the abnormal response order;

[0019] Match the alarm information template to generate corresponding processing suggestions;

[0020] Optimize the instruction suggestion information using natural language processing algorithms;

[0021] Push the alarm information in grades and collect feedback information.

[0022] Preferably, the alarm feature extraction and coding include:

[0023] The alarm level is divided into a first-level alarm and a second-level alarm; the first-level alarm is that a single status data exceeds the preset safety threshold; the second-level alarm is that multiple status data exceed the preset safety threshold, or the equipment self-check fails. Specifically:

[0024] Alarm level ;

[0025] where, represents the indicator function, and the value of the indicator function is 1 when the condition is satisfied, otherwise the value is 0;

[0026] Status parameter ; represents the th safety threshold of the status parameter;

[0027] Mark the alarm types, including:

[0028] Generate a label list according to the type of abnormal status characteristic values;

[0029] Within the time window Count the number of devices triggering the same type of alarm ;

[0030] Read the current operation stage of the target device from the device log, and the current operation stage includes preheating, sterilization, and cooling;

[0031] The calculation of the alarm priority and the determination of the alarm response order include:

[0032] Construct an alarm priority scoring model ;

[0033] Among them, respectively represent the weight coefficients, represents the alarm type weight; represents the alarm type, that is, the type of abnormal status characteristic value; represents the current operation stage weight; ; ;

[0034] Determine the order of abnormal responses according to the alarm priority.

[0035] Preferably, the matching of the alarm information template and the generation of corresponding processing suggestions include:

[0036] Pre-define the information template preprocessing suggestions in different scenarios, including:

[0037] Level 1 alarm template: [Device ID] has [Alarm type] <Current status parameter value [x] and safety threshold [y]> at [Time], it is recommended to <Suggested instruction information>;

[0038] Level 2 alarm template: [Device ID] has a composite fault <[Alarm type 1]+[Alarm type 2]> at [Time], stop sterilization, execute [Self-check protocol] and contact [Responsible person];

[0039] The optimization of the instruction suggestion information by using natural language processing algorithms includes:

[0040] Based on the fault, use the lightweight machine learning algorithm CRF to replace the placeholders in the alarm information template with real-time data, and the placeholders include [Time], [y], [x];

[0041] Combine the alarm type and the device ID to optimize the sentence structure of the alarm information template;

[0042] The hierarchical push of alarm information and collection of feedback information include:

[0043] Adopt different push methods based on the priority level:

[0044] If , the alarm priority is high, and it is pushed through text messages, the App, and audible and visual alarms;

[0045] If , the alarm priority is medium, and it is alarmed through system pop-ups and emails;

[0046] If , the alarm priority is low, the alarm status is written into the log, and manual review is notified;

[0047] Check the log to determine whether the alarm status is updated and give feedback.

[0048] Preferably, based on the sterilization task, obtain the biological monitoring data and chemical monitoring data of the corresponding target device, and judge whether the sterilization effect is qualified based on the chemical monitoring data and biological monitoring data, including:

[0049] Integrate the device operation data, chemical monitoring data, and biological monitoring data into a structured data matrix;

[0050] Obtain the reasons for unqualified sterilization effect through the analysis of the structured data matrix.

[0051] Preferably, the integration of the device operation data, chemical monitoring data, and biological monitoring data into a structured data matrix includes:

[0052] Obtain the operation status characteristic sequence; collect the image data of the chemical indicators before and after sterilization;

[0053] Obtain the color change status data of the chemical indicators by analyzing the collected image data of the chemical indicators before and after sterilization;

[0054] Collect the biological monitoring result data from the biological monitoring instrument;

[0055] Align the operation status characteristic sequence, the color change status data of the chemical indicators, and the biological monitoring result data of the target device in the same sterilization batch on the time axis;

[0056] Obtain the operation status characteristic sequences of all devices , and construct the monitoring characteristic sequence ; where respectively represent the th device identification ID, the pressure change value, average sterilization temperature, equivalent sterilization time, average humidity, and average gas concentration during the sterilization process of the same batch; respectively represent the state characteristic values of the biological monitoring package, the state characteristic values of the biological monitoring negative experimental group, several state characteristic values of chemical monitoring, and the sterilization effect state value during the sterilization process of the same batch;

[0057] Construct a data matrix , where, represents the number of devices;

[0058] Average sterilization temperature Among them, represents the total number of batches;

[0059] Pressure change value ; Among them, represents the average pressure of multiple batches;

[0060] Equivalent sterilization time ; Among them, represents the sterilization end time, and the temperature variable is determined according to the reference bacterial sterilization kinetic index.

[0061] Preferably, judging whether the sterilization effect is qualified based on the chemical monitoring data and the biological monitoring data includes:

[0062] Obtain the state characteristic values of the biological monitoring package ;

[0063] Obtain the state characteristic values of the biological monitoring negative experimental group ;

[0064] Obtain the state characteristic values of the chemical monitoring results ;

[0065] If and and , then judge that the sterilization effect is qualified, that is ; Otherwise, judge that the sterilization effect is unqualified, that is ;

[0066] The reasons for obtaining unqualified sterilization effects through the analysis of the structured data matrix include:

[0067] Obtain , extract multiple elements to form an input feature vector ;

[0068] Input the input feature vector into the logistic regression model to judge the reasons for unqualified:

[0069] The logistic regression model is:

[0070] ;

[0071] Among them, respectively represent the influence weight of the equivalent sterilization time, the influence weight of pressure change on non - compliance, the influence weight of insufficient temperature on non - compliance, the influence weight of unqualified chemical monitoring, and the interaction effect weight of temperature and chemical monitoring; represents the probability of non - compliance;

[0072] Use the maximum likelihood estimation MLE algorithm to solve the logistic regression model and output and the significance value, that is, the p - value;

[0073] If the significance value corresponding to each influence weight is less than the preset significance threshold, it is determined that the influence of this influence weight on the non - compliance of the sterilization effect is not significant;

[0074] Find the significance value with the largest numerical value, and the corresponding feature is the main reason for the non - compliance of sterilization, and focus on optimizing the corresponding feature.

[0075] Preferably, by analyzing the image data of the chemical indicator before and after sterilization, obtaining the discoloration state data of the chemical indicator, including:

[0076] Collect the RGB images of the chemical indicator before and after sterilization, mark the indicator area in the RGB image, and generate an XML or JSON annotation file;

[0077] Based on the annotation, crop the indicator area through the ROI algorithm;

[0078] Convert the image of the indicator area to the HSV space to enhance color sensitivity;

[0079] Input the images of the indicator area before and after sterilization into the CNN network, and output the chemical monitoring compliance status;

[0080] The CNN network includes an input layer, a fully - connected layer, and an output layer. The input layer is used to receive the pre - processed image of the indicator area, and the fully - connected layer is used for indicator area cropping and discoloration degree analysis; the output layer is used to output the classification result, that is, chemical monitoring compliance or chemical monitoring non - compliance.

[0081] The present invention also provides a computer - readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method for identifying the status, faults, and handling abnormalities of the disinfection supply center equipment.

[0082] The beneficial effects of the present invention:

[0083] 1. The present invention collects the operation status data of multiple devices in real time, monitors the operation status of the devices, and determines the alarm level according to the types and quantities of abnormal device operation parameters identified. Based on the alarm level, the alarm priority is calculated by combining the number of alarm devices, alarm types, etc. According to the alarm priority, different alarm methods are used for alarm grading.

[0084] 2. The present invention collects chemical monitoring data and biological monitoring data in real time to determine whether the sterilization effect is qualified, and combines the device operation status data to use a pre-trained CNN model to determine the reasons for the unqualified sterilization effect, so as to take countermeasures. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 It is a flowchart of the method for identifying device status, faults and handling abnormalities in the disinfection supply center of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0086] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments in the following description are only examples, and those skilled in the art can think of other obvious deformations. The basic principles defined in the following description can be applied to other implementation schemes, deformation schemes, improvement schemes, equivalent schemes and other technical schemes without departing from the spirit and scope of the present invention.

[0087] It can be understood that the term "a" should be understood as "at least one" or "one or more". That is, in one embodiment, the number of one element can be one, while in other embodiments, the number of this element can be multiple. The term "a" cannot be understood as a limitation on the quantity.

[0088] Refer to Figure 1 , the technical solution provided by the present invention is: a method for identifying device status, faults and handling abnormalities in the disinfection supply center, including the following steps:

[0089] Step 1: Obtain a sterilization task, and collect the operation status data of different target devices at a preset frequency according to the sterilization task; the operation status data includes the pressure, temperature, sterilization time, humidity and gas concentration inside the device.

[0090] Step 2: Construct an abnormal response framework, analyze the obtained operation status data, and implement different response strategies according to the analysis results. Specifically, it includes the following steps:

[0091] Step 2.1: Obtain the operation status data, and after normalization processing, form an operation status feature sequence: , where respectively represent the device at Pressure, temperature, sterilization duration, humidity, and gas concentration at a moment;

[0092] Step 2.2: Construct an exception response framework, which includes an input layer, a preprocessing layer, a priority ranking layer, an information generation layer, and an output layer;

[0093] Step 2.3: Input the operating state feature sequence into the exception alarm framework for alarm feature extraction and encoding. Specifically:

[0094] The alarm levels are divided into first-level alarm and second-level alarm; the first-level alarm is that a single status data exceeds the preset safety threshold; the second-level alarm is that multiple status data exceed the preset safety threshold, or the device self-check fails. Specifically:

[0095] Alarm level ; In some embodiments, when the device cannot perform self-check or the self-check status is abnormal, the alarm level takes the value of 2.

[0096] Among them, represents the indicator function, and when the condition is satisfied, the value of the indicator function is 1, otherwise the value is 0;

[0097] Status parameter ; represents the th safety threshold of the status parameter;

[0098] Mark the alarm types, specifically including:

[0099] Generate a label list according to the type of exception status feature value;

[0100] Within the time window , count the number of devices triggering the same type of alarm ;

[0101] Read the current operating stage of the target device from the device log, and the current operating stage includes preheating, sterilization, and cooling.

[0102] Step 2.4: Calculate the alarm priority to determine the exception response order, specifically including:

[0103] Construct an alarm priority scoring model ;

[0104] Among them, respectively represent the weight coefficients, represents the alarm type weight; represents the alarm type, that is, the type of exception status feature value; represents the current operating stage weight; ; ;

[0105] Determine the order of abnormal responses according to the alarm priority.

[0106] For example, a secondary alarm for a single device, i.e., abnormal temperature and pressure, and the current operation stage of the device is the sterilization stage;

[0107] At this time, the alarm priority of this device:

[0108] .

[0109] In this embodiment, the types of alarms include too high temperature (the alarm type weight is taken as 1), abnormal pressure (the alarm type weight is taken as 0.8), and abnormal humidity (the alarm type weight is taken as 0.6); the current operation stages include the preheating stage (the current operation stage weight is taken as 0.3), the sterilization stage (the current operation stage weight is taken as 1), and the cooling stage (the current operation stage weight is taken as 0.5).

[0110] Step 2.5, Match the alarm information template and generate corresponding processing suggestions, including:

[0111] Pre-define the information template preprocessing suggestions under different scenarios, specifically:

[0112] Primary alarm template: [Device ID] occurred [Alarm type] <[Current state parameter value [x] and safety threshold [y]]> at [Time], it is recommended <[Suggested instruction information]>;

[0113] Secondary alarm template: [Device ID] occurred a compound fault <[Alarm type 1]+[Alarm type 2]> at [Time], stop sterilization, execute [Self-check protocol] and contact [Responsible person];

[0114] Step 2.6, Use natural language processing algorithms to optimize the instruction suggestion information; push the alarm information in levels and collect feedback information.

[0115] Among them, using natural language processing algorithms to optimize the instruction suggestion information includes:

[0116] Based on the fault, use the lightweight machine learning algorithm CRF to replace the placeholder in the alarm information template with real-time data, and the placeholders include [Time], [y], [x];

[0117] Combine the alarm type and device ID to optimize the sentence structure of the alarm information template.

[0118] For example, when a device has an abnormal temperature, use the primary alarm template, that is: Device A occurred too high temperature <[Current state parameter value 135°C and safety threshold 132°C]> at 13:00, it is recommended <[Check the temperature sensor and execute the cooling process]>;

[0119] Use the CRF algorithm to optimize the instruction suggestion:

[0120] First, for the action "inspection" corresponding to excessively high temperature, when the device type is a sterilizer, obtain the sterilization process from the database and match the sterilization process;

[0121] When a temperature anomaly occurs, the optimized instruction suggestion is: Please check the temperature sensor of the sterilizer, stop sterilization and execute the cooling process.

[0122] For example, when temperature anomalies and pressure anomalies occur during the sterilization stage of 3 sterilization devices, the alarm level is 2 at this time; the alarm priority is ;

[0123] Match the secondary template, the alarm priority is high priority, and the alarm information output by means of SMS and App push is that composite failures (temperature anomaly + pressure anomaly) occurred in sterilization devices A, B, and C at 8:00, stop sterilization, execute the self-inspection protocol and contact person in charge A.

[0124] Among them, hierarchical push of alarm information and collection of feedback information include:

[0125] Adopt different push methods based on the priority level:

[0126] If , the alarm priority is high priority, and it is pushed through SMS, App push and audible and visual alarms;

[0127] If , the alarm priority is medium priority, and it is alarmed through system pop-up windows and emails;

[0128] If , the alarm priority is low priority, write the alarm status into the log, notify manual review; regularly check the log to determine whether the alarm status is updated, and feedback the inspection result information to the host computer.

[0129] Step 3: Based on the sterilization task, obtain the biological monitoring data and chemical monitoring data of the corresponding target device, and judge whether the sterilization effect is qualified based on the chemical monitoring data and biological monitoring data. If so, return to Step 1; otherwise, enter Step 4, including the following steps:

[0130] Judge whether the sterilization effect is qualified based on the chemical monitoring data and biological monitoring data, specifically:

[0131] Obtain the status characteristic value of the biological monitoring package ;

[0132] Obtain the status characteristic value of the negative experimental group of biological monitoring ;

[0133] Obtain the status characteristic value of the chemical monitoring result ;

[0134] If and and , it is determined that the sterilization effect is qualified, that is ; otherwise, it is determined that the sterilization effect is unqualified, that is .

[0135] Step 4: Construct a data matrix using the equipment operation status data, biological monitoring data, and chemical monitoring data, and obtain the reasons for the unqualified sterilization effect based on the analysis of the data matrix. The biological monitoring data includes the status characteristics of the biological test package obtained from the biological monitoring instrument, and the status characteristics of the biological test package include that the biological indicator is negative and the biological indicator is positive.

[0136] It includes the following steps:

[0137] Step 4.1: Integrate the equipment operation data, chemical monitoring data, and biological monitoring data into a structured data matrix. Specifically: Obtain the operation status characteristic sequence; Collect the image data of the chemical indicator before and after sterilization;

[0138] By analyzing the collected image data of the chemical indicator before and after sterilization, obtain the discoloration status data of the chemical indicator;

[0139] Collect the biological monitoring result data from the biological monitoring instrument;

[0140] Align the operation status characteristic sequence, the discoloration status data of the chemical indicator, and the biological monitoring result data of the target equipment in the same sterilization batch on the time axis;

[0141] Obtain the operation status characteristic sequences of all equipment , and construct the monitoring characteristic sequence ; where respectively represent the identification ID of the th equipment, the pressure change value, the average sterilization temperature, the equivalent sterilization time, the average humidity, and the average gas concentration during the sterilization process of the same batch; respectively represent the status characteristic values of the biological monitoring package, the status characteristic values of the biological monitoring negative experimental group, the status characteristic values of several chemical monitoring, and the sterilization effect status value during the sterilization process of the same batch;

[0142] Construct the data matrix , where represents the number of equipment;

[0143] Average sterilization temperature where represents the total number of batches;

[0144] Pressure change value ; where Represents the average pressure of multiple batches;

[0145] Equivalent sterilization time ; where, Represents the end time of sterilization, and the temperature variable Is determined according to the reference bacterial sterilization kinetic index; for example, taking Bacillus stearothermophilus spores as an example, °C, °C.

[0146] Step 4.2. Obtain the reasons for unqualified sterilization effect through the analysis of the structured data matrix, specifically:

[0147] Obtain , extract multiple elements to form an input feature vector ;

[0148] Input the input feature vector into the logistic regression model to judge the unqualified reasons:

[0149] The logistic regression model is:

[0150] ;

[0151] Where, Respectively represent the influence weight of the equivalent sterilization time, the influence weight of the pressure change on unqualified, the influence weight of insufficient temperature on unqualified, the influence weight of unqualified chemical monitoring, and the interaction effect weight of temperature and chemical monitoring; Represents the unqualified probability;

[0152] Use the maximum likelihood estimation MLE algorithm to solve the logistic regression model and output And the significance value, that is, the p-value;

[0153] If the significance value corresponding to each influence weight is less than the preset significance threshold, it is judged that the influence of this influence weight on the unqualified sterilization effect is not significant;

[0154] Find the significance value with the largest numerical value, and the corresponding feature is the main reason for the unqualified sterilization, and focus on optimizing its corresponding feature.

[0155] For example, The corresponding p-values are 0.45, 0.3, 0.2, 0.05. It can be seen that the equivalent sterilization time has the greatest influence and is solved by extending the sterilization time;

[0156] The important influence of pressure fluctuation is the second, optimize the equipment pressure control module to reduce the pressure fluctuation.

[0157] For example, the biological monitoring of a certain batch of sterilization is unqualified, and the chemical monitoring shows compliance;

[0158] The average sterilization temperature of the equipment is 128 °C (lower than the safety threshold temperature of 132 °C), the pressure fluctuation is 0.25 bar (exceeding the pressure fluctuation threshold of 0.1 bar), and the equivalent sterilization time is 5.2 minutes (less than the equivalent sterilization threshold of 6 minutes); the alarm level at this time is 2 because two status parameters are abnormal (temperature, pressure);

[0159] Through the analysis of the logistic regression model, , the corresponding p-value is 0.04, , the corresponding p-value is 0.2, and the significance of both is relatively obvious. It is judged that the insufficient average sterilization temperature (low temperature) leads to unqualified biological monitoring of sterilization, and at the same time, the misjudgment of chemical monitoring is superimposed, resulting in unqualified sterilization effect.

[0160] It is solved by calibrating the temperature sensor and replacing the chemical indicator card.

[0161] Among them, the state characteristics of the chemical indicator are obtained through the analysis of the image data, and the specific steps are as follows:

[0162] Collect high-definition images of the chemical indicator before and after sterilization, mark the indicator area in the image, and generate an XML or JSON annotation file;

[0163] Based on the annotation, the indicator area is cropped through the ROI algorithm;

[0164] Convert the image of the indicator area into the HSV space to enhance color sensitivity;

[0165] Input the images of the indicator area before and after sterilization into the CNN network, and output the chemical monitoring compliance status;

[0166] The CNN network includes an input layer, a fully connected layer, and an output layer. The input layer is used to receive the preprocessed image of the indicator area, and the fully connected layer is used for indicator area cropping and discoloration degree analysis; the output layer is used to output the classification result, that is, chemical monitoring compliance or chemical monitoring non-compliance.

[0167] The present invention also provides a computer-readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the method for the disinfection supply center to identify the equipment status, faults, and handle abnormalities.

[0168] Embodiments disclosed in the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. Embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for performing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part, and / or installed from a removable medium. When the computer program is executed by a central processing unit (CPU), the above functions defined in the methods of the present invention are performed. It should be noted that the computer-readable medium in the present invention can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. The computer-readable storage medium can, for example, but is not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection having one or more wire segments, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or combined with an instruction execution system, apparatus, or device. In the present invention, the computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, which carries the computer-readable program code. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium can also be any computer-readable medium other than the computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or combined with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wireless segments, wire segments, optical cables, RF, etc., or any suitable combination of the above.

[0169] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur in a different order than that noted in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0170] Those skilled in the art should understand that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The objectives of the present invention have been fully and effectively achieved. The functions and structural principles of the present invention have been demonstrated and explained in the embodiments. Without departing from the said principles, any changes or modifications may be made to the embodiments of the present invention.

Claims

1. A method for identifying the status, faults and handling abnormalities of equipment used in the disinfection supply center, characterized in that, The method includes: Step 1: Obtain a sterilization task, and collect operation status data of different target devices at a preset frequency based on the sterilization task; Step 2: Construct an exception response framework, analyze the obtained operation status data, and implement different response strategies according to the analysis results, including: Obtain the operating status data, and after normalization, form an operating status feature sequence: , where respectively represent the pressure, temperature, sterilization duration, humidity, and gas concentration of the device at the moment of pressure, temperature, sterilization duration, humidity, and gas concentration; Step 3: Based on the sterilization task, obtain biological monitoring data and chemical monitoring data of the corresponding target device, and judge whether the sterilization effect is qualified based on the chemical monitoring data and the biological monitoring data. If so, return to Step 1; otherwise, enter Step 4; Step 4: Use the device operation status data, biological monitoring data, and chemical monitoring data to construct a data matrix, and obtain the reasons for the unqualified sterilization effect based on the analysis of the data matrix, including: Integrate the device operation data, chemical monitoring data, and biological monitoring data into a structured data matrix, including: Obtain the operation status feature sequence; collect image data of chemical indicators before and after sterilization; Obtain the color change status data of the chemical indicator by analyzing the collected image data of the chemical indicator before and after sterilization; Collect biological monitoring result data from a biological monitoring instrument; Align the operation status feature sequence, the color change status data of the chemical indicator, and the biological monitoring result data of the target device in the same sterilization batch on the time axis; Obtain the operation status feature sequences of all devices , and construct the monitoring feature sequences ; among them, respectively represent the identification ID of the th device, the pressure change value, average sterilization temperature, equivalent sterilization time, average humidity, and average gas concentration during the sterilization process of the same batch; respectively represent the state characteristic values of the biological monitoring package, the state characteristic values of the biological monitoring negative experimental group, the state characteristic values of chemical monitoring, and the sterilization effect state values during the sterilization process of the same batch; Construct a data matrix , where represents the number of devices; Obtain the reasons for the unqualified sterilization effect through the analysis of the structured data matrix, including: Obtain , extract multiple elements to form an input feature vector ; Input the input feature vector into a logistic regression model to judge the unqualified reasons.

2. The method for identifying the status, faults and handling abnormalities of the equipment used in the disinfection supply center according to claim 1, characterized in that The operation status data includes the pressure, temperature, sterilization time, humidity, and gas concentration inside the device; The chemical monitoring data includes the image data of chemical indicators before and after sterilization, and obtains the state characteristics of the chemical indicator by analyzing the image data; The biological monitoring data includes the state characteristics of a biological test package obtained from a biological monitoring instrument, and the state characteristics of the biological test package include that the biological indicator is negative and the biological indicator is positive.

3. The method for identifying the status, faults and handling abnormalities of the equipment in the disinfection supply center according to claim 2, wherein, The construction of the exception response framework, the analysis of the obtained operation status data, and the implementation of different response strategies according to the analysis results include: The analysis of the obtained operation status data and the implementation of different response strategies according to the analysis results include: Input the operation status feature sequence into an exception alarm framework for alarm feature extraction and encoding; Calculate the alarm priority to determine the exception response order; Match the alarm information template to generate corresponding processing suggestions; Optimize the instruction suggestion information using natural language processing algorithms; Push the alarm information at different levels and collect feedback information.

4. The method for identifying the status, faults and handling abnormalities of the equipment in the sterile supply center according to claim 3, characterized in that, The alarm feature extraction and encoding include: Divide the alarm levels into first-level alarms and second-level alarms; the first-level alarm is that a single status data exceeds a preset safety threshold; the second-level alarm is that multiple status data exceed a preset safety threshold, or the device self-check fails. Specifically: Alarm level ; Among them, represents an indicator function. When the condition is satisfied, the value of the indicator function is 1; otherwise, the value is 0. Status parameter ; Indicates the safety threshold of the th status parameter; Mark the alarm types, including: Generate a label list according to the type of the abnormal status characteristic value; Count the number of devices that trigger the same type of alarm within the time window ; Read the current operation stage of the target device from the device log, and the current operation stage includes preheating, sterilization, and cooling; The calculation of the alarm priority to determine the alarm response order includes: Construct an alarm priority scoring model ; Among them, respectively represent weight coefficients, represents the weight of the alarm type; represents the type of the alarm type, that is, the type of the abnormal state eigenvalue; represents the weight of the current operation stage; ; ; Determine the exception response order according to the alarm priority.

5. The method for identifying the status, faults and handling abnormalities of the equipment in the disinfection supply center according to claim 4, characterized in that, Match the alarm information template and generate corresponding processing suggestions, including: Pre-define information template pre-processing suggestions for different scenarios, including: Level 1 alarm template: [Device ID] had an [Alarm type] at [Time] <Current status parameter value [x] and safety threshold [y]>, suggest <Suggested instruction information>; Level 2 alarm template: [Device ID] had a compound fault <[Alarm type 1]+[Alarm type 2]> at [Time], stop sterilization, execute [Self-check protocol] and contact [Responsible person]; Optimize the instruction suggestion information using natural language processing algorithms, including: Based on the fault, use the lightweight machine learning algorithm CRF to replace the placeholders in the alarm information template with real-time data, and the placeholders include [Time], [y], [x]; Combine the alarm type and device ID to optimize the sentence structure of the alarm information template; Push alarm information and collect feedback information in a hierarchical manner, including: Adopt different push methods based on the priority level: If , the alarm priority is high, and it alarms through text messages, App push, and audible and visual alarms; If , the alarm priority is medium priority, and the system will pop up a window and send an email for alarm; If , the alarm priority is low priority. Write the alarm status to the log and notify manual review; Check the log to determine whether the alarm status is updated and give feedback.

6. The method for identifying the status, faults and handling abnormalities of the equipment used in the disinfection supply center according to claim 5, characterized in that, Integrate the device operation data, chemical monitoring data, and biological monitoring data into a structured data matrix, including: Average sterilization temperature Wherein, represents the total number of batches; Pressure change value ; wherein, represents the average pressure of multiple batches; Equivalent sterilization time ; wherein, represents the end time of sterilization, and the temperature variable is determined according to the reference bacterial sterilization kinetic index.

7. The method for identifying the status, faults and handling abnormalities of the equipment in the disinfection supply center according to claim 6, characterized in that, Judge whether the sterilization effect is qualified based on the chemical monitoring data and biological monitoring data, including: Obtain the status characteristic value of the biological monitoring package ; Obtain the status characteristic values of the negative experimental group for biological monitoring ; Obtain the status characteristic value of the chemical monitoring result ; If and and , it is determined that the sterilization effect is qualified, that is ; otherwise, it is determined that the sterilization effect is unqualified, that is ; The logistic regression model is: ； Among them, respectively represent the influence weight of the equivalent sterilization time, the influence weight of the pressure change on nonconformity, the influence weight of insufficient temperature on nonconformity, the influence weight of unqualified chemical monitoring, and the interaction effect weight of temperature and chemical monitoring; represents the nonconformity probability; Use the maximum likelihood estimation (MLE) algorithm to solve the logistic regression model and output and the significance value, i.e., the p-value; If the significance value corresponding to each influence weight is less than the preset significance threshold, it is judged that the influence of this influence weight on the unqualified sterilization effect is not significant; Find the significance value with the largest numerical value, and the corresponding feature is the main reason for the unqualified sterilization, and focus on optimizing the corresponding feature.

8. The method for identifying the status, faults and handling abnormalities of the equipment in the disinfection supply center according to claim 7, characterized in that, Obtain the color change status data of the chemical indicator by analyzing the image data of the chemical indicator before and after sterilization, including: Collect the RGB images of the chemical indicator before and after sterilization, mark the indicator area in the RGB image, and generate an XML or JSON annotation file; Based on the annotation, crop the indicator area through the ROI algorithm; Convert the image of the indicator area to the HSV space to enhance color sensitivity; Input the images of the indicator area before and after sterilization into the CNN network and output the chemical monitoring compliance status; The CNN network includes an input layer, a fully connected layer, and an output layer. The input layer is used to receive the pre-processed image of the indicator area, and the fully connected layer is used for indicator area cropping and color change degree analysis; the output layer is used to output the classification result, that is, chemical monitoring compliance or chemical monitoring non-compliance.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, and the computer program is executed by a processor to implement the method for identifying the status, faults, and handling abnormalities of the disinfection supply center device described in any one of claims 1-8 above.

Images