Big data-based central sterile supply department instrument whole-process tracing system

Abstract

The application discloses a kind of based on big data's disinfection supply center instrument whole-process tracing system, it is related to medical instrument sterilization monitoring technical field, can solve how to accurately perceive the actual state in the drying process of pressure steam sterilizer inside instrument package non-invasively and intelligently trace abnormality problem, comprising: data acquisition module is used to obtain the power supply current data of pressure steam sterilizer in drying stage;Event encapsulation module is used to determine discrete heating event data and discrete pump load event data according to power supply current data;State evaluation module is used to calculate drying state index set according to the time sequence association between discrete heating event data and discrete pump load event data;Decision tracing module is used to judge whether instrument package reaches drying standard based on drying state index set, and in the case where instrument package does not reach drying standard, the reason that does not reach drying standard is classified and traced.

Description

Technical Field

[0001] This invention relates to the field of medical device sterilization monitoring technology, specifically to a big data-based full-process traceability system for medical devices in a disinfection supply center. Background Technology

[0002] Pressure steam sterilization is a crucial step in ensuring the sterility of medical devices in hospital sterilization supply centers, and the effectiveness of its drying stage directly determines the quality and safety of sterile packs. Thoroughly removing residual moisture from the device packs during the sterilization process is essential to prevent microbial growth and subsequent storage; therefore, reliable monitoring and quality control of the drying process are a vital foundation of the hospital's infection control system.

[0003] Currently, monitoring of the drying stage generally relies on preset fixed-time programs or simple threshold judgments of vacuum pump operating current. These methods can only indirectly reflect whether the equipment is operating, but cannot deeply perceive the actual degree of dryness inside the instrument pack or the true state of energy transfer efficiency. This leads to problems of insufficient or excessive drying in actual operation, which not only poses the risk of infection caused by wet packs, but also wastes energy, and it is difficult to accurately trace the root cause once a problem occurs. Summary of the Invention

[0004] To address the current technical challenge of how to non-invasively and accurately sense the actual state inside instrument packs during the drying process in a pressure steam sterilizer and intelligently trace any abnormalities, this invention aims to provide a big data-based end-to-end traceability system for instruments in a sterilization supply center. The specific technical solution adopted is as follows: In a first aspect, the present invention provides a big data-based end-to-end traceability system for medical devices in a sterilization supply center, comprising: a data acquisition module for acquiring power supply current data of a pressure steam sterilizer during the drying stage; an event encapsulation module for determining discrete heating event data and discrete pump load event data based on the power supply current data; wherein, the discrete heating event data characterizes independent events that convert electrical energy into heat energy and input it into the sterilization chamber, and the discrete pump load event data characterizes independent response events in which the vacuum pump performs work to remove gas from the chamber; a state assessment module for calculating a set of drying state indicators based on the temporal correlation between the discrete heating event data and the discrete pump load event data; wherein, the set of drying state indicators characterizes the physical state of energy transfer and conversion efficiency within the medical device pack during the drying process; and a decision traceability module for determining whether the medical device pack meets the drying standard based on the set of drying state indicators, and, if the medical device pack does not meet the drying standard, classifying and tracing the reasons for not meeting the drying standard.

[0005] Secondly, this invention provides a big data-based method for full-process traceability of medical devices in a sterilization supply center, comprising: acquiring power supply current data of a pressure steam sterilizer during the drying stage; determining discrete heating event data and discrete pump load event data based on the power supply current data; wherein, discrete heating event data is used to characterize independent events that convert electrical energy into heat energy and input it into the sterilization chamber, and discrete pump load event data is used to characterize independent response events in which the vacuum pump performs work to remove gas from the chamber; calculating a drying state index set based on the temporal correlation between the discrete heating event data and the discrete pump load event data; wherein, the drying state index set is used to characterize the physical state of energy transfer and conversion efficiency inside the medical device pack during the drying process; and determining whether the medical device pack meets the drying standard based on the drying state index set, and classifying and tracing the reasons for not meeting the drying standard if the medical device pack does not meet the drying standard.

[0006] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer execution instructions, and when the electronic device is running, the processor executes the computer execution instructions stored in the memory to cause the electronic device to perform the big data-based full-process traceability method for sterilization supply center equipment as described in the second aspect and any possible implementation of the second aspect.

[0007] This invention offers the following advantages: By constructing a discrete event time-series correlation model of electrothermal input and pneumatic output, and innovatively utilizing non-invasively acquired total power supply current of the sterilizer, it achieves precise, dynamic perception and intelligent closed-loop control of the actual drying state inside the instrument pack during the pressure steam sterilizer drying process. This solution fundamentally solves the technical problems of misjudging deep wet pack risks, energy waste, and difficulty in tracing the root causes of anomalies caused by traditional methods relying on fixed time or simple current thresholds. By decoupling and encapsulating continuous current signals into independent energy input and response events, and then calculating the microscopic physical characteristics characterizing phase transition hysteresis and latent heat erosion, the system can quantify changes in energy transfer efficiency within the chamber. This allows for the determination of the drying endpoint at a physical level, and intelligent differentiation of different abnormal causes such as overloading, pipeline blockage, or mechanical failure. Ultimately, it forms a complete digital traceability closed loop that includes objective physical evidence and supports quality backtracking, significantly improving the safety of sterilization packs and the level of intelligence in supply center management. Attached Figure Description

[0008] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the architecture of a big data-based full-process traceability system for medical instruments in a disinfection supply center, provided as an embodiment of the present invention. Figure 2 This is a schematic diagram of the architecture of an event encapsulation module provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of the architecture of a state assessment module provided in one embodiment of the present invention; Figure 4 This is a flowchart illustrating a big data-based method for tracing the entire process of medical instruments in a disinfection supply center, as provided in one embodiment of the present invention. Detailed Implementation

[0010] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0011] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0012] The specific solution of the big data-based full-process traceability system for medical instruments in a sterilization supply center provided by this invention is described below with reference to the accompanying drawings. Under the monitoring of this system, it is recommended or controlled that the pressure steam sterilizer operate in a low duty cycle pulse heating mode during the drying stage, and the interval between individual heating pulses should be greater than the associated time window. This ensures that each heating event triggers an independent and distinguishable aerodynamic response event, facilitating precise timing matching of the system.

[0013] For example, such as Figure 1The diagram shown is an architectural representation of a big data-based traceability system for medical instruments in a disinfection supply center (hereinafter referred to as the traceability system) according to an embodiment of the present invention. The traceability system 10 includes: a data acquisition module 11, an event encapsulation module 12, a status assessment module 13, and a decision traceability module 14. The modules are described below in sequence: (1) Data acquisition module 11.

[0014] The data acquisition module 11 is responsible for physically connecting to the power supply circuit of the pressure steam sterilizer, and collecting the power supply current data of the entire cycle during the drying stage in real time, providing the most original electrical signal source for all subsequent analyses.

[0015] Optionally, the data acquisition module 11 is used to acquire the power supply current data of the pressure steam sterilizer during the drying stage.

[0016] Among them, the power supply current data refers to the continuous current waveform collected in real time from the main power supply line of the pressure steam sterilizer. This data is essentially the total operating current of the sterilizer during the drying stage. Its waveform changes comprehensively reflect the on-off heating state of the heating tube (pure resistive load) and the pumping work state of the vacuum pump (motor inductive load). It is the physical carrier of the complete energy input and output information of the equipment.

[0017] Specifically, the data acquisition module 11 can be implemented using a high-precision current sensor (such as a Hall effect current sensor) installed on the main power supply line of the sterilizer. This sensor continuously acquires the current waveform at a preset high sampling frequency (e.g., thousands of times per second) to ensure that it can capture rapid current fluctuations caused by the on / off switching of the heating tube and changes in the load of the vacuum pump.

[0018] Finally, the collected continuous time series current data is uploaded to the event encapsulation module 12 in real time, serving as the starting point for the entire intelligent traceability process.

[0019] (2) Event encapsulation module 12.

[0020] The event encapsulation module 12 is responsible for receiving the raw power supply current data from the data acquisition module 11 and converting the continuous current waveform into a discrete event sequence with clear physical meaning, namely discrete heating event data and discrete pump load event data.

[0021] Optionally, the event encapsulation module 12 is used to determine discrete heating event data and discrete pump load event data based on the power supply current data. The discrete heating event data characterizes independent events that convert electrical energy into heat energy and input it into the sterilization chamber, while the discrete pump load event data characterizes independent response events where the vacuum pump performs work to remove gas from the chamber.

[0022] It should be explained that discrete heating event data and discrete pump load event data are two core data objects defined in this system. These two types of discrete event data decouple the continuous equipment operating current into a series of micro-events that occur sequentially on the time axis and have clear energy attributes, providing standardized data units for subsequently establishing accurate input and response pairs.

[0023] Discrete heating event data is the identification of each independent heating tube energizing cycle from the heating current sequence separated from the mixed current. Each such data object records a complete energy input event, with its core attribute "single input energy" quantifying the total electrical energy consumed within that cycle (i.e., the total amount converted into heat energy), and "input duration" recording the duration of the heating tube's operation.

[0024] Discrete pump load event data is identified from the vacuum pump current sequence as current fluctuations caused by increased load due to the extraction of gas from the chamber. Each such data object records a corresponding aerodynamic response event, with its core attribute "work done in a single response" quantifying the additional mechanical energy consumed by the vacuum pump to remove that portion of gas, and "response duration" characterizing the width of the load fluctuation signal.

[0025] For example, such as Figure 2 As shown, the event encapsulation module 12 can be further divided into a signal separation submodule 121, a heating event encapsulation submodule 122, and a pump event encapsulation submodule 123 to achieve a refined processing flow. Furthermore, when the event encapsulation module 12 determines discrete heating event data and discrete pump load event data based on the power supply current data, the aforementioned submodules implement this through the following steps: S101. Based on the supply current data, determine the reference current and reference conversion parameters. The reference current characterizes the basic power consumption required for the vacuum pump to maintain operation, and the reference conversion parameters characterize the efficiency of converting thermal energy into mechanical work.

[0026] It is understandable that the reference current is specifically the steady-state current consumed by the vacuum pump when maintaining the basic vacuum level. The reference conversion parameter is specifically a dimensionless proportionality coefficient, reflecting the theoretical efficiency of converting the thermal energy input to the system into the mechanical work of the vacuum pump under ideal dry conditions.

[0027] For example, the signal separation submodule 121 performs parameter initialization during the pre-vacuum stage (when the sterilization chamber is empty) before each sterilization procedure enters the formal drying stage. This specifically includes two sub-steps: a) Determine the pump maintaining current baseline: Collect the total power supply current of the sterilizer during the period when the vacuum pump operates independently to maintain the basic vacuum level. The arithmetic mean of the current during this period is calculated using a moving average filter and defined as the pump sustaining current reference. (That is, the reference current). This parameter represents the inherent current required for the equipment to overcome mechanical friction and the inherent resistance of the pipeline in the current batch, and is specifically used in S103 to determine the current amplitude threshold.

[0028] b) Determine the standard energy conversion ratio: Monitor the first trial heating pulse during the pre-vacuum stage. Record the electrical energy consumed by this pulse as the reference input, and record the mechanical work generated by the subsequent change in vacuum pump load due to the thermal expansion of the air inside the chamber as the reference output. Calculate the ratio of the reference output to the reference input, defined as the standard energy conversion ratio α. This parameter is only valid for the current sterilization batch and is used as a normalized benchmark in subsequent calculations to determine whether energy has been consumed by the latent heat of moisture, specifically applied in subsequent S202.

[0029] S102. Based on the current characteristics of different types of loads in the power supply current data, the power supply current data is divided into heating current sequence and vacuum pump current sequence.

[0030] Optionally, the signal separation submodule 121 divides the power supply current data into a heating current sequence and a vacuum pump current sequence through the following steps: a) Perform frequency domain analysis on the power supply current data, extract the active current component that is in phase with the power supply voltage, and determine the active current component as the heating current sequence.

[0031] Specifically, the signal separation submodule 121 uses short-time Fourier transform to analyze the total power supply current of the sterilizer. Time-frequency decomposition is performed. Since the heating element is a purely resistive load, its current waveform is in phase with the voltage waveform; however, the vacuum pump is an inductive motor load, and its current waveform exhibits phase lag and contains higher harmonics. The system sampling window is set to 20ms (corresponding to one cycle of a 50Hz power frequency). Within each window, the system extracts the active component at the fundamental frequency that is in phase with the voltage and reconstructs it into a heating current sequence.

[0032] b) Remove the heating current sequence from the power supply current data to obtain the vacuum pump current sequence.

[0033] Furthermore, the signal separation submodule 121 receives the total current. Subtract the heating current sequence from the middle part, and reconstruct the remaining part (including the fundamental reactive component and the higher harmonic components) into the vacuum pump current sequence.

[0034] S103. Perform event identification and parameter extraction on the heating current sequence to generate discrete heating event data. The discrete heating event data shall include at least the single input energy and the input duration.

[0035] For example, the heating event encapsulation submodule 122 performs the following operations in order to identify events and extract parameters from the heating current sequence: a) Determine the current amplitude threshold based on the reference current, identify current pulse segments that are greater than the current amplitude threshold from the heating current sequence, and determine each identified current pulse segment as a heating event.

[0036] First, the heating event encapsulation submodule 122 sets the trigger threshold to the pump sustaining current reference. 1.2 times that of the signal separation submodule. Afterwards, this submodule monitors in real time the heating current sequence output by the signal separation submodule 121. , where t represents the time variable. When When the instantaneous value rises and exceeds this threshold, it is marked as the start time of a heating event. , where n is the index of the heating event; when The moment when the instantaneous value falls below this threshold is marked as the end of the event. To filter out interference, only current pulse segments lasting longer than 200 milliseconds are identified as valid heating events.

[0037] b) For each heating event, calculate the single input energy based on the power of the heating current during the duration of the heating event, and use the duration of each heating event as the input duration.

[0038] Furthermore, the heating event encapsulation submodule 122 encapsulates each identified heating event. For the nth event, the time difference between its end time and start time is calculated to obtain the heating duration of the event. The duration is used as the input. Simultaneously, the voltage during the duration of this event is... With heating current Integrate the product of the two components, and use the result as the single-input thermal energy of the event. .Include and The data structure is encapsulated as a thermal input pulse data. This refers to discrete heating event data. Its mathematical relationship is: Here, This is the instantaneous voltage value collected from the main power supply line of the sterilizer.

[0039] It should be pointed out that, This refers to the instantaneous value of the power supply voltage collected in real time, or the rated voltage value when the voltage is stable.

[0040] S104. Perform event identification and parameter extraction on the vacuum pump current sequence to generate discrete pump load event data. The discrete pump load event data shall include at least the work done in a single response and the response duration.

[0041] For example, when the pump event encapsulation submodule 123 performs event identification and parameter extraction on the vacuum pump current sequence, it specifically performs the following operations: a) Determine the AC fluctuation sequence from the vacuum pump current sequence, identify the peaks greater than the preset fluctuation threshold from the AC fluctuation sequence, and determine the fluctuation corresponding to each identified peak as a pump load event.

[0042] Specifically, the pump event encapsulation submodule 123 first processes the vacuum pump current sequence. A high-pass digital filter is applied to remove the DC component, resulting in an AC fluctuation sequence characterizing the load variation. Subsequently, peak detection is performed on the AC fluctuation sequence. When the amplitude of the detected peak exceeds the preset fluctuation detection threshold, the complete current fluctuation corresponding to that peak is identified as a valid pump load event, and its peak time is recorded. It should be noted that the complete current fluctuation corresponding to the peak refers to the current fluctuation range between two points centered on the identified peak and extending to both sides until the amplitude drops to half of the peak amplitude, which is determined as the complete current fluctuation corresponding to a single pump load event.

[0043] For example, a fluctuation detection threshold is used to distinguish between effective load events of the vacuum pump and background noise. Its value can be derived statistically from current data during no-load operation of the equipment. Specifically, under stable operating conditions with no load in the sterilizer during the pre-vacuum phase, an AC fluctuation sequence of the vacuum pump current is collected. The initial threshold is calculated by analyzing the statistical characteristics of the sequence amplitude (e.g., taking 2 to 3 times its root mean square value, or a certain percentage of the peak value, such as 50%). This threshold can be fine-tuned during system deployment; an empirical range is 1% to 5% of the vacuum pump's rated operating current. The core of setting this rule is to ensure that the threshold is higher than the amplitude of most electrical and mechanical background noise, while being lower than the amplitude of typical pneumatic load fluctuations caused by actual heating events, thereby reliably detecting effective pump load events.

[0044] b) For each pump load event, calculate the work done in a single response based on the power of the AC fluctuation during the duration of the pump load event, and take the half-width of the fluctuation waveform of each pump load event as the response duration.

[0045] Furthermore, the pump event encapsulation submodule 123 encapsulates each identified pump load event. For the m-th event, the full width at half maximum (FWHM) of its current fluctuation waveform is measured to obtain the response duration of the event. Meanwhile, the voltage during the duration of the event... With alternating current fluctuation Integrate the product of the two factors and calculate the result as the work done in a single response of the event. .Include , and its peak time The data structure is encapsulated as a pump load response data. This refers to discrete pump load event data. Its mathematical relationship is: Here, Both are instantaneous voltage values ​​collected from the main power supply line of the sterilizer.

[0046] Thus, the event encapsulation module 12 completes a fundamental transformation from continuous, mixed analog current signals to discrete, semantically clear digital event models. This transformation process not only decouples the macroscopic equipment operating current into two independent event flows: microscopic "energy input" and "aerodynamic response," but more importantly, it lays the data structure foundation for subsequently establishing accurate temporal causal relationships and quantifying energy transfer efficiency. This makes it possible to understand the complex physical drying process inside the chamber through non-invasive current analysis.

[0047] (3) Status assessment module 13.

[0048] The state assessment module 13 is responsible for receiving two discrete event data streams from the event encapsulation module 12, establishing the temporal correlation between them and performing feature calculations, and finally obtaining a set of drying state indicators that can comprehensively quantify the physical state of the drying process.

[0049] Optionally, the state assessment module 13 is used to calculate a set of drying state indicators based on the temporal correlation between discrete heating event data and discrete pump load event data. The set of drying state indicators characterizes the physical state of energy transfer and conversion efficiency within the instrument pack during the drying process.

[0050] Specifically, the drying state index set is a core state vector defined by this system, used to digitally characterize the real-time drying process within the sterilization chamber. It consists of three key statistical indicators: average duration extension ratio... This is used to quantify the time broadening of the gas generation process; a higher value indicates a more significant phase transition hysteresis caused by the evaporation of residual moisture; average energy loss ratio This is used to quantify the proportion of input heat energy that is absorbed by the latent heat of water and fails to be converted into mechanical work; a higher value indicates that the energy is more severely "swallowed up"; no-response input ratio. This is used to quantify the proportion of ineffective energy input that fails to elicit any aerodynamic response after heating; an elevated value directly indicates potential severe water accumulation or tubing blockage. These three indicators together form a multi-dimensional state observation window, allowing the previously invisible dryness state inside the instrument pack (whether it's rapid expansion of dry air, slow evaporation of moisture, or condensation blockage) to be clearly quantified and tracked.

[0051] This transformation process enables the system to move beyond monitoring the actions of a single device and instead gain a continuous understanding of the phase change state and thermodynamic efficiency of the media (air and moisture) inside the chamber from a physical perspective, providing stable and objective data for accurately judging the degree of dryness.

[0052] For example, such as Figure 3 As shown, the state assessment module 13 may include a time-series matching submodule 131, a feature calculation submodule 132, and an index aggregation submodule 133, to respectively complete event association, feature extraction, and state aggregation. Furthermore, when calculating the dryness state index set, the state assessment module 13 is implemented by the aforementioned submodules performing the following steps: S201. Based on the preset associated time window, match the corresponding discrete pump load event data for each discrete heating event data to obtain input and response data pairs.

[0053] Understandably, the purpose of this step is to establish a causal mapping between heat input and aerodynamic response. Since there is a physical delay in the transfer of heat energy to the instrument pack and the resulting moisture vaporization, a "hysteresis observation" strategy is required for matching.

[0054] a) For each discrete heating event data, search for discrete pump load event data within the associated time window after the start time.

[0055] Specifically, the timing matching submodule 131 sets an associated time window. The length of this window This is a preset value based on the thermal conductivity of the sterilizer chamber and the maximum allowable loading capacity; in this embodiment, it is set to 15 seconds. For the nth thermal input pulse data generated in S103... (where n is the index of the heating event), and its starting time is The system is in After Within seconds, iterate through all generated pump load response data. Search for its peak time Data falling within this range.

[0056] b) If at least one discrete pump load event data is found within the associated time window, then select the discrete pump load event data from the searched discrete pump load event data that has the smallest time difference between the peak time and the start time of the discrete heating event data.

[0057] In this case, if in the window One or more pump load response data were found within the search. The timing matching submodule 131 will calculate each peak time With the start time of the heating event The system executes comparison logic to select the pump load response data with the smallest time difference.

[0058] c) The selected discrete pump load event data is determined as the corresponding discrete pump load event data to form input and response data pairs.

[0059] Specifically, the timing matching submodule 131 will use the pump load response data selected in the previous step. , with the current thermal input pulse data Pairing is performed. This pair of paired data is identified as a causally related input-response data pair, denoted as... =( , This input and response data pair This provides support for subsequent calculations of microscopic features.

[0060] d) If no discrete pump load event data is found within the associated time window, the current discrete heating event data is marked as an unresponsive input event.

[0061] If in the associated time window If no pump load response data is found, it indicates that the energy of this heating input has been completely absorbed (e.g., for the latent heat of phase change of moisture), and no gas expansion sufficient to trigger a change in vacuum pump load has been generated. The timing matching submodule 131 will then use the current heat input pulse data... This is marked as "No-response hot input". This marking is the direct basis for subsequent statistics on the percentage of no-response inputs.

[0062] S202. Calculate the duration extension characteristic and energy loss characteristic for each input-response data pair. The duration extension characteristic is used to characterize the time lag of the aerodynamic response relative to the thermal energy input, and the energy loss characteristic is used to characterize the proportion of energy loss during the energy transfer process.

[0063] Understandably, the purpose of this step is to extract the microscopic features of the physical state of quantification of drying from each successfully matched data pair. The duration extension feature reflects the waveform broadening caused by the phase transition of the medium, and the energy loss feature reflects the proportion of energy "swallowed" by the latent heat of moisture.

[0064] a) Calculate the duration extension characteristics based on the response duration of discrete pump load event data and the input duration of discrete heating event data.

[0065] For the k-th input-response data pair This includes thermal input pulse data and pump load response data. The feature calculation submodule 132 calculates extended features over time. For example, the calculation formula is as follows: in, This represents the duration extension ratio (duration extension feature) of the k-th data pair. This represents the paired pump load response data. Response duration Represents paired heating pulse data The duration of heating, This represents a correction factor that is a very small positive number, such as 10 to the power of negative 5, to prevent the computer program from crashing or generating invalid values ​​due to a zero denominator or zero input, thereby ensuring the robustness and feasibility of the algorithm under extreme conditions.

[0066] Using the above formula, the ratio obtained by dividing the duration of the pump response by the duration of heating is used to quantify the time broadening factor of the pneumatic response waveform relative to the heating pulse waveform. If the medium is dry air, this ratio should be close to 1; if there is a liquid water phase change, the vaporization process will smooth and broaden the response waveform, resulting in a ratio significantly greater than 1.

[0067] b) Obtain the reference transformation parameters, and calculate the energy loss characteristics based on the reference transformation parameters, the single input energy of the discrete heating event data in the input and response data pair, and the single response work of the discrete pump load event data in the input and response data pair.

[0068] First, the feature calculation submodule 132 obtains the standard energy efficiency conversion ratio α (i.e., the baseline conversion parameter) determined during the pre-vacuum phase from the system initialization module. Then, it calculates the energy loss characteristics. For example, the calculation formula is as follows: in, This represents the energy loss ratio (energy loss characteristic) of the k-th data pair. This represents the paired pump load response data. The work done in a single response, This represents the standard energy efficiency conversion ratio (also known as the baseline conversion parameter). Represents paired heating pulse data The single input of thermal energy.

[0069] After the above calculations, the feature calculation submodule 132 uses the benchmark transformation parameters. Regarding the input heat energy After correction, the theoretical work done in the response under ideal dry air conditions is obtained. Then, the work done in the measured response is calculated. The ratio to this theoretical value. Finally, subtracting this ratio from 1 gives the proportion of the input energy that was not converted into effective mechanical work—the "loss" portion. The closer this value is to 1, the more energy is absorbed by the latent heat of water.

[0070] Furthermore, for heating events marked as non-responsive input events in S201, since the response waveform cannot be extracted for calculation, the feature calculation submodule 132 directly assigns a saturation penalty value: its duration extension ratio. Set to a preset maximum upper limit value (e.g., 5.0), its energy loss ratio Set to 1.0. This ensures that silent energy absorption caused by severe wet packs or blockages can be identified as a high-risk condition in subsequent statistics.

[0071] S203. Based on the total duration extension characteristics, energy loss characteristics, and the proportion of discrete heating event data that failed to match the corresponding discrete pump load event data within the preset statistical period, a set of drying status indicators is generated.

[0072] Understandably, the purpose of this step is to smooth and aggregate a large number of transient micro-features to generate a set of macro-indicators that can robustly reflect the current overall dryness state in order to support decision-making.

[0073] Specifically, the indicator aggregation submodule 133 sets a sliding time window of 60 seconds as the preset statistical period. Within this window, the indicator aggregation submodule 133 performs the following operations: Aggregate micro-features: Collect all duration extension ratios output by the feature calculation submodule 132 within this time period. And energy loss ratio This includes the penalty value for unresponsive events.

[0074] Count the number of unresponsive events: Count the number of heating events marked as "unresponsive input events" within this time period.

[0075] Calculate macroeconomic indicators: Calculate the following three statistics, which together constitute the dryness state indicator set. : a) Average duration stretch ratio : For all within the statistics window Find the arithmetic mean.

[0076] b) Average energy loss ratio : For all within the statistics window Find the arithmetic mean.

[0077] c) Percentage of unresponsive inputs : The proportion of the number of unresponsive input events to the total number (N) of all processed heating events within the statistics window.

[0078] Thus, the state assessment module 13 completes the transformation from discrete, independent input-output event streams to continuous, quantifiable drying process state indicators. By establishing a precise temporal relationship between heating events and pump load events, and extracting microscopic features characterizing time lag and energy loss, this module aggregates a series of isolated energy transfer events into a set of macroscopic state indicators that comprehensively reflect the energy transfer and conversion efficiency within the equipment package. This transformation process enables the system to move beyond monitoring the actions of individual devices and instead continuously gain insight into the phase change state and thermodynamic efficiency of the media (air, moisture) within the chamber from a physical perspective, providing stable and objective data for accurately determining the degree of dryness.

[0079] (4) Decision traceability module 14.

[0080] The decision traceability module 14 is the "decision brain" of the system. It receives the drying status index set updated in real time by the status assessment module 13, and automatically completes the determination of the drying endpoint and performs root cause classification and traceability when an abnormality occurs based on preset intelligent rules.

[0081] Optionally, the decision traceability module 14 is used to determine whether the instrument pack meets the drying standard based on the drying status index set, and to classify and trace the reasons for not meeting the drying standard if the instrument pack does not meet the drying standard.

[0082] For example, the decision tracing module 14 implements the above functions in the following two ways: (Case 1) The average duration extension, average energy loss, and percentage of no-response input in the dry state index set meet the preset convergence conditions within the first consecutive judgment duration.

[0083] It should be noted that the above convergence conditions are: the average duration extension value of the dry state index set is within the first preset range, the average energy loss value of the dry state index set is lower than the first preset threshold, and the proportion of no-response input in the dry state index set is zero.

[0084] In this case, the decision traceability module 14 determines that the instrument package has met the drying standard.

[0085] Specifically, the decision traceability module 14 monitors in real time the set of drying status indicators output by the status assessment module 13. This module sets a stability determination window. (For example, in this embodiment, it can be set to 30 seconds) as the continuous first determination duration. The decision tracing module 14 continuously checks the most recent During the period, Does it simultaneously meet the following three physical convergence conditions: a) Morphological regression benchmark: Average duration extension ratio Converging to the interval [1.0 - 1 / 2] 1.0+ Within this range, which is also the first preset range, To extend the specific tolerance (for example, 0.1 in this embodiment), this condition indicates that the gas waveform of the pump response no longer undergoes the time broadening caused by the liquid-gas phase transition, and the medium in the chamber has recovered to the rigid gas characteristics of dry air.

[0086] b) Zero energy loss: Average energy loss ratio Below the loss tolerance (That is, the first preset threshold, which can be 0.05). This condition indicates that almost all of the input electrothermal energy is converted into pneumatic work and is no longer absorbed by the latent heat of moisture.

[0087] c) Corresponding channels are intact: percentage of inputs without response The value is 0. This condition indicates that all energy inputs have triggered an effective aerodynamic response, and the energy transfer channels are unobstructed.

[0088] If and only if the above three conditions are consecutive If the drying conditions are met simultaneously and continuously within the specified time, the decision traceability module 14 determines that the instrument packs in the sterilization chamber have reached the drying standard. Subsequently, the module sends a stop command to the sterilizer's controller to end the drying process.

[0089] (Scenario 2) Within the second consecutive judgment period, the set of drying state indicators does not meet the preset convergence condition.

[0090] In this case, the decision tracing module 14 determines the category of reasons for not meeting the drying standard based on the numerical combination relationship of the average duration extension value, average energy loss value, and proportion of no response input in the drying status index set.

[0091] Optionally, the cause categories include slow drying due to the absorbency of the instrument packaging, energy transfer disruption due to blockage in the sterilizer tubing, and abnormal response due to mechanical failure of the sterilizer vacuum pump.

[0092] Specifically, after the drying process reaches its preset maximum running time (exemplarily set to 20 minutes in this embodiment), the decision tracing module 14 checks whether the convergence condition in Case 1 is still not met. If not, the anomaly diagnosis process is triggered. The decision tracing module 14 analyzes the current... The numerical characteristics of each component are determined, and the following logical branch judgments are executed based on their physical meanings: Branch 1, judged as "slow drying"): If the system detects the average time extension ratio Percentage of inputs that consistently show a value significantly higher than 1.0 (e.g., consistently greater than 1.5) but have no response. Maintain a low level (e.g., close to 0). This combination of characteristics indicates that energy input can trigger a pneumatic response, but the response waveform is severely broadened. This physically corresponds to the continued, slow evaporation of moisture within the chamber, which has not been completely drained due to insufficient drying time or overloading. The system determines the cause to be excessive instrument packaging or overly absorbent dressings.

[0093] Branch 2: Determined as an energy transfer blockage: If the system detects a percentage of unresponsive input... Significantly increased (e.g., greater than 20%), or average energy loss ratio Extremely high (e.g., close to 1.0) but average duration stretch ratio There was no corresponding increase. This characteristic indicates that energy was injected into the heater, but no corresponding pneumatic response was detected at the vacuum pump end. Physically, this corresponds to the energy being completely absorbed by the stagnant water zone (no gas generated), or the generated gas failing to reach the pump end due to pipe blockage. The system determined the cause to be a blockage in the drain pipe or a malfunctioning steam trap.

[0094] Branch 3, Determined as an abnormal response: If the dryness state index set The values ​​exhibit irregular, high-frequency random oscillations, and the three indicators lose a stable physical correlation (e.g., and The trends of change are contradictory, and (The energy flow fluctuates wildly). This indicates that the energy transfer process has lost its causal stability, and the system determines that the problem is either a decrease in the mechanical efficiency of the vacuum pump or a failure in the transmission mechanism.

[0095] This embodiment involves multiple thresholds (such as associated window length). Parameters such as statistical window length and drying judgment threshold are used, and their values ​​are based on the physical characteristics of the equipment and engineering experience. For example, the associated window... The 15-second time limit is based on the time constant of the heat conduction process. The first preset range used for drying determination (based on the elongation tolerance) (determined) and the first preset threshold (i.e., loss tolerance) ) is a tolerance range determined based on dry air reference experiments, for example 0.1 is acceptable. A value of 0.05 is acceptable. These thresholds are calibrated according to the specific sterilizer model during system deployment and are allowed to be adjusted within a certain range to adapt to different operating conditions.

[0096] Thus, the decision traceability module 14 realizes a decisive transformation from state monitoring of the drying process to intelligent closed-loop control and root cause tracing. By matching the numerical characteristics of the drying state index set with the physical convergence conditions of the drying endpoint and the logical characteristics of different abnormal modes, this module can automatically and accurately determine the drying completion node, and intelligently distinguish the root causes such as "overloading," "pipeline blockage," or "mechanical failure" when the target is not met. This transformation process maps the physical state data calculated at the front end into executable decisions that directly guide equipment operation (shutdown) and personnel intervention (maintenance, loading optimization). It not only solves the problem of wet packaging or energy waste that may be caused by traditional timed control, but also provides digital quality traceability capabilities based on physical evidence.

[0097] The above describes the big data-based full-process traceability system 10 for medical devices in a sterilization supply center and its included modules. Based on the above technical solution, the traceability system provided by this invention innovatively utilizes the non-invasively collected total power supply current of the sterilizer to construct a discrete event time-series correlation model of "electrothermal input-pneumatic output," achieving accurate, dynamic perception and intelligent closed-loop control of the actual drying state inside the medical device pack during the pressure steam sterilizer drying process. This solution fundamentally solves the technical problems of misjudging the risk of deep wet packs, energy waste, and difficulty in tracing the root causes of anomalies caused by traditional methods relying on fixed time or simple current thresholds. By decoupling and encapsulating continuous current signals into independent energy input and response events, and then calculating the microscopic physical characteristics characterizing phase transition hysteresis and latent heat erosion, the system can quantify the changes in energy transfer efficiency within the chamber, thereby determining the drying endpoint at the physical level and intelligently distinguishing different abnormal causes such as overloading, pipeline blockage, or mechanical failure. Ultimately, it forms a full-process digital traceability closed loop that includes objective physical evidence and supports quality backtracking, significantly improving the safety of sterilization packs and the level of intelligence in supply center management.

[0098] For example, such as Figure 4 The diagram shown is a flowchart illustrating a big data-based method for tracing the entire process of medical instruments in a disinfection supply center, according to an embodiment of the present invention. The method includes the following steps: S1. Obtain the power supply current data of the pressure steam sterilizer during the drying stage.

[0099] Optionally, this step is performed by the data acquisition module 11 described above, and the specific implementation process can be found above.

[0100] S2. Based on the power supply current data, determine the discrete heating event data and the discrete pump load event data. The discrete heating event data characterizes the independent events that convert electrical energy into heat energy and input it into the sterilization chamber, while the discrete pump load event data characterizes the independent response events of the vacuum pump performing work to remove gas from the chamber.

[0101] Optionally, this step is performed by the event encapsulation module 12 described above, and the specific implementation process can be found above.

[0102] S3. Calculate the drying state index set based on the temporal correlation between discrete heating event data and discrete pump load event data. The drying state index set characterizes the physical state of energy transfer and conversion efficiency within the instrument pack during the drying process.

[0103] Optionally, this step is performed by the state assessment module 13 described above, and the specific implementation process can be found above.

[0104] S4. Based on the set of drying status indicators, determine whether the instrument pack meets the drying standard, and if the instrument pack does not meet the drying standard, classify and trace the reasons for not meeting the drying standard.

[0105] Optionally, this step is performed by the decision tracing module 14 described above, and the specific implementation process can be found above.

[0106] The technical effects of the above-mentioned big data-based full-process traceability method for medical equipment in the disinfection supply center can be found in the previous description of the technical effects of the traceability system 10, which will not be repeated here.

[0107] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A big data-based end-to-end traceability system for medical instruments in a disinfection supply center, characterized in that, The system includes: The data acquisition module is used to acquire the power supply current data of the pressure steam sterilizer during the drying stage; An event encapsulation module is used to determine discrete heating event data and discrete pump load event data based on the power supply current data; wherein, the discrete heating event data is used to characterize an independent event of converting electrical energy into heat energy and inputting it into the sterilization chamber, and the discrete pump load event data is used to characterize an independent response event of the vacuum pump doing work to remove gas from the chamber. The state assessment module is used to calculate a set of drying state indicators based on the temporal correlation between the discrete heating event data and the discrete pump load event data; wherein, the set of drying state indicators is used to characterize the physical state of energy transfer and conversion efficiency inside the instrument pack during the drying process; The decision traceability module is used to determine whether the instrument pack meets the drying standard based on the set of drying status indicators, and to classify and trace the reasons for not meeting the drying standard if the instrument pack does not meet the drying standard.

2. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 1, characterized in that, When the event encapsulation module determines discrete heating event data and discrete pump load event data based on the power supply current data, it specifically performs the following steps: Based on the power supply current data, a reference current and a reference conversion parameter are determined; wherein, the reference current is used to characterize the basic power consumption required for the vacuum pump to maintain operation, and the reference conversion parameter is used to characterize the efficiency of converting thermal energy into mechanical work; Based on the current characteristics of different types of loads in the power supply current data, the power supply current data is divided into heating current sequence and vacuum pump current sequence; The heating current sequence is subjected to event identification and parameter extraction to generate discrete heating event data; wherein, the discrete heating event data includes at least a single input energy and an input duration; The vacuum pump current sequence is subjected to event identification and parameter extraction to generate discrete pump load event data; wherein, the discrete pump load event data includes at least the work done in a single response and the response duration.

3. The big data-based full-process traceability system for medical equipment in a disinfection supply center according to claim 2, characterized in that, When the event encapsulation module divides the power supply current data into heating current sequence and vacuum pump current sequence based on the current characteristics of different types of loads in the power supply current data, it specifically performs the following steps: Frequency domain analysis is performed on the power supply current data to extract the active current component that is in phase with the power supply voltage, and the active current component is determined as the heating current sequence. The heating current sequence is obtained by removing the power supply current data.

4. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 2, characterized in that, When the event encapsulation module performs event identification and parameter extraction on the heating current sequence to generate the discrete heating event data, it specifically performs the following steps: A current amplitude threshold is determined based on the reference current, current pulse segments greater than the current amplitude threshold are identified from the heating current sequence, and each identified current pulse segment is identified as a heating event. For each heating event, the single input energy is calculated based on the power of the heating current during the duration of the heating event, and the duration of each heating event is used as the input duration.

5. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 2, characterized in that, When the event encapsulation module performs event identification and parameter extraction on the vacuum pump current sequence to generate the discrete pump load event data, it specifically performs the following steps: An AC fluctuation sequence is determined from the vacuum pump current sequence, and peaks greater than a preset fluctuation threshold are identified from the AC fluctuation sequence. Each identified peak is defined as a pump load event. For each pump load event, the work done in a single response is calculated based on the power of the AC fluctuation during the duration of the pump load event, and the half-width of the fluctuation waveform of each pump load event is taken as the response duration.

6. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 1, characterized in that, When the state assessment module calculates the drying state index set based on the temporal correlation between the discrete heating event data and the discrete pump load event data, it specifically performs the following steps: According to the preset associated time window, the corresponding discrete pump load event data is matched for each discrete heating event data to obtain input and response data pairs; Calculate the duration extension feature and energy loss feature for each input and response data pair; wherein the duration extension feature is used to characterize the time lag of the aerodynamic response relative to the thermal energy input, and the energy loss feature is used to characterize the loss ratio during the energy transfer process; The drying state index set is generated based on all the duration extension characteristics, energy loss characteristics, and the proportion of discrete heating event data that failed to match the corresponding discrete pump load event data within the preset statistical period.

7. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 6, characterized in that, When the state assessment module matches the corresponding discrete pump load event data for each discrete heating event data, it specifically performs the following steps: For each discrete heating event data, search for the discrete pump load event data within the associated time window after the start time; If at least one discrete pump load event data is found within the associated time window, the discrete pump load event data with the smallest time difference between the peak time and the start time of the discrete heating event data is selected from the searched discrete pump load event data; the selected discrete pump load event data is determined as the corresponding discrete pump load event data to form the input and response data pair. If the discrete pump load event data is not found within the associated time window, the current discrete heating event data is marked as a no-response input event.

8. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 6, characterized in that, When calculating the duration extension characteristics and energy loss characteristics of each input-response data pair, the state assessment module specifically performs the following steps: The duration extension feature is calculated based on the response duration of the discrete pump load event data and the input duration of the discrete heating event data; The reference conversion parameters are obtained, and the energy loss characteristics are calculated based on the reference conversion parameters, the single input energy of the discrete heating event data in the input and response data pair, and the single response work of the discrete pump load event data in the input and response data pair.

9. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 1, characterized in that, When the decision traceability module determines whether the instrument pack meets the drying standard based on the drying status index set, and classifies and traces the reasons for not meeting the drying standard if the instrument pack does not meet the drying standard, it specifically performs the following steps: If the average duration extension, average energy loss, and percentage of no-response input in the set of drying state indicators meet the preset convergence conditions within a continuous first determination period, then the instrument pack is determined to have reached the drying standard; wherein, the preset convergence conditions are: the average duration extension of the set of drying state indicators is within a first preset range, the average energy loss of the set of drying state indicators is lower than a first preset threshold, and the percentage of no-response input in the set of drying state indicators is zero. If the set of drying state indicators fails to meet the preset convergence condition within the second consecutive determination period, the category of cause for not meeting the drying standard is determined based on the numerical combination relationship of the average duration extension value, the average energy loss value, and the proportion of no-response input in the set of drying state indicators; wherein, the category of cause includes slow drying due to the absorbency of the instrument packaging, energy transfer blockage due to the blockage of the sterilizer pipeline, and abnormal response due to mechanical failure of the sterilizer vacuum pump.

10. The big data-based full-process traceability system for medical instruments in a disinfection supply center according to claim 2, characterized in that, When determining the reference current and reference conversion parameters based on the power supply current data, the event encapsulation module specifically performs the following steps: The power supply current data is collected under the condition that the pressure steam sterilizer is performing the pre-vacuum stage and there is no instrument load in the sterilization chamber, and the average value of the power supply current data during the pre-vacuum stage and the period when the vacuum pump maintains the basic vacuum level is calculated as the reference current. Obtain the input energy reference value corresponding to a heating event during the pre-vacuum stage, and the response work reference value corresponding to a pump load event triggered by the heating event, and determine the reference conversion parameter based on the input energy reference value and the response work reference value.

Images