Intelligent inspection system for hospital supply room equipment

Abstract

The invention discloses an intelligent inspection system for equipment in a hospital supply room, relates to the technical field of safety detection of medical equipment in the hospital supply room, and aims to solve the technical problems of insufficient result accuracy, difficulty in pollution tracing and conflict between equipment calibration and emergency treatment detection requirements in endoscopic pathogen detection in the current hospital supply room. The multi-pathogen parallel detection module is used for carrying out multi-pathogen parallel detection on endoscope surface eluent; the material adaptation analysis module is used for acquiring endoscope material information and judging the suitability of the detection reagent and the endoscope material; the AI interpretation module is used for receiving the detection signal and the adaptive result, carrying out detection signal interpretation and pathogen concentration prediction and generating a detection report; and the data linkage module is used for binding full-process data of the endoscope. The method has the advantages that the detection result is accurate and reliable, the pollution source can be positioned, the process can be optimized, and the requirements of equipment calibration and emergency treatment detection can be balanced to guarantee the clinical efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical instrument safety detection in hospital supply room, and more particularly to a hospital supply room equipment intelligent inspection system. BACKGROUND

[0002] The endoscope pathogen detection scheme currently used in the hospital supply room generally has the core technical problem of "insufficient detection result accuracy and lack of dynamic adaptation ability". The existing scheme relies on fixed thresholds for detection result interpretation, without fully considering the interference of detection environment factors on signal interpretation, nor adjusting the concentration calculation method in combination with the material characteristics of the endoscope. From the environmental impact, changes in the detection environment will directly affect the stability of the detection signal, leading to a mismatch between the interpretation standard and the actual detection conditions; from the material impact, different materials of the endoscope have different adsorption abilities for pathogens, and the material performance will change over time, but the existing scheme uses a unified concentration calculation logic, which can easily underestimate or overestimate the pollution level of some endoscopes. This problem may on the one hand lead to the flow of endoscopes that have not been thoroughly cleaned into the clinic, posing a risk of cross-infection; on the other hand, it may cause the misjudgment of endoscopes that are otherwise qualified as unqualified, thereby leading to excessive cleaning or unnecessary scrapping, increasing medical costs, and occupying detection resources, affecting overall detection efficiency.

[0003] If only the above precision problem is optimized, two types of derived problems will still be faced: first, even if the high concentration pollution of the endoscope is accurately detected, it is difficult to quickly locate the source of the pollution, and it is difficult to determine whether the pollution is caused by improper operation in the cleaning process, pollution during clinical use, or pathogen residue caused by the material characteristics of the endoscope, making it difficult to reduce the pollution rate fundamentally; second, the core equipment for accurate detection needs to be calibrated regularly to maintain stable performance, but the calibration process often takes a certain amount of time, and if there is an emergency endoscope detection demand during calibration, the existing scheme cannot effectively balance the contradiction between equipment calibration and emergency detection, which may cause the interruption of emergency detection process and affect the clinical treatment efficiency. In view of this, we propose a hospital supply room equipment intelligent inspection system. SUMMARY

[0004] The purpose of the present application is to provide a hospital supply room equipment intelligent inspection system to solve the technical problems of insufficient result accuracy, pollution traceability difficulty, and conflict between equipment calibration and emergency detection demand in the current endoscope pathogen detection in the hospital supply room.

[0005] To solve the above technical problems, the present application provides the following technical scheme: a hospital supply room equipment intelligent inspection system, comprising: a microfluidic detection module for parallel detection of multiple pathogens in the endoscope surface eluent; The material adaptation analysis module is configured to acquire endoscope material information and determine the adaptability of the detection reagent to the endoscope material. The AI interpretation module is configured to receive the detection signal and the adaptation result, perform detection signal interpretation and pathogen concentration prediction, and generate a detection report. The data linkage module is configured to bind endoscope full-process data, realize data tracing, pollution tracing, early warning pushing, and cleaning process optimization suggestion output. The device calibration module is configured to monitor the running state of the microfluidic detection module, adjust the calibration priority according to the clinical demand, and automatically trigger the calibration.

[0006] Preferably, the microfluidic detection module comprises a multi-channel microfluidic chip, each channel of the multi-channel microfluidic chip is provided with a specific recognition element for different pathogens, and each channel is configured with a fluorescent marker of different wavelength to distinguish the detection signal.

[0007] Preferably, the material adaptation analysis module is in communication connection with the microfluidic detection module and the hospital endoscope management system, and the acquired endoscope material information includes the main body material and the surface coating material. The adaptability determination specifically analyzes whether the detection reagent has the risk of corrosion, discoloration or coating falling of the endoscope material, and switches to a low-stimulation detection reagent or shortens the contact time of the detection reagent with the endoscope surface when the risk exists.

[0008] Preferably, the AI interpretation module is in communication connection with the microfluidic detection module and the material adaptation analysis module. The detection signal interpretation adopts a dynamic threshold method: by real-time acquisition of detection environment parameters, chip reaction time data and adaptation results, a multi-dimensional correction model is constructed and the interpretation threshold is automatically adjusted. The pathogen concentration prediction specifically includes: collecting detection signals at multiple time points, inversely calculating the initial concentration through a deep learning model, correcting the concentration value in combination with the influence coefficient of the endoscope material on pathogen adhesion, dividing the pollution level and outputting the treatment suggestion containing differentiated re-cleaning parameters. The AI interpretation module also regularly summarizes detection data, manual review results and material loss feedback, updates model parameters and material influence coefficients. The automatic adjustment of the interpretation threshold is realized through the following algorithm formula: ; In the formula, is the adjusted dynamic interpretation threshold, is the reference threshold, is the deviation value of the detection environment temperature from the standard temperature, is the deviation value of the actual reaction time of the chip from the standard reaction time, is the temperature influence coefficient, This is the time-related influence coefficient. The revised formula for calculating pathogen concentration is as follows: ; In the formula, This is the corrected pathogen concentration. The initial concentration calculated for the deep learning model. The material influence coefficient, This refers to the fatigue factor of endoscope materials.

[0009] Preferably, the data linkage module is communicatively connected to the AI ​​interpretation module, the hospital cleaning equipment system, and the clinical information system, respectively, and assigns a unique electronic identifier combining an RFID tag and a QR code to each endoscope, through which the following data is associated: Cleaning data: operating parameters of cleaning equipment, type of cleaning reagent, and cleaning duration; Test data: pathogen type, concentration, interpretation criteria, personnel information, and material compatibility results; Clinical usage data: Departments using the product, duration of use, patient disease type, and adherence to operating procedures; Material data: Endoscope material type, service life, and historical material wear and tear records.

[0010] Preferably, the pollution source analysis of the data linkage module is as follows: when a high concentration of pollution is detected, a source tracing chart is generated, marking the probability of pollution caused by cleaning, clinical use, and material characteristics, and combining historical data to locate high-frequency pollution links; The warning push is as follows: when a certain endoscope is contaminated with high concentrations multiple times within a preset cycle, a risk warning will be pushed to the clinic and the cleaning equipment process will be upgraded. The recommended optimization of the cleaning process is as follows: for high-frequency contamination steps, provide solutions for adjusting cleaning reagents, extending the time, or adding new steps; The probability of pollution stages is calculated using the following algorithm formula: ; In the formula, For the first The probability of contamination at each stage. For the first The weighting coefficients of each stage, For the first Corrected pathogen concentrations associated with each step. Corresponding to the cleaning process, Corresponding to clinical use, Regarding the corresponding material properties, The index is set to the process, with values ​​of 1, 2, and 3, corresponding to the three contamination-related processes: cleaning, clinical use, and material characteristics, respectively.

[0011] Preferably, the device calibration module is integrated into the microfluidic detection module to monitor the intensity of the light source, the pressure of the chip channel, and the sensitivity of the detector. Adjust calibration priority according to clinical needs: when there is a demand for grade I emergency detection completed within 30 minutes and the index deviation is acceptable, temporarily delay calibration, prioritize detection, and calibrate immediately after the task is completed; when there is no emergency demand or deviation exceeds the limit, calibrate immediately; The automatic calibration process is as follows: a built-in standard sample storage unit is used to extract standard samples and inject them into the chip, the AI interpretation module calculates the deviation value and automatically adjusts the parameters of the microfluidic detection module; after calibration, part of the eluate is extracted for repeated detection to verify consistency, and if the consistency is not consistent, recalibration is performed, and a calibration log containing the deviation value, adjustment record, and results is generated and uploaded to the hospital equipment management system; The deviation value is calculated by the following algorithm formula: ; In the formula, is the comprehensive deviation of the device, is the light source intensity deviation, calculated as , is the actual light source intensity, is the standard light source intensity, is the chip channel pressure deviation, calculated as , is the actual pressure, is the standard pressure, is the detector sensitivity deviation, calculated as , is the actual sensitivity, is the standard sensitivity; The parameter adjustment amount of the microfluidic detection module is calculated by the following algorithm formula: ; In the formula, is the chip channel pressure calibration adjustment amount, is the deviation correction coefficient.

[0012] Preferably, it also includes a device locking module, which is in communication with the AI interpretation module and the material adaptation analysis module. When the AI interpretation module determines that the endoscopic detection is unqualified or the material adaptation analysis module determines that there is a risk of material damage, the device locking module automatically locks the endoscope to prevent it from flowing to the next link until the preset processing process of re-washing, reagent replacement, or maintenance is completed, and records the locking reason and processing process.

[0013] Preferably, the microfluidic detection module, AI interpretation module, data linkage module, equipment calibration module, and material adaptation analysis module achieve data interaction through the hospital's internal network. Data transmission is encrypted, and the patient data associated with the endoscope is desensitized.

[0014] Preferably, it also includes an emergency backup module connected in parallel with the microfluidic detection module. When the microfluidic detection module fails or is in calibration and there is an urgent need for detection, the emergency backup detection unit is automatically activated. The emergency backup detection unit adopts a simplified multi-pathogen detection process to ensure basic detection needs.

[0015] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention enables parallel detection of multiple pathogens through a microfluidic detection module. Combined with the dynamic threshold adjustment and material influence coefficient correction functions of the AI ​​interpretation module, the detection results can be adapted to different environmental conditions and endoscope material characteristics, effectively improving the accuracy of the detection results. This avoids the clinical risks caused by underestimating the degree of contamination and reduces the waste of resources and efficiency loss caused by misjudgment.

[0016] 2. This invention also establishes a unique electronic identifier for each endoscope by using a data linkage module, linking it to data from the entire process, including cleaning, clinical use, and materials. Combined with the pollution source tracing analysis function, it identifies high-frequency pollution links, making rectification measures more targeted, helping to reduce the probability of endoscope contamination from the source, and improving the overall management efficiency of the hospital supply room.

[0017] 3. The present invention also dynamically adjusts the calibration priority through the equipment calibration module and is equipped with an emergency backup module. While ensuring the equipment calibration effect, it can respond to emergency testing needs in a timely manner, avoid delays in emergency testing caused by equipment calibration, and balance testing accuracy and clinical service efficiency. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the system framework of the present invention. Detailed Implementation

[0019] To facilitate understanding of the technical solution of the present invention by those skilled in the art, the technical solution of the present invention will now be further described in conjunction with the accompanying drawings.

[0020] Example 1, as Figure 1 As shown, the present invention provides an intelligent inspection system for hospital supply room equipment, comprising: A microfluidic detection module is used for parallel detection of multiple pathogens in the eluent on the endoscope surface. The material compatibility analysis module is used to obtain endoscope material information and determine the compatibility between the test reagents and the endoscope material. The AI interpretation module is configured to receive the detection signal and the adaptation result, perform detection signal interpretation and pathogen concentration prediction, and generate a detection report. The data linkage module is configured to bind endoscope full-process data, realize data tracing, pollution tracing, early warning pushing, and cleaning process optimization suggestion output. The device calibration module is configured to monitor the running state of the microfluidic detection module, adjust the calibration priority according to clinical needs, and automatically trigger calibration.

[0021] In the embodiment of the present application, the microfluidic detection module comprises a multi-channel microfluidic chip, each channel of the multi-channel microfluidic chip is provided with a specific recognition element (specific antibody or nucleic acid probe) for different pathogens, and each channel is configured with a fluorescent marker of different wavelength to distinguish the detection signal. The microfluidic detection module can complete the synchronous detection of multiple pathogens through single sampling.

[0022] In the embodiment of the present application, the material adaptation analysis module is in communication connection with the microfluidic detection module and the hospital endoscope management system, and the obtained endoscope material information includes the main body material and the surface coating material. The adaptability judgment specifically refers to analyzing whether the detection reagent has the risk of corrosion, discoloration or coating falling off on the endoscope material, and switching to a low-stimulation detection reagent or shortening the contact time of the detection reagent with the surface of the endoscope when there is a risk.

[0023] In the embodiment of the present application, the AI interpretation module is in communication connection with the microfluidic detection module and the material adaptation analysis module. The detection signal interpretation adopts a dynamic threshold method: by collecting detection environment parameters, chip reaction time data and adaptation results in real time, a multi-dimensional correction model is constructed and the interpretation threshold is automatically adjusted. The pathogen concentration prediction specifically refers to collecting detection signals at multiple time points, inversely calculating the initial concentration through a deep learning model, correcting the concentration value in combination with the influence coefficient of the endoscope material on pathogen adhesion, dividing the pollution level and outputting the treatment suggestion containing differentiated re-cleaning parameters. The AI interpretation module also regularly summarizes detection data, manual review results and material loss feedback, updates model parameters and material influence coefficients. The automatic adjustment of the interpretation threshold is realized through the following algorithm formula: ; In the formula, is the adjusted dynamic interpretation threshold, i.e. the fluorescence signal judgment standard adapted to the current detection environment and reaction condition, which is used to distinguish the signal critical value of "qualified / unqualified"; is the reference threshold, i.e. the preset fluorescence signal judgment reference value under the standard detection environment (such as standard temperature, standard reaction time). The deviation value of the actual temperature from the standard temperature, a positive value indicates that the actual temperature is higher than the standard temperature, and a negative value indicates that the actual temperature is lower than the standard temperature; The deviation value of the actual reaction time of the chip from the standard reaction time, a positive value indicates that the actual reaction time is longer than the standard time, and a negative value indicates that the actual reaction time is shorter than the standard time; The temperature influence coefficient, generated by the AI judgment module through historical detection data self-learning, is used to quantify the influence degree of unit temperature deviation on the judgment threshold; The time influence coefficient, also generated by the AI judgment module self-learning, is used to quantify the influence degree of unit reaction time deviation on the judgment threshold; The core of this algorithm is to realize "self-adaptive threshold adjustment of environment and reaction conditions". First, take the baseline threshold as the basis, and then dynamically correct according to the detection environment temperature deviation and the chip reaction time deviation Temperature deviation will affect the fluorescence signal intensity (such as temperature too high may enhance the signal), so through term to offset the temperature interference; Reaction time deviation will affect the binding degree of pathogens and recognition elements (such as too short reaction time may lead to weak signal), so through term to compensate for the time difference, and finally get the accurate judgment threshold adapted to the current detection conditions; The calculation formula of the corrected pathogen concentration is: ; In the formula, is the corrected pathogen concentration, which reflects the real pathogen number on the endoscope surface after eliminating the influence of endoscope material; is the initial concentration calculated by the deep learning model, which is the pathogen concentration calculated by the fluorescence signal intensity without considering the material influence; is the material influence coefficient, updated by the AI judgment module according to the endoscope material type (such as rubber, resin, metal coating) and historical material loss record, which quantifies the adsorption capacity of different materials to pathogens; is the endoscope material fatigue factor, calculated by the material adaptation analysis module according to the service life of the endoscope. The longer the service life, the greater the material fatigue factor, which reflects the change of material adsorption capacity with time; The core of this algorithm is to solve the interference of endoscope material on the detection result. First, get the initial concentration , combined with the modification of endoscope material properties, different materials (such as rubber, resin) have different adsorption capacities for pathogens, and the longer the material is used, the stronger the adsorption capacity may be, and the initial concentration without modification is easy to underestimate the actual pollution degree, so by quantifying the influence of the material, the modified concentration reflecting the real pollution situation is finally obtained ; Through the dynamic threshold adjustment algorithm, false judgments caused by environmental temperature fluctuations and reaction time differences can be avoided, ensuring the consistency of the interpretation standard under different detection conditions; combined with the concentration calculation algorithm of the material modification, the interference of material adsorption on the detection result can be accurately eliminated, avoiding the flow of unqualified endoscopes into the clinic due to the low estimation of the pollution degree, or causing the endoscope to be excessively scrapped due to the overestimation of the pollution degree, and the design of the AI self-learning update coefficient can make the algorithm continuously adapt to new endoscope materials and pathogen variations, improving long-term detection accuracy and system adaptability.

[0024] In the embodiments of the present application, the data linkage module is respectively in communication connection with the AI interpretation module, the hospital cleaning equipment system and the clinical information system, and a unique electronic identifier combined with an RFID tag and a two-dimensional code is allocated for each endoscope, and the following data is associated through the identifier: cleaning data: cleaning equipment operating parameters, cleaning reagent types and cleaning time; detection data: pathogen species, concentration, interpretation basis, detection personnel information and material adaptation result; clinical use data: using department, use time, patient disease type and operation specification execution situation; material data: endoscope material type, service life and historical material loss record.

[0025] In the embodiments of the present application, the pollution traceability analysis of the data linkage module is: when high concentration pollution is detected, a traceability chart is generated, and the probabilities of cleaning, clinical use and material characteristics causing pollution are marked, and high-frequency pollution links are located combined with historical data; early warning push: when a certain endoscope has high concentration pollution for multiple times within a preset period, a risk warning is pushed to the clinic and the cleaning equipment process is upgraded; cleaning process optimization suggestion: for high-frequency pollution links, output the scheme of cleaning reagent adjustment, time extension or additional steps.

[0026] Among them, the pollution link probability is calculated by the following algorithm formula: ; In the formula, is the pollution probability of the th link, expressed in percentage, and the larger the value is, the higher the possibility of causing pollution is, corresponding to the cleaning link, corresponding to the clinical use link, Corresponding material properties; For the first The weighting coefficients for each step are derived by the data linkage module based on historical pollution data (e.g., if the cleaning step has a high pollution percentage in historical data, then...). (More extensively), quantifying the historical contribution of each stage in a pollution incident; For the first Corrected pathogen concentrations associated with each step. The time represents the concentration of pathogens remaining on the endoscope surface after cleaning, reflecting the impact of incomplete cleaning; The concentration of pathogens newly added after clinical use reflects the impact of clinical operation contamination; The time represents the increase in pathogen concentration caused by material adsorption, reflecting the influence of material properties.

[0027] The index is set to the stage, with values ​​of 1, 2, and 3, corresponding to the three pollution-related stages: cleaning, clinical use, and material characteristics, respectively. This index is used to iterate through and calculate the total weighted concentration. The core of this algorithm is to achieve quantitative traceability of contamination responsibility across multiple stages. First, a weight coefficient is assigned to each contamination-related stage (cleaning, clinical use, material properties). (Reflecting the historical frequency of contamination at each stage), combined with the corrected pathogen concentrations associated with each stage. The pollution probability of each stage is obtained by calculating the ratio of "stage-weighted concentration / total weighted concentration". This allows us to pinpoint the main sources of pollution. This contamination source tracing algorithm upgrades traditional "fuzzy attribution" to "quantitative probability tracing," accurately pinpointing specific sources of contamination such as incomplete cleaning, improper clinical procedures, or material adsorption. This avoids ineffective rectification due to the inability to clearly identify the responsible party. Furthermore, by dynamically adjusting weighting coefficients based on historical data, the tracing results better reflect the hospital's actual operations. This provides data support for optimizing cleaning processes (such as adjusting reagents or duration for high-probability cleaning steps) and improving clinical operating procedures (such as strengthening disinfection requirements for high-probability clinical procedures), thereby reducing the risk of endoscopic contamination at its source.

[0028] In an embodiment of the present invention, the device calibration module is integrated into the microfluidic detection module to monitor the light source intensity, chip channel pressure, and detector sensitivity. Adjust calibration priority based on clinical needs: If there is a Level I emergency test requirement that can be completed within 30 minutes and the index deviation is acceptable, postpone calibration and prioritize testing. Calibrate immediately after the task is completed. If there is no emergency requirement or the deviation exceeds the limit, calibrate immediately. The automatic trigger calibration process is: a built-in standard sample storage unit, a standard sample injection chip, a deviation value calculation by an AI interpretation module and automatic adjustment of the microfluidic detection module parameters; after calibration, part of the eluate is extracted for repeated detection to verify consistency, and if inconsistent, recalibration is performed, and a calibration log containing the deviation value, adjustment record and result is generated and uploaded to the hospital equipment management system; wherein the deviation value is calculated by the following algorithm formula: ; In the formula, is the equipment comprehensive deviation, dimensionless, quantifying the overall deviation degree of the microfluidic detection module in the three dimensions of light source, pressure and sensitivity, and the larger the value, the lower the device running accuracy; is the light source intensity deviation, relative deviation (dimensionless), and the calculation method is , which quantifies the deviation degree of the actual light source intensity from the standard value; is the actual light source intensity, i.e. the light source intensity value collected in real time at the device calibration time; is the standard light source intensity, i.e. the standard light source intensity value preset when the microfluidic detection module is shipped or debugged; is the chip channel pressure deviation, relative deviation (dimensionless), and the calculation method is , which quantifies the deviation degree of the actual channel pressure from the standard value; is the actual pressure, i.e. the pressure value collected in real time in the chip microchannel at the device calibration time; is the standard pressure, i.e. the standard pressure value preset for the chip channel pressure of the microfluidic detection module, which is also the reference value for calibration parameter adjustment; is the detector sensitivity deviation, relative deviation (dimensionless), and the calculation method is , which quantifies the deviation degree of the actual detector sensitivity from the standard value; is the actual sensitivity, i.e. the signal response sensitivity value collected in real time by the fluorescence detector at the device calibration time; is the standard sensitivity, i.e. the standard sensitivity value preset for the detector of the microfluidic detection module; The core of the algorithm is the comprehensive quantification of multi-dimensional device deviation. First, the single deviation of the light source intensity, chip channel pressure and detector sensitivity is calculated respectively , , (All of these are relative deviations). Then, by taking the square root of the sum of squares, the deviations of the three dimensions are integrated into a comprehensive deviation value. This avoids misjudging the overall condition of the equipment due to a single indicator bias, and enables a comprehensive assessment of the equipment's operational accuracy. The parameter adjustment amount of the microfluidic detection module is calculated using the following algorithm formula: ; In the formula, This is the chip channel pressure calibration adjustment amount, which is the amount of pressure change required to adjust the actual pressure of the chip channel to the standard value. A positive value indicates that the pressure needs to be increased, and a negative value indicates that the pressure needs to be decreased. The deviation correction factor is derived from... (Probability of contamination during cleaning process) is dynamically adjusted. The higher, The larger the value, the more thorough the calibration is in scenarios requiring high detection accuracy; The core of this algorithm is to dynamically adjust the calibration intensity as needed, based on the overall equipment deviation. Based on, combined with standard pressure Then, through the deviation correction coefficient Adjusting the calibration range; a higher probability of contamination during the cleaning process indicates a higher requirement for detection accuracy. The larger the value, the more calibration adjustment is required. The larger the size, the more accurate the equipment will be in critical demand scenarios. Through a multi-dimensional comprehensive deviation algorithm, the operating status of the equipment can be fully evaluated, avoiding the decrease in detection accuracy caused by ignoring deviations in other indicators due to the normality of a single indicator. Combined with an algorithm that dynamically adjusts the calibration intensity based on the probability of contamination during the cleaning process, "on-demand calibration" can be achieved. In scenarios with high cleaning contamination risk and urgent need for detection accuracy, the equipment accuracy can be ensured by increasing the calibration adjustment amount. In non-critical scenarios, over-calibration can be avoided, which would lead to prolonged equipment downtime. At the same time, the design of consistency verification after calibration can ensure that the calibration effect meets the standards and avoid incomplete calibration affecting subsequent detection, thus providing equipment-level assurance for the reliability of endoscopic detection results.

[0029] In an embodiment of the present invention, a device locking module is also included. The device locking module is communicatively connected to the AI ​​judgment module and the material compatibility analysis module. When the AI ​​judgment module determines that the endoscope is unqualified or the material compatibility analysis module determines that there is a risk of material damage, the endoscope is automatically locked to prevent it from flowing into the next stage until the preset processing flow of rewashing, reagent replacement or repair is completed, and the reason for locking and the processing process are recorded.

[0030] In the embodiment of the application, the microfluidic detection module, the AI interpretation module, the data linkage module, the equipment calibration module and the material adaptation analysis module realize data interaction through a hospital internal network, data transmission adopts encryption processing, and patient data associated with the endoscope is desensitized.

[0031] In the embodiment of the application, an emergency backup module is further included in parallel with the microfluidic detection module, when the microfluidic detection module fails or is in a calibration state and there is an emergency detection demand, an emergency backup detection unit is automatically started, and the emergency backup detection unit adopts a simplified multi-pathogen detection process to guarantee basic detection demand.

[0032] The embodiment of the application discloses a preferred embodiment, but is not limited thereto, and a person of ordinary skill in the art can easily understand the spirit of the application according to the above-mentioned embodiment, and make different inferences and changes, as long as they do not deviate from the spirit of the application, they are within the protection scope of the application.

Claims

1. A hospital supply room equipment intelligent inspection system, characterized in that, The application relates to a medical endoscope quality control system, which comprises the following modules: a microfluidic detection module for parallel detection of multiple pathogens in endoscope surface eluate; a material adaptation analysis module for obtaining endoscope material information and judging the adaptability of detection reagents to endoscope materials; an AI interpretation module for receiving detection signals and adaptability results, interpreting the detection signals, predicting pathogen concentrations, and generating a detection report; a data linkage module for binding endoscope whole-process data to realize data tracing, pollution tracing, early warning pushing, and cleaning process optimization suggestion output; a device calibration module for monitoring the running state of the microfluidic detection module, adjusting the calibration priority according to clinical requirements, and automatically triggering calibration.

2. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The microfluidic detection module comprises a multi-channel microfluidic chip, each channel of the multi-channel microfluidic chip is provided with a specific recognition element for different pathogens, and each channel is provided with a fluorescent marker of different wavelength to distinguish the detection signals.

3. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The material adaptation analysis module is in communication connection with the microfluidic detection module and the hospital endoscope management system, and the obtained endoscope material information comprises main body material and surface coating material; The adaptability judgment specifically refers to analyzing whether the detection reagent has the risk of corrosion, discoloration or coating falling of the endoscope material, and switching to a low-stimulation detection reagent or shortening the contact time of the detection reagent and the endoscope surface when the risk exists.

4. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The AI interpretation module is in communication connection with the microfluidic detection module and the material adaptation analysis module; The detection signal interpretation adopts a dynamic threshold mode: by collecting detection environment parameters, chip reaction time data and adaptability results in real time, a multi-dimensional correction model is constructed, and the interpretation threshold is automatically adjusted; The pathogen concentration prediction specifically refers to collecting detection signals at multiple time points, inversely calculating the initial concentration through a deep learning model, correcting the concentration value by combining the influence coefficient of the endoscope material on pathogen adhesion, dividing the pollution level and outputting the treatment suggestion containing differential re-washing parameters; The AI interpretation module also regularly summarizes detection data, manual review results and material loss feedback, updates model parameters and material influence coefficients; The automatic adjustment of the interpretation threshold is realized through the following algorithm formula: ； In the formula, is the adjusted dynamic interpretation threshold value, is the reference threshold value, is the deviation value of the detection ambient temperature from the standard temperature, is the deviation value of the actual reaction time of the chip from the standard reaction time, is the temperature influence coefficient, is the time influence coefficient; The calculation formula of the corrected pathogen concentration is: ； In the formula, is the corrected pathogen concentration, is the initial concentration calculated by the deep learning model, is the material influence coefficient, is the endoscope material fatigue factor.

5. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The data linkage module is in communication connection with the AI interpretation module, a hospital cleaning equipment system and a clinical information system, and a unique electronic identifier of an RFID tag and a two-dimensional code combination is allocated to each endoscope, and the following data are associated through the identifier: Cleaning data: cleaning equipment operation parameters, cleaning reagent types and cleaning time; Detection data: pathogen types, concentrations, interpretation basis, detection personnel information and material adaptability results; Clinical use data: use department, use time, patient disease type and operation specification execution situation; Material data: endoscope material type, service life and historical material loss record.

6. A hospital supply room equipment intelligent inspection system according to claim 5, wherein, The pollution tracing analysis of the data linkage module is: when high-concentration pollution is detected, a tracing chart is generated, the probabilities of cleaning, clinical use and material characteristics causing pollution are marked, and high-frequency pollution links are located in combination with historical data; The early warning pushing is: when an endoscope is polluted at high concentration for multiple times within a preset period, a risk early warning is pushed to the clinic, and a cleaning equipment process upgrade is triggered. The cleaning process optimization suggestion is: for high-frequency pollution links, output cleaning reagent adjustment, time extension or new step scheme; In the formula, the pollution link probability is calculated by the following algorithm formula: ; In the formula, the pollution probability of the first link, the weight coefficient of the first link, the corrected pathogen concentration associated with the first link, corresponding to the cleaning link, corresponding to the clinical use link, corresponding to the material property link, is a link index, taking values 1, 2, 3, respectively corresponding to the three pollution associated links of cleaning, clinical use, and material property.

7. A hospital supply room equipment intelligent inspection system according to claim 6, wherein, The device calibration module is integrated in the microfluidic detection module, and monitors light source intensity, chip channel pressure and detector sensitivity; The calibration priority is adjusted according to clinical needs: when there is a Ⅰ level emergency detection demand that can be completed within 30 minutes and the index deviation is acceptable, the calibration is temporarily delayed, the detection is prioritized, and the calibration is performed immediately after the task is completed; when there is no emergency demand or the deviation is out of limit, the calibration is performed immediately; The automatic calibration process is: a standard sample storage unit is built in, the standard sample is extracted and injected into the chip, the deviation value is calculated by the AI judgment module and the microfluidic detection module parameters are automatically adjusted; after calibration, part of the eluate is repeatedly detected to verify consistency, if inconsistent, recalibration, at the same time, generate calibration log containing deviation value, adjustment record and result, upload to hospital equipment management system; The deviation value is calculated by the following algorithm formula: ； In the formula, is the equipment comprehensive deviation, is the light source intensity deviation, and the calculation method is , is the actual light source intensity, is the standard light source intensity, is the chip channel pressure deviation, and the calculation method is , is the actual pressure, is the standard pressure, is the detector sensitivity deviation, and the calculation method is , is the actual sensitivity, is the standard sensitivity; The microfluidic detection module parameter adjustment amount is calculated by the following algorithm formula: ； In the formula, is the chip channel pressure calibration adjustment amount, is the deviation correction coefficient.

8. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The device locking module is also included, which is in communication connection with the AI judgment module and the material adaptation analysis module, when the AI judgment module determines that the endoscopic detection is unqualified or the material adaptation analysis module determines that there is a risk of material damage, the endoscope is automatically locked to prohibit flowing into the next link until the preset processing process of re-washing, reagent replacement or maintenance is completed, and the locking reason and processing process are recorded.

9. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, The microfluidic detection module, AI judgment module, data linkage module, device calibration module and material adaptation analysis module realize data interaction through the hospital internal network, data transmission adopts encryption processing, and patient data associated with the endoscope is desensitized.

10. The system for intelligent inspection of hospital supply room equipment of claim 1, wherein, An emergency backup module parallel to the microfluidic detection module is also included, when the microfluidic detection module fails or is in calibration state and there is an emergency detection demand, the emergency backup detection unit is automatically started, the emergency backup detection unit adopts a simplified version of multi-pathogen detection process to guarantee basic detection demand.

Images