Disinfection supply detection method and system

Abstract

The invention relates to a disinfection supply detection method and system, and relates to the field of disinfection detection, and the method comprises the steps: obtaining the types and number of medical instruments disinfected in a disinfection supply room, and obtaining the instrument type distribution; performing instrument complexity classification according to instrument category distribution to obtain instrument complexity, performing instrument disinfection test sampling inspection quantity calculation according to the instrument complexity to obtain the sampling inspection quantity, randomly extracting the medical instruments with the sampling inspection quantity, and acquiring a medical instrument image set; and performing medical instrument surface damage identification on the medical instrument image set to obtain a surface damage parameter set, and combining the instrument complexity to calculate the casual inspection number of the damaged areas to obtain the casual inspection number of the damaged areas, and performing random sampling biological detection on the extracted medical instruments to obtain a detection result. The technical problems of inaccurate detection, low reliability and low efficiency after disinfection supply are solved.

Description

Technical Field

[0001] The present invention relates to the field of disinfection detection, and particularly to a disinfection supply detection method and system. Background Art

[0002] Currently, medical devices are disinfected through centralized disinfection supply to improve the disinfection efficiency of medical devices. For the disinfected medical devices, it is necessary to detect whether the disinfection is complete. Traditional detection methods often rely on sampling inspection or simple automated equipment for simple automatic detection, which is prone to incomplete detection due to human negligence or insufficient equipment accuracy. Moreover, the cost of detecting all medical devices is too high and the efficiency is too low. Therefore, the current disinfection supply detection of medical devices is carried out by simple sampling detection means, which will lead to incomplete detection and the situation of undetected incomplete disinfection. Summary of the Invention

[0003] The present invention aims at the technical problem that the disinfection supply detection of medical devices in the prior art is incomplete, resulting in the situation of undetected incomplete disinfection, and proposes a disinfection supply detection method and system.

[0004] The technical solution of the present invention to solve the above technical problems is as follows:

[0005] In a first aspect, the present invention provides a disinfection supply detection method, including: obtaining the categories and quantities of medical devices being disinfected in a disinfection supply room to obtain the distribution of device categories;

[0006] According to the distribution of device categories, classifying the complexity of devices to obtain the device complexity, and calculating the number of samples to be randomly selected for device disinfection testing according to the device complexity, obtaining the number of samples and randomly selecting the number of medical devices of the samples, and collecting a set of medical device images;

[0007] Identifying surface damage of medical devices in the set of medical device images to obtain a set of surface damage parameters, combining the device complexity, calculating the number of samples to be randomly selected for the damaged area, obtaining the number of samples for the damaged area and randomly sampling and biologically testing the selected medical devices to obtain the test results.

[0008] In a second aspect, the present invention provides a disinfection supply detection system, including:

[0009] A device category distribution acquisition module for obtaining the categories and quantities of medical devices being disinfected in a disinfection supply room to obtain the distribution of device categories;

[0010] The instrument image sampling module is used to classify the instrument complexity according to the instrument category distribution, obtain the instrument complexity, calculate the sampling quantity for the instrument disinfection test according to the instrument complexity, obtain the sampling quantity, randomly select the medical instruments of the sampling quantity, and collect and obtain the medical instrument image set;

[0011] The sampling and testing module is used to identify the surface damage of the medical instruments in the medical instrument image set, obtain the surface damage parameter set, calculate the sampling quantity for the damaged area in combination with the instrument complexity, obtain the sampling quantity for the damaged area, and conduct random sampling biological testing on the selected medical instruments to obtain the test results.

[0012] The beneficial effects of the present invention are as follows: The disinfection supply detection method of the present invention classifies and calculates the sampling quantity by obtaining the categories and quantities of medical instruments in the disinfection supply room and combining the instrument complexity. By considering the overall complexity of medical instruments for sampling quantity calculation and configuration, it improves the comprehensiveness of sampling while ensuring the inspection efficiency as much as possible, realizes high-quality disinfection supply detection of medical instruments, and effectively avoids the problem of insufficient sampling caused by a large variety of instrument types or complex structures. At the same time, by randomly selecting medical instruments and collecting the image set, combined with the surface damage recognition technology, it can accurately identify the damaged areas of the instruments and quantify the damage parameters, introduce the technical direction of the increased probability of bacterial residue caused by surface damage of medical instruments, and calculate and configure the sampling quantity for the damaged area, significantly improving the effectiveness and pertinence of the instrument disinfection supply detection. Dynamically adjusting the sampling quantity based on the instrument complexity and damage parameters further optimizes the detection process, ensures different degrees of detection for instruments with different complexities and different damage situations during the disinfection supply detection process, improves the detection quality and detection sufficiency, and ensures the detection efficiency, thus effectively solving the problems of insufficient disinfection and missed detection in the prior art. The overall solution of the present invention significantly improves the comprehensiveness, accuracy and reliability of the disinfection supply detection of medical instruments through an intelligent and data-driven detection method, providing a strong guarantee for medical safety. Description of the Drawings

[0013] Figure 1 It is a flowchart of a disinfection supply detection method provided by the present invention;

[0014] Figure 2 It is a structural diagram of a disinfection supply detection system provided by the present invention.

[0015] Reference numerals: Instrument category distribution acquisition module 11, Instrument image sampling module 12, Sampling and testing module 13. Detailed Embodiments

[0016] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative efforts belong to the scope of protection of the present invention.

[0017] In the description of the present invention, the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the present invention, "a plurality of" means two or more, unless otherwise specifically defined.

[0018] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration, or explanation". Any embodiment described as "for example" in the present invention is not necessarily construed as being more preferred or having more advantages than other embodiments. In order for any person skilled in the art to implement and use the present invention, the following description is given. In the following description, details are set forth for purposes of explanation. It should be understood that those of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other instances, well-known structures and processes are not elaborated in detail to avoid unnecessary details from obscuring the description of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the broadest scope that conforms to the principles and features disclosed in the present invention.

[0019] Embodiment 1, as Figure 1 shown, the embodiment of the present invention provides a disinfection supply detection method, and the method specifically includes the following steps:

[0020] S100: Obtain the categories and quantities of medical devices being disinfected in the disinfection supply room to obtain the distribution of device categories;

[0021] In the embodiments of the present application, when disinfecting medical devices in batches in the disinfection supply room, first, the categories and quantities of all medical devices currently being disinfected are obtained through data collection, so as to generate the distribution of device categories for medical device sampling inspection decision analysis and analyze the complexity of the medical devices in the current disinfection batch.

[0022] The step S100 in the method provided by the embodiments of the present application includes:

[0023] Obtain the category of each medical device being disinfected in the current disinfection supply room to obtain a plurality of device categories;

[0024] Collect the quantities of medical devices of the multiple device categories to obtain multiple device category quantities, label the multiple device categories, and obtain the device category distribution.

[0025] In the embodiments of the present application, first, the category information of each medical device currently undergoing disinfection is obtained through data collection. Exemplarily, before or after disinfection, the category information of each medical device can be obtained through barcode scanning or RFID radio frequency identification, and identified through the barcodes or RFID tags preset on the medical devices. Furthermore, multiple device categories of multiple medical devices currently undergoing disinfection are obtained. For example, surgical forceps, endoscopes, syringes, etc., thus generating "multiple device categories". Taking surgical forceps as an example, its category can be further subdivided into subcategories such as straight forceps and curved forceps, while endoscopes can be divided into gastroscopes, colonoscopes, etc. according to their uses. In this way, multiple device categories are obtained.

[0026] Furthermore, collect the specific quantity of medical devices under each device category, that is, the quantity of medical devices of the same medical device category in a batch of medical devices. For example, 50 surgical forceps, 20 endoscopes, and 100 syringes, to obtain "multiple device category quantities".

[0027] Furthermore, use the multiple device category quantities to label the multiple device categories to form a device category distribution. For example, through tagging processing, information such as "surgical forceps - 50", "endoscope - 20", "syringe - 100" is associated and stored as data binary groups, and finally a device category distribution is generated. This distribution data not only reflects the proportion of the quantities of various devices in the sterile supply room (such as surgical forceps having the highest proportion), but also provides a basic support for its subsequent complexity classification and sampling inspection strategy.

[0028] S200: According to the device category distribution, perform device complexity classification to obtain device complexity, and according to the device complexity, calculate the sampling inspection quantity of device disinfection tests to obtain the sampling inspection quantity and randomly select the medical devices of the sampling inspection quantity, and collect to obtain a medical device image set;

[0029] In the embodiments of the present application, after obtaining the device category distribution of the medical devices undergoing disinfection in the sterile supply room, according to the structural complexity of different medical devices, perform device complexity classification and calculate the device complexity of the current batch of disinfected medical devices. The greater the device complexity, for example, the smaller the minimum size or the more the structure, the greater the probability of residual bacteria and other contaminants, and the detection scale needs to be increased.

[0030] Therefore, according to the complexity of the instruments, the number of samples for the instrument disinfection test is calculated. The size of the number of samples is positively correlated with the complexity of the instruments. After obtaining the number of samples, randomly select the corresponding number of medical devices from the current batch of medical devices, and collect the images of the randomly selected medical devices to obtain a set of medical device images, so as to identify the surface damage of the randomly selected medical devices, analyze the degree of surface damage of the current batch of medical devices, and further quantify the impact of the surface damage of medical devices on the contamination residues such as bacteria and viruses for detection.

[0031] Step S200 in the method provided by the embodiment of the present application includes:

[0032] Input each instrument category in the instrument category distribution into the instrument complexity classification table, and classify to obtain the complexity distribution;

[0033] Calculate the mean value of all complexities in the complexity distribution to obtain the instrument complexity;

[0034] Adjust and calculate the preset number of samples for the disinfection test according to the ratio of the instrument complexity to the historical average complexity to obtain the number of samples;

[0035] Extract medical devices according to the number of samples and collect images to obtain a set of medical device images.

[0036] In the embodiment of the present application, after obtaining the instrument category distribution in the sterile supply room, each instrument category (such as surgical forceps, endoscopes, syringes, etc.) is input into a pre-constructed instrument complexity classification table for classification, and the complexity corresponding to each instrument category is obtained. Different instrument categories have different complexities, and the complexity is marked for each currently disinfected medical device to obtain the complexity distribution.

[0037] Among them, the construction steps of the instrument complexity classification table include:

[0038] Obtain multiple sample instrument categories, and label to obtain multiple sample complexities according to the minimum size of each sample instrument category, where the size of the minimum size is negatively correlated with the size of the sample complexity;

[0039] Use the multiple sample instrument categories and the multiple sample complexities to construct an index relationship to obtain the instrument complexity classification table.

[0040] In the embodiment of the present application, according to the disinfection record data in the historical time of the sterile supply room, obtain all the instrument categories of the medical devices to be disinfected as multiple sample instrument categories.

[0041] Further, obtain the minimum size of each sample instrument category. For example, measure the minimum size inside the medical device of each sample instrument category through a measuring device. For example, the minimum size of an endoscope may be the lens diameter (such as 2 mm), and the minimum size of a syringe may be the inner diameter of the needle (such as 0.3 mm). In this way, multiple minimum sizes of multiple sample instrument categories are measured.

[0042] Based on the multiple minimum sizes, identify the sample complexity of multiple sample instrument categories. The size of the minimum size is negatively correlated with the size of the sample complexity. That is, the smaller the minimum size, the greater the sample complexity. For example, calculate the ratio of each minimum size to the mean of all minimum sizes, and calculate the reciprocal of this ratio as the sample complexity. Label multiple sample instrument categories to obtain multiple sample complexities. Optionally, the multiple minimum sizes can also be sorted in descending order, and the sample complexity can be identified according to the sorting. For example, label the sample complexity of the largest minimum size as 1, and label the sample complexity of the second smallest minimum size as 2. In this way, multiple sample complexities are obtained through labeling.

[0043] Optionally, the sample complexity can also be labeled based on a formula. For example, complexity = 10 - minimum size × 2, and multiple sample complexities are obtained.

[0044] Use the multiple sample instrument categories and the multiple sample complexities to construct an index relationship, that is, construct a corresponding relationship between the multiple sample instrument categories and the corresponding meanings of the multiple sample complexities. For example, construct it through a database table to obtain an instrument complexity classification table.

[0045] Through the above steps, the complexity is labeled according to the minimum size of the instrument category and the classification table is constructed, which can quantify the complexity of thoroughly disinfecting different categories of medical devices, ensure the accuracy of complexity classification, and provide a reliable basis for the precise adjustment of subsequent sampling inspection strategies.

[0046] Further, calculate the mean of all complexities in the complexity distribution of the current batch of medical devices to obtain the instrument complexity, which reflects the overall structural complexity of the current batch of medical devices.

[0047] Further, calculate the ratio of the instrument complexity to the historical average complexity. The historical average complexity can be the mean of the instrument complexities of multiple batches of medical devices disinfected within a historical period, which reflects the historical average instrument structural complexity level of the medical device disinfection in the sterile supply department. For example, the historical average complexity is 4.5, and the current instrument complexity is 5.3, then the ratio is 1.18. The larger this ratio, the more complex the overall structure of the current batch of medical devices being disinfected, and the greater the probability of pollution residue.

[0048] Then, this ratio is used to adjust and calculate the preset number of samples for disinfection testing to obtain the number of samples. Exemplarily, the ratio is multiplied by the preset number of samples for disinfection testing and rounded up to obtain the adjusted number of samples. The preset number of samples is a fixed number set for the sampling inspection of medical devices previously, for example, it is 100. Then, after multiplying the ratio by the preset number of samples, for example, using 100×1.18 = 118 pieces, the number of samples is 118.

[0049] According to this number of samples, randomly select the corresponding number of medical devices from the current batch of medical devices, and collect the surface images of each medical device to obtain a medical device image set. The medical device images include image features of damage on the surface of the medical device, such as scratches, etc., which will be used as the data basis for subsequent analysis of contamination residues caused by damage to the medical device.

[0050] For example, the images of the surface of the medical device can be collected by an industrial camera, and the image size can be set for unified collection to ensure the data quality of the medical device image set.

[0051] By adjusting the number of samples of medical devices according to the structural complexity of the batch of medical devices, and considering the structural complexity to adjust the disinfection supply detection strategy, the reliability and comprehensiveness of disinfection supply detection can be improved. High-number sampling inspections are carried out on batches of medical devices with complex structures, which improves the pertinence, comprehensiveness and reliability of sampling inspections, and also ensures the detection efficiency.

[0052] S300: Identify the surface layer damage of the medical devices in the medical device image set to obtain a set of surface layer damage parameters. Combine the device complexity to calculate the number of samples for the damaged area, obtain the number of samples for the damaged area, and conduct random sampling biological testing on the selected medical devices to obtain the test results.

[0053] In the embodiment of the present application, after obtaining the medical device image set, the surface layer damage of the medical devices in the medical device images is identified through image recognition in machine learning to obtain a set of surface layer damage parameters. The surface layer damage parameters are, for example, the scratch size on the surface layer of the medical device, etc.

[0054] Then, combining the two dimensions of device complexity and the set of layer damage parameters, calculate the number of samples for the damaged area, obtain the number of samples for the damaged area, and conduct random sampling biological testing on the selected medical devices to determine whether there is an incomplete disinfection situation.

[0055] By comprehensively considering the two dimensions of the complexity of the medical device and the damage parameters of the medical device, taking into account the influence of the characteristics of the medical device on incomplete disinfection of the supply, setting a targeted sampling inspection strategy, improving the reliability, accuracy and efficiency of sampling inspections, further reducing the occurrence of missed inspections of incomplete disinfection, and reasonably utilizing disinfection detection resources.

[0056] Step S300 in the method provided by the embodiment of the present application includes:

[0057] Using supervised training of an integrated convolutional neural network, training an instrument damage recognition channel including J instrument damage recognition branches, where J is a positive integer;

[0058] Obtaining the maximum instrument complexity, and calculating the number of recognition branches K in combination with the instrument complexity, as shown in the following formula:

[0059]

[0060] where K is the number of recognition branches, ROUNDUP is the rounding-up operation, number r is the instrument complexity, number max is the maximum instrument complexity, and J is the number of all instrument damage recognition branches;

[0061] Randomly select K instrument damage recognition branches, and input each medical instrument image in the medical instrument image set into the K instrument damage recognition branches respectively, identify and obtain multiple sets of recognition surface damage parameters, and calculate the mean value to obtain the surface damage parameter set.

[0062] In the embodiment of the present application, first, in order to improve the accuracy of medical instrument surface damage recognition and reasonably utilize the image recognition computing power resources, supervised training of an integrated convolutional neural network is used to train an instrument damage recognition channel including J instrument damage recognition branches, where J is a positive integer. It is used for subsequent damage recognition of medical instrument images.

[0063] In the embodiment of the present application, using supervised training of an integrated convolutional neural network to train an instrument damage recognition channel including J instrument damage recognition branches includes:

[0064] According to the usage detection data of medical instruments, collect a sample medical instrument image set, label the size of the damage in each sample medical instrument image, and obtain a sample surface damage parameter set;

[0065] Randomly select J pieces of supervised training data from the sample medical instrument image set and the sample surface damage parameter set with replacement;

[0066] Based on the integrated convolutional neural network, construct a network structure of J instrument damage recognition branches;

[0067] Use the J pieces of supervised training data to perform supervised training on the J instrument damage recognition branches respectively until the test meets the requirements, and combine to obtain the instrument damage recognition channel.

[0068] In the embodiments of the present application, according to the usage and disinfection supply detection data of medical devices within a historical period, medical device images obtained by collecting surface images of medical devices previously are used as a sample medical device image set, which serves as the input features for subsequent training of J device damage recognition branches.

[0069] Furthermore, the size of the damage in each sample medical device image is marked. For example, the maximum width of the scratch in each sample medical device image is measured and marked as the damage size, such as 0.2mm, 0.5mm, 0.4mm, etc., which serves as the surface damage parameter. In this way, the surface damage parameters of each sample medical device image in the sample medical device image set are marked to obtain a sample surface damage parameter set, which serves as the output features for subsequent training of J device damage recognition branches.

[0070] Furthermore, in order to improve the accuracy of identifying the surface damage parameters in medical device images, ensemble learning is adopted to train J different device damage recognition branches to identify the surface damage parameters of medical device images through multiple different branches, and then the final result is obtained by combining the results of multiple branches to improve the accuracy. Therefore, the J different device damage recognition branches are trained with different training data.

[0071] Specifically, J pieces of supervised training data are randomly selected with replacement from the sample medical device image set and the sample surface damage parameter set. The amount of data in each piece of supervised training data is the same, but the data is not completely the same and there may be partial overlap. For example, the sample medical device image set and the sample surface damage parameter set include 10,000 groups of data, and each group of data includes a sample medical device image and the corresponding sample surface damage parameter. Each time, 4,000 groups of data are randomly selected with replacement as a piece of supervised training data. In this way, J pieces of supervised training data are obtained. The J pieces of supervised training data are not completely the same as each other, and J device damage recognition branches with different performances can be trained to ensure diversity and accuracy, thereby improving the recognition accuracy of the overall device damage recognition channel. Moreover, the amount of training data for a single device damage recognition branch is small, which can improve the convergence efficiency and avoid overfitting.

[0072] Furthermore, based on the convolutional neural network for image recognition in machine learning, an ensemble learning idea is adopted to construct the network structure of J device damage recognition branches. For example, J is 20. The network structures of the J device damage recognition branches are the same, but different network structure parameters are obtained through different supervised training data in subsequent supervised training.

[0073] Each instrument damage recognition branch includes an input layer, multiple convolutional layers, multiple pooling layers, a fully connected layer, and an output layer. Further, J pieces of supervised training data are used to train the J instrument damage recognition branches respectively. For example, the first piece of training data is used to train the first branch network, the second piece of training data is used to train the second branch network, and so on. During the training process, the network parameters are optimized through a loss function (such as cross-entropy loss) until the recognition accuracy rate on the test set reaches a preset threshold (such as 95%).

[0074] During the training process, for example, input the sample medical device image into the instrument damage recognition branch, obtain the surface damage parameters output by it, calculate the error loss through the cross-entropy loss function combined with the real sample surface damage parameters, and then adjust the network parameters such as weights, gradually reducing the error loss until the recognition loss on the test set is less than 5%, that is, after the recognition accuracy rate reaches the preset threshold (such as 95%), the training is completed. The test set is obtained by dividing the J pieces of supervised training data. For example, each piece of supervised training data is divided into a training set and a test set according to a ratio of 8:2 for training and testing.

[0075] After the supervised training of the J instrument damage recognition branches is completed, the J trained instrument damage recognition branches are combined to obtain an instrument damage recognition channel. The trained instrument damage recognition channel can be used to recognize the inner surface damage parameters of the integrated medical device image.

[0076] Further, after the instrument damage recognition channel is trained, calculate the number of current medical device surface damage recognition branches according to the instrument complexity in the foregoing content. Among them, the greater the instrument complexity, the greater the probability of contamination residue in the current batch of medical devices, so it is necessary to improve the accuracy of medical device surface damage recognition to ensure accurate recognition of medical device surface damage and improve the reliability and accuracy of medical device disinfection supply detection. On the contrary, if the instrument complexity is smaller, the probability of contamination residue in the current batch of medical devices is smaller, so the image recognition computing power can be saved, basic medical device surface damage recognition can be carried out, and a relatively small number of spot checks can be performed to improve the processing efficiency and save computing power on the premise of meeting the reliability of medical device disinfection supply detection.

[0077] Specifically, obtain the maximum instrument complexity, that is, the complexity of the largest medical device category, for example, the largest sample complexity in the instrument complexity classification table in the foregoing content.

[0078] Then, according to the maximum instrument complexity and the instrument complexity, calculate the number of recognition branches K as follows:

[0079]

[0080] Wherein, K is the number of recognition branches, and K is greater than or equal to 1 and less than or equal to J, ROUNDUP is the rounding-up operation, number r is the device complexity, number max is the maximum device complexity, and J is the number of all device damage recognition branches. In this way, the greater the current device complexity, the greater the number of recognition branches K, enabling more accurate recognition of device surface damage and improving the accuracy and reliability of the sampling inspection strategy formulation. For example, if the maximum device complexity is 10, the number J of all cutting damage recognition branches is 20, and the current device complexity is 4.5, then calculate 4.5 / 10*20 and round up, and K is obtained as 9, that is, the number of recognition branches is 9.

[0081] Further, according to the number K of recognition branches, randomly select K device damage recognition branches from the J device damage recognition branches, and input each medical device image in the medical device image set collected above into the K device damage recognition branches respectively, and respectively obtain the recognition surface damage parameter sets of each medical device image. Each recognition surface damage parameter set includes K recognition surface damage parameters respectively obtained by the K device damage recognition branches.

[0082] Further, integrate the K recognition surface damage parameters in each recognition surface damage parameter set, specifically calculate the mean value of multiple recognition surface damage parameter sets, and obtain the surface damage parameters of multiple medical device images as the surface damage parameter set.

[0083] By adopting integrated machine learning in the embodiments of the present application to train the device damage recognition channel, considering the device complexity of the current batch of medical devices, and recognizing the surface damage parameters of medical devices, the recognition accuracy and reliability can be improved, and the recognition efficiency and computing power can be saved, thereby improving the reliability of the disinfection supply detection decision of medical devices.

[0084] The step S300 in the method provided by the embodiments of the present application further includes:

[0085] Calculate the average surface damage parameter according to the surface damage parameter set;

[0086] Adjust and calculate the preset disinfection test sampling quantity according to the ratio of the average surface damage parameter to the historical average surface damage parameter to obtain the sampling quantity of the damaged area;

[0087] Calculate and allocate the sampling probability according to the complexity of the extracted medical devices;

[0088] Randomly sample and biologically detect the damaged areas of the extracted medical devices according to the sampling quantity of the damaged area and the sampling probability to obtain the detection results.

[0089] In the embodiments of the present application, based on the set of surface damage parameters obtained by recognition and calculation, which includes the surface damage parameters of the medical devices with a randomly selected sampling quantity, the average value of multiple surface damage parameters is calculated to obtain the average surface damage parameter.

[0090] Further, the historical average surface damage parameter for the recognition of the surface damage of medical devices within the historical time is obtained. For example, in the historical data of the recognition of the surface damage of medical devices before, the average value of all the historical surface damage parameters is calculated to obtain the historical average surface damage parameter, which reflects the average level of the surface damage size of medical devices within the historical time.

[0091] Based on the ratio of the average surface damage parameter to the historical average surface damage parameter, the preset sampling quantity for the disinfection test is adjusted and calculated. The larger the current average surface damage parameter, the greater the degree of surface damage of the current batch of medical devices, and the greater the probability of contamination residues such as bacteria and viruses. Therefore, a larger sampling quantity for the damaged area needs to be set to conduct a more comprehensive and reliable sampling inspection to avoid the situation of incomplete disinfection and undetected problems. The preset sampling quantity for the disinfection test is the quantity of the previously preset random sampling inspection for the disinfection of medical devices, for example, it is 100.

[0092] Exemplarily, the ratio of the average surface damage parameter to the historical average surface damage parameter is calculated. For example, if the average surface damage parameter is 0.35 mm and the historical average surface damage parameter is 0.3 mm, then the ratio is 0.35 mm / 0.3 mm = 1.17. Then, the preset sampling quantity for the disinfection test is multiplied by this ratio for adjustment and calculation. For example, 1.17 * 100 = 117 is used as the sampling quantity for the damaged area.

[0093] Further, according to the complexity of the medical devices with the randomly selected sampling quantity, the distribution calculation of the sampling probability is carried out. For example, the ratio of the complexity of each medical device corresponding to the device category to the sum of the complexities of the device categories of all the randomly selected medical devices is calculated as the sampling probability. For example, the complexity corresponding to the device category of a certain medical device is 3, while the sum of the complexities of the device categories of the randomly selected sampling quantity of medical devices is 341. Then the sampling probability of this medical device is 0.88%. In this way, the multiple sampling probabilities of the medical devices with the randomly selected sampling quantity are calculated.

[0094] Among them, the greater the complexity of the medical device, the greater the probability of being sampled, that is, the greater the probability of the medical device with a greater probability of contamination residues such as bacteria and viruses being sampled, so as to improve the reliability and pertinence of the disinfection supply detection.

[0095] Then, according to the sampling quantity and sampling probability of the damaged area, randomly select the medical devices with the sampling quantity of the damaged area from the medical devices with the sampling quantity. Herein, a medical device can be sampled multiple times. For example, randomly sample 117 times to obtain 117 medical devices, and there can be duplicate medical devices among them. In this way, the sampling is completed, and samples are taken from the damaged areas of the sampled medical devices for biological detection to determine whether there are contaminants such as bacteria and viruses remaining in the damaged areas of the medical devices, and further determine whether there is incomplete disinfection.

[0096] Exemplarily, use a sterile sampling tool (such as a sterile cotton swab, a micro brush or a micropipette) to sample the scratched area on the surface of the medical device. For example, for the scratch on the tip of a surgical forceps, a micro brush can be used to scrape the residue in the scratch. Then, prepare the sample of the residue and conduct detection, such as culture detection or PCR detection, etc., to detect whether there are residues such as bacteria and viruses, so as to determine whether the disinfection is thorough, and finally obtain the detection result.

[0097] The detection result can include the presence or absence of contamination residue, and the determination output is made according to the biological detection result.

[0098] In the embodiment of the present invention, by combining the influence of the structural complexity of the medical device and the surface layer damage on the disinfection residue, a sampling detection strategy after disinfection supply is formulated, and targeted sampling tests can be carried out on batches of medical devices with different structural complexities and surface damage degrees, so as to improve the accuracy and reliability of the disinfection supply detection, maximize the detection rate of incomplete disinfection, ensure the detection efficiency, and improve the quality of disinfection supply.

[0099] A disinfection supply detection method provided by an embodiment of the present invention has at least the following technical effects:

[0100] The disinfection supply detection method provided by the embodiments of the present invention calculates the classification and sampling quantity by obtaining the types and quantities of medical devices in the disinfection supply room and combining with the device complexity. By considering the overall complexity of medical devices for sampling quantity calculation and configuration, it improves the comprehensiveness of sampling while ensuring the inspection efficiency as much as possible, realizes high-quality disinfection supply detection of medical devices, and effectively avoids the problem of insufficient sampling caused by a large variety of device types or complex structures. At the same time, by randomly selecting medical devices and collecting an image set, combined with the surface damage recognition technology, it can accurately identify the damaged areas of the devices and quantify the damage parameters, introduce the technical direction of the increased probability of bacterial residue caused by surface damage of medical devices, and calculate and configure the sampling quantity of the damaged areas, significantly improving the effectiveness and pertinence of device disinfection supply detection. Dynamically adjusting the sampling quantity based on the device complexity and damage parameters further optimizes the detection process, ensures different degrees of detection for devices with different complexities and different damage situations during the disinfection supply detection process, improves the detection quality and detection sufficiency, and ensures the detection efficiency, thus effectively solving the problems of insufficient disinfection and missed detection in the prior art. The overall solution significantly improves the comprehensiveness, accuracy, and reliability of medical device disinfection supply detection through an intelligent and data-driven detection method, providing a strong guarantee for medical safety.

[0101] Embodiment 2, as Figure 2 shown, with the same inventive concept as a disinfection supply detection method in Embodiment 1, the embodiments of the present invention also provide a disinfection supply detection system. The explanation of the disinfection supply detection method in Embodiment 1 also applies to a disinfection supply detection system. The system includes:

[0102] A device category distribution acquisition module 11, configured to obtain the types and quantities of medical devices to be disinfected in the disinfection supply room, and obtain the device category distribution;

[0103] A device image sampling module 12, configured to classify the device complexity according to the device category distribution, obtain the device complexity, calculate the sampling quantity for device disinfection testing according to the device complexity, obtain the sampling quantity, and randomly select the medical devices of the sampling quantity, and collect a medical device image set;

[0104] A sampling detection module 13, configured to perform surface damage recognition of medical devices on the medical device image set, obtain a surface damage parameter set, combine with the device complexity, calculate the sampling quantity of the damaged area, obtain the sampling quantity of the damaged area, and perform random sampling biological detection on the selected medical devices to obtain a detection result.

[0105] Further, the device category distribution acquisition module 11 is further configured to:

[0106] Obtain the category of each medical device being disinfected in the current sterile supply room to obtain multiple device categories;

[0107] Collect the quantities of medical devices of the multiple device categories to obtain multiple device category quantities, and label the multiple device categories to obtain the device category distribution.

[0108] Furthermore, the device image sampling module 12 is further configured to:

[0109] Input each device category within the device category distribution into the device complexity classification table, and classify to obtain the complexity distribution;

[0110] Calculate the mean value of all complexities within the complexity distribution to obtain the device complexity;

[0111] Adjust and calculate the preset disinfection test sampling quantity according to the ratio of the device complexity to the historical average complexity to obtain the sampling quantity;

[0112] Extract medical devices according to the sampling quantity and collect images to obtain a medical device image set.

[0113] Among them, the construction steps of the device complexity classification table include:

[0114] Obtain multiple sample device categories, and label to obtain multiple sample complexities according to the minimum size of each sample device category, where the size of the minimum size is negatively correlated with the size of the sample complexity;

[0115] Use the multiple sample device categories and the multiple sample complexities to construct an index relationship to obtain the device complexity classification table.

[0116] Furthermore, the sampling detection module 13 is further configured to:

[0117] Identify surface damage of the medical devices in the medical device image set to obtain a surface damage parameter set, including:

[0118] Adopt supervised training of an integrated convolutional neural network to train a device damage recognition channel including J device damage recognition branches, where J is a positive integer;

[0119] Obtain the maximum device complexity, and combine the device complexity to calculate the number of recognition branches K as follows:

[0120]

[0121] Among them, K is the number of recognition branches, ROUNDUP is the rounding-up operation, number r is the device complexity, number maxLet \(I\) be the maximum device complexity and \(J\) be the number of all device damage recognition branches;

[0122] Randomly select \(K\) device damage recognition branches, input each medical device image in the medical device image set into the \(K\) device damage recognition branches respectively, identify and obtain multiple sets of recognition surface damage parameters, and calculate the mean value to obtain the surface damage parameter set.

[0123] Among them, using the supervised training of the integrated convolutional neural network, training the device damage recognition channel including \(J\) device damage recognition branches, including:

[0124] According to the usage detection data of the medical device, collect a sample medical device image set, label the size of the damage in each sample medical device image, and obtain the sample surface damage parameter set;

[0125] Randomly select \(J\) pieces of supervised training data from the sample medical device image set and the sample surface damage parameter set with replacement;

[0126] Based on the integrated convolutional neural network, construct the network structure of \(J\) device damage recognition branches;

[0127] Use the \(J\) pieces of supervised training data to perform supervised training on the \(J\) device damage recognition branches respectively until the test meets the requirements, and combine to obtain the device damage recognition channel.

[0128] Furthermore, the sampling detection module 13 is also used for:

[0129] Calculate the average surface damage parameter according to the surface damage parameter set;

[0130] Adjust and calculate the preset disinfection test sampling quantity according to the ratio of the average surface damage parameter to the historical average surface damage parameter to obtain the sampling quantity of the damaged area;

[0131] Calculate and allocate the sampling probability according to the complexity of the extracted medical device;

[0132] Randomly sample and biologically detect the damaged area of the extracted medical device according to the sampling quantity of the damaged area and the sampling probability to obtain the detection result.

[0133] It should be noted that in the above embodiments, the descriptions of each embodiment have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0134] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to the embodiments of the present invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as the combination of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general-purpose computers, special-purpose computers, embedded computers, or other programmable data processing devices to generate a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0136] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or multiple flows and / or blocks Figure 1 or multiple blocks.

[0138] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications to these embodiments once they know the basic inventive concept.

[0139] Obviously, those skilled in the art can make various changes and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalent technologies, the present invention is also intended to include these modifications and variations.

Claims

1. A disinfection supply detection method, characterized in that: The method comprises: Obtain the categories and quantities of medical devices sterilized in the sterilization supply room and obtain the distribution of device categories; According to the distribution of the device categories, classify the device complexity to obtain the device complexity, and calculate the number of random inspections for device disinfection tests based on the device complexity, obtain the random inspection number, and randomly select the medical devices of the random inspection number to obtain a medical device image set; The surface damage of medical devices is identified on the medical device image set to obtain a surface damage parameter set. Combined with the complexity of the device, the number of random inspections of the damaged area is calculated to obtain the number of random inspections of the damaged area and random sampling biological testing is performed on the sampled medical devices to obtain the test results.

2. The disinfection supply detection method according to claim 1, characterized in that: Obtain the categories and quantities of medical devices sterilized in the sterilization supply room and obtain the distribution of device categories, including: Obtain the category of each medical device being sterilized in the current sterilization supply room, and obtain multiple device categories; The number of medical devices of the multiple device categories is collected to obtain the number of multiple device categories, and the multiple device categories are labeled to obtain the device category distribution.

3. The disinfection supply detection method according to claim 1, characterized in that: According to the distribution of the device categories, the device complexity is classified to obtain the device complexity, and the number of random inspections for device disinfection tests is calculated based on the device complexity, including: Input each device category in the device category distribution into the device complexity classification table, and classify to obtain a complexity distribution; Calculating the mean of all complexities in the complexity distribution to obtain the complexity of the device; According to the ratio of the complexity of the device to the historical average complexity, the preset number of random inspections for disinfection testing is adjusted and calculated to obtain the number of random inspections; Medical devices are sampled according to the sampling quantity and images are collected to obtain a medical device image set.

4. The disinfection supply detection method according to claim 3, characterized in that: The steps of constructing the device complexity classification table include: Acquire multiple sample device categories, and annotate and obtain multiple sample complexities according to the minimum size of each sample device category, wherein the size of the minimum size is negatively correlated with the size of the sample complexity; An index relationship is constructed using the multiple sample instrument categories and the multiple sample complexities to obtain an instrument complexity classification table.

5. The disinfection supply detection method according to claim 1, characterized in that: Performing medical device surface damage identification on the medical device image set to obtain a surface damage parameter set includes: The supervised training of the integrated convolutional neural network is adopted to train the instrument damage recognition channel including J instrument damage recognition branches, where J is a positive integer; The maximum device complexity is obtained, and the number of identification branches K is calculated based on the device complexity, as follows: Among them, K is the number of identification branches, ROUNDUP is the rounding up operation, and number r is the complexity of the device, number max is the maximum device complexity, J is the number of all device damage identification branches; K device damage recognition branches are randomly selected, each medical device image in the medical device image set is input into the K device damage recognition branches respectively, a plurality of surface damage recognition parameter sets are obtained by identification, and the surface damage parameter set is obtained by calculating the mean.

6. The disinfection supply detection method according to claim 5, characterized in that: The supervised training of the integrated convolutional neural network is adopted to train the instrument damage recognition channel including J instrument damage recognition branches, including: According to the use and detection data of medical devices, a sample medical device image set is collected, the size of the damage in each sample medical device image is marked, and a sample surface damage parameter set is obtained; Randomly select J supervised training data from the sample medical device image set and the sample surface damage parameter set with replacement; Based on the integrated convolutional neural network, a network structure with J device damage recognition branches was constructed; The J pieces of supervised training data are used to perform supervised training on the J instrument damage recognition branches respectively until the test meets the requirements, and then the instrument damage recognition channel is obtained by combination.

7. The disinfection supply detection method according to claim 1, characterized in that: Combined with the complexity of the device, the number of random inspections of the damaged area is calculated, the number of random inspections of the damaged area is obtained, and random sampling biological tests are performed on the selected medical devices, including: According to the surface damage parameter set, an average surface damage parameter is calculated; According to the ratio of the average surface damage parameter to the historical average surface damage parameter, the preset disinfection test sampling quantity is adjusted and calculated to obtain the sampling quantity of the damaged area; Calculate the probability of obtaining random inspections based on the complexity of the medical devices sampled; According to the number of random inspections and the probability of random inspections of the damaged areas, random sampling biological inspections are performed on the damaged areas of the selected medical devices to obtain the inspection results.

8. A disinfection supply detection system, characterized in that: The system comprises: The device category distribution acquisition module is used to obtain the categories and quantities of medical devices sterilized in the sterilization supply room and obtain the device category distribution; An instrument image sampling module is used to classify the complexity of instruments according to the instrument category distribution, obtain the complexity of instruments, and calculate the number of random inspections for instrument disinfection tests according to the complexity of instruments, obtain the number of random inspections and randomly select medical instruments of the number of random inspections to collect and obtain a medical instrument image set; The sampling detection module is used to identify surface damage of medical devices in the medical device image set, obtain a surface damage parameter set, calculate the number of random inspections of the damaged area in combination with the complexity of the device, obtain the number of random inspections of the damaged area and perform random sampling biological testing on the sampled medical devices to obtain the test results.

Images