A method and system for managing high-value medical consumables
Through digital processing and weight assignment of high-value medical consumables usage situations, automatic identification and accurate matching of high-value consumables are achieved, manual operation errors and inventory shortages are solved, and the correctness of use and continuous supply of inventory is ensured.
Patent Information
- Application Number
- CN202411139471.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-20
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The use of high-value medical consumables in the prior art is usually manually controlled by obligors, which are prone to omissions and errors, and when the high-value consumables are regularly replenished, it is easy to cause temporary shortage of inventory.
By extracting the characteristics of the usage situation of high-value consumables based on historical data, digitizing and giving weights, calculating the correlation with the actual usage situation, selecting matching high-value consumables, and combining and replacing them when inventory is insufficient, ensuring the accuracy of use and Sufficient inventory.
Automatic identification and accurate matching of high-value consumables is achieved, manual operation errors are avoided, and the correctness of use is ensured, and the temporary shortage of high-value consumables is avoided through combined replacement use.
Smart Images

Figure CN118675720B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for managing high-value medical consumables. Background Art
[0002] High-value medical consumables refer to medical consumables that directly act on the human body, have high risks, have strict requirements for safety management and are relatively high in value, such as vascular intervention, non-vascular intervention, orthopedic implants, neurosurgery, electrophysiology, pacemakers, extracorporeal circulation and blood purification, ophthalmic materials, dentistry and artificial valves and other consumables.
[0003] With the continuous development and innovation of science and technology, the variety of high-value consumables has continued to increase, and the usage has also increased significantly, increasing nursing manpower and management costs. Due to the high cost of high-value consumables, they are used in more urgent situations. Therefore, there are high requirements for the correct use of high-value consumables, otherwise it will lead to waste of consumables and untimely treatment. However, the existing use of high-value consumables is usually manually controlled by volunteers. In the complicated medical work, omissions are inevitable, leading to errors. In addition, high-value consumables are replenished regularly, and when they are used, it is easy to cause temporary shortages in inventory. Summary of the invention
[0004] In order to solve the above-mentioned technical problems, a method and system for managing high-value medical consumables are provided. This technical solution solves the problem raised in the above-mentioned background technology that the use of high-value consumables is usually manually controlled by voluntary personnel, and omissions are inevitable in complicated medical work, leading to errors. In addition, high-value consumables are replenished regularly, and when they are used, they are prone to temporary shortages in inventory.
[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is:
[0006] A method for managing high-value medical consumables, comprising:
[0007] Based on the historical data, at least one usage scenario of all high-value consumables is obtained, and features of the usage scenario are extracted to obtain at least one identification feature;
[0008] Use the switching mechanism to convert the identification features into digital features and form the judgment interval of the digital features;
[0009] Mapping digital features to high-value consumables, and summarizing digital features mapped to the same high-value consumables into a digital feature set;
[0010] Assign weights to numerical features based on their importance;
[0011] Acquire a current actual usage scenario, calculate a correlation between the actual usage scenario and the digital feature set, and select at least one digital feature set as a candidate digital feature set based on the correlation;
[0012] Performing a matching check between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set;
[0013] The high-value consumables corresponding to the target digital feature set are used as target high-value consumables;
[0014] Based on the target high-value consumables, replaceable high-value consumables are formed;
[0015] When the inventory of the target high-value consumables is higher than the preset value, the target high-value consumables are used as medical consumables for actual usage scenarios;
[0016] When the inventory of the target high-value consumables does not exceed the preset value, the target high-value consumables and the replaceable high-value consumables are used in combination, and the inventory of the target high-value consumables is replenished.
[0017] Preferably, extracting features from the usage context to obtain at least one identification feature comprises the following steps:
[0018] Obtain usage context, including patient medical records and the purpose of patient consumables;
[0019] Extract relevant information about the use of patient consumables from the patient medical records of the use scenario, classify and identify the relevant information, and divide the relevant information into basic information elements during the classification and identification;
[0020] Based on big data, the proportion of the cases where consumables and basic information elements appear at the same time to the cases where consumables appear is counted as the discrimination ratio;
[0021] When the discrimination ratio of the basic information element is greater than a preset value, the basic information element is necessary information, otherwise, the basic information element is non-essential information;
[0022] Clustering the necessary information to obtain at least one clustering information;
[0023] The clustering information is paired and summarized with the patient consumable usage to obtain identification features.
[0024] Preferably, the method of using a switching mechanism to convert the identification feature into a digital feature and forming a judgment interval of the digital feature comprises the following steps:
[0025] Summarize and remove duplicates of the cluster information and the usage of patient consumables in the identification features, generate at least one relationship element after removing duplicates, and pair the relationship element with the cluster information or the usage of patient consumables, wherein at least one relationship element is a number with different values;
[0026] Replace the clustering information in the identification feature with the relationship element corresponding to the use of the patient's consumables to obtain a digital feature;
[0027] The clustering information in the identification features is classified to obtain non-indicator clustering information and indicator clustering information, wherein the indicator clustering information is information with indicators, and the non-indicator clustering information is judgment information without indicators;
[0028] Extracting indicators from the indexed clustering information in the identification features to obtain judgment indicators, and pairing the judgment indicators with the indexed clustering information;
[0029] Identify indexed clustering information corresponding to the same judgment index in different identification features, and obtain a value range of the judgment index based on at least one indexed clustering information corresponding to the same judgment index;
[0030] The value range of the judgment index corresponds to the indexed clustering information, and the value range of at least one judgment index is used as the judgment interval of the digital feature.
[0031] Preferably, mapping the digital features to high-value consumables comprises the following steps:
[0032] Based on at least one usage of the high-value consumable, pairing the patient consumable usage with the high-value consumable that includes the patient consumable usage;
[0033] The digital features are mapped to high-value consumables corresponding to the patient consumable usage purposes in the digital features.
[0034] Preferably, assigning weights to digital features according to their importance comprises the following steps:
[0035] Obtain high-value consumables corresponding to the digital features as target high-value consumables;
[0036] Based on historical data, the probability of the target high-value consumables appearing when clustering information appears is counted as the weighted probability;
[0037] The clustering information in the identification feature corresponding to the digital feature is obtained, the weight probability of the clustering information corresponding to the digital feature is averaged to obtain the feature weight, and the feature weight is assigned to the digital feature.
[0038] Preferably, calculating the correlation between the actual usage scenario and the digital feature set, and selecting at least one digital feature set based on the correlation comprises the following steps:
[0039] Extracting features from the actual usage scenario to obtain at least one feature to be identified, and converting the feature to be identified into a digital feature to be tested;
[0040] Obtaining a calibration digital feature from the digital feature set, wherein the relationship element of the calibration digital feature is consistent with the relationship element in the digital feature to be measured, and the value of the judgment index of the calibration digital feature is within the value range of the judgment index in the digital feature to be measured;
[0041] The feature weights of the calibrated digital features in the digital feature set are superimposed to obtain a correlation value;
[0042] When the correlation value is greater than a preset value, the digital feature set is used as a candidate digital feature set.
[0043] Preferably, the matching verification between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set comprises the following steps:
[0044] Acquire patient images in actual usage scenarios, identify organs in the patient images, and obtain target organs;
[0045] identifying a digital feature associated with the target organ from the candidate digital feature set as an associated digital feature;
[0046] Based on historical data, the repair ratio of the high-value consumables corresponding to the candidate digital feature set to the relationship elements in the associated digital features is obtained, wherein when the relationship element corresponds to non-indicator clustering information, when the non-indicator clustering information is repaired, the repair ratio of the relationship element is 100%, otherwise, the repair ratio of the relationship element is 0, and when the relationship element corresponds to the indexed clustering information, the repair ratio of the relationship element is the ratio of the judgment indicator in the indexed clustering information that is repaired;
[0047] Accumulating the repair ratios of the relationship elements in the associated digital features in the candidate digital feature set to obtain a comprehensive ratio;
[0048] When the comprehensive ratio is greater than the preset ratio, the candidate digital feature set is used as the target digital feature set.
[0049] Preferably, the forming of replaceable high-value consumables based on the target high-value consumables comprises the following steps:
[0050] Removing candidate digital feature sets from at least one digital feature set to obtain at least one alternative digital feature set;
[0051] Based on the candidate digital feature sets corresponding to the target high-value consumables, select an alternative digital feature set with the greatest overlap with the candidate digital feature set as the similar digital feature set;
[0052] The calculation method of the overlap between the alternative digital feature set and the candidate digital feature set is as follows:
[0053] Count the number of identical digital features in the replaceable digital feature set and the candidate digital feature set as the target number;
[0054] The target number is divided by the total number of digital features in the alternative digital feature set and the candidate digital feature set to obtain the overlap between the alternative digital feature set and the candidate digital feature set.
[0055] Preferably, the combined use of the target high-value consumables and the replaceable high-value consumables comprises the following steps:
[0056] The target mean is obtained by taking the average of the inventory of target high-value consumables and replaceable high-value consumables;
[0057] Subtract the target high-value consumables and the replaceable high-value consumables from the target mean respectively to obtain at least one allocation value;
[0058] Connect the assigned values in sequence to form a ratio to obtain a combined ratio;
[0059] When in use, the target high-value consumables and replaceable high-value consumables are allocated in usage according to the combination ratio.
[0060] A medical high-value consumables management system, used to implement the above-mentioned medical high-value consumables management method, comprising:
[0061] A feature extraction module, wherein the feature extraction module obtains at least one usage scenario of all high-value consumables based on historical data, extracts features of the usage scenario, and obtains at least one identification feature;
[0062] A feature switching module, which uses a switching mechanism to convert the identification features into digital features and form a judgment interval of the digital features;
[0063] A mapping allocation module, wherein the mapping allocation module maps digital features to high-value consumables, and the digital features mapped to the same high-value consumables are aggregated into a digital feature set;
[0064] A weighting module, wherein the weighting module assigns weights to the digital features according to the importance of the digital features;
[0065] A set selection module, wherein the set selection module obtains a current actual usage context, calculates a correlation between the actual usage context and the digital feature set, and selects at least one digital feature set as a candidate digital feature set based on the correlation;
[0066] A matching verification module, wherein the matching verification module performs a matching degree verification on the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set, and uses the high-value consumables corresponding to the target digital feature set as the target high-value consumables;
[0067] A consumable replacement module, wherein the consumable replacement module forms replaceable high-value consumables based on target high-value consumables;
[0068] A consumables usage module, which uses the target high-value consumables as medical consumables for actual usage scenarios when the inventory of the target high-value consumables is higher than a preset value, and uses the target high-value consumables and replaceable high-value consumables in combination when the inventory of the target high-value consumables does not exceed the preset value, and replenishes the inventory of the target high-value consumables.
[0069] Compared with the prior art, the present invention has the following beneficial effects:
[0070] By setting up a feature switching module, a weighting module, a set selection module, a matching verification module and a consumables usage module, the usage scenario is digitally converted, which can facilitate automatic identification and judgment, thereby providing high-value consumables that match the usage scenario, and verifying the high-value consumables and the usage scenario to ensure that there are no errors in their use. When using them, taking into account the inventory situation, for high-value consumables with less inventory, similar high-value consumables are used for combination replacement, so that they can be supported until the next purchase, thereby avoiding errors in manual operation and temporary shortages of high-value consumables. BRIEF DESCRIPTION OF THE DRAWINGS
[0071] Figure 1 It is a schematic diagram of the process of the medical high-value consumables management method of the present invention;
[0072] Figure 2 A schematic diagram of a process of extracting features from a usage scenario to obtain at least one identification feature according to the present invention;
[0073] Figure 3 A schematic diagram of a flow chart of using the switching mechanism of the present invention to convert identification features into digital features and form a judgment interval of the digital features;
[0074] Figure 4 A schematic diagram of the process of mapping digital features to high-value consumables of the present invention;
[0075] Figure 5 A schematic diagram of a process of assigning weights to digital features according to the importance of the digital features of the present invention;
[0076] Figure 6 A schematic diagram of a flow chart of calculating the relevance between an actual use scenario and a digital feature set and selecting at least one digital feature set based on the relevance according to the present invention;
[0077] Figure 7 A schematic diagram of a flow chart of the present invention for performing matching verification between a candidate digital feature set and an actual usage scenario to obtain at least one target digital feature set;
[0078] Figure 8 A schematic diagram of a process for forming replaceable high-value consumables based on target high-value consumables according to the present invention;
[0079] Fig. 9 The present invention is a flow chart of the combined use of target high-value consumables and replaceable high-value consumables. DETAILED DESCRIPTION
[0080] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.
[0081] Reference Figure 1 As shown, a method for managing high-value medical consumables includes:
[0082] Based on the historical data, at least one usage scenario of all high-value consumables is obtained, and features of the usage scenario are extracted to obtain at least one identification feature;
[0083] Use the switching mechanism to convert the identification features into digital features and form the judgment interval of the digital features;
[0084] Mapping digital features to high-value consumables, and summarizing digital features mapped to the same high-value consumables into a digital feature set;
[0085] Assign weights to numerical features based on their importance;
[0086] Acquire a current actual usage scenario, calculate a correlation between the actual usage scenario and the digital feature set, and select at least one digital feature set as a candidate digital feature set based on the correlation;
[0087] Performing a matching check between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set;
[0088] The high-value consumables corresponding to the target digital feature set are used as target high-value consumables;
[0089] Based on the target high-value consumables, replaceable high-value consumables are formed;
[0090] When the inventory of the target high-value consumables is higher than the preset value, the target high-value consumables are used as medical consumables for actual usage scenarios;
[0091] When the inventory of the target high-value consumables does not exceed the preset value, the target high-value consumables and the replaceable high-value consumables are used in combination, and the inventory of the target high-value consumables is replenished.
[0092] Reference Figure 2As shown, extracting features from the usage scenario to obtain at least one identification feature includes the following steps:
[0093] Obtain usage context, including patient medical records and the purpose of patient consumables;
[0094] Extract relevant information about the use of patient consumables from the patient medical records of the use scenario, classify and identify the relevant information, and divide the relevant information into basic information elements during the classification and identification;
[0095] Based on big data, the proportion of the cases where consumables and basic information elements appear at the same time to the cases where consumables appear is counted as the discrimination ratio;
[0096] When the discrimination ratio of the basic information element is greater than a preset value, the basic information element is necessary information, otherwise, the basic information element is non-essential information;
[0097] Clustering the necessary information to obtain at least one clustering information;
[0098] The clustering information is paired and summarized with the patient consumable usage to obtain identification features.
[0099] Identification features are used to judge the usage scenario. Otherwise, the usage scenario cannot be directly identified, and the corresponding high-value consumables cannot be matched to the usage scenario. Features can be extracted based on the patient's medical history and the purpose of the patient's consumables, so as to extract sufficient elements to determine the usage scenario. However, the identification feature is text information, and its identification still has certain difficulties. The computer cannot recognize it quickly. Therefore, it needs to be converted into digital features so that the computer can quickly recognize it.
[0100] Reference Figure 3 As shown, using the switching mechanism, converting the identification features into digital features and forming the judgment interval of the digital features includes the following steps:
[0101] Summarize and remove duplicates of the cluster information and the usage of patient consumables in the identification features, generate at least one relationship element after removing duplicates, and pair the relationship element with the cluster information or the usage of patient consumables, wherein at least one relationship element is a number with different values;
[0102] Replace the clustering information in the identification feature with the relationship element corresponding to the use of the patient's consumables to obtain a digital feature;
[0103] The clustering information in the identification features is classified to obtain non-indicator clustering information and indicator clustering information, wherein the indicator clustering information is information with indicators, and the non-indicator clustering information is judgment information without indicators;
[0104] Extracting indicators from the indexed clustering information in the identification features to obtain judgment indicators, and pairing the judgment indicators with the indexed clustering information;
[0105] Identify indexed clustering information corresponding to the same judgment index in different identification features, and obtain a value range of the judgment index based on at least one indexed clustering information corresponding to the same judgment index;
[0106] The value range of the judgment index corresponds to the indexed clustering information, and the value range of at least one judgment index is used as the judgment interval of the digital feature.
[0107] The method of converting into digital features is to deduplicate the information in the identification features and assign a numerical value to each information. Different information is assigned different score values. Therefore, the information can be judged according to the numerical value. Information corresponding to the same numerical value is the same information, and information corresponding to different numerical values is different information. At the same time, due to the differences in clustering information, some are indicator information, and some are simple judgment information, such as being sick or not. Therefore, it is necessary to form a judgment interval for the indicator information. Therefore, the value range of the judgment indicator is additionally formed to determine whether the corresponding item in the actual usage scenario is within the value range of the judgment indicator.
[0108] Reference Figure 4 As shown, mapping digital features to high-value consumables includes the following steps:
[0109] Based on at least one usage of the high-value consumable, pairing the patient consumable usage with the high-value consumable that includes the patient consumable usage;
[0110] The digital features are mapped to high-value consumables corresponding to the patient consumable usage purposes in the digital features.
[0111] Reference Figure 5 As shown in Figure 1, assigning weights to numerical features according to their importance includes the following steps:
[0112] Obtain high-value consumables corresponding to the digital features as target high-value consumables;
[0113] Based on historical data, the probability of the target high-value consumables appearing when clustering information appears is counted as the weighted probability;
[0114] The clustering information in the identification feature corresponding to the digital feature is obtained, the weight probability of the clustering information corresponding to the digital feature is averaged to obtain the feature weight, and the feature weight is assigned to the digital feature.
[0115] Each digital feature has a different weight of influence on high-value consumables, so they need to be distinguished. Otherwise, indiscriminate judgment based on digital features will lead to errors in the selection of high-value consumables.
[0116] Reference Figure 6 As shown, calculating the correlation between the actual usage scenario and the digital feature set, and selecting at least one digital feature set based on the correlation includes the following steps:
[0117] Extracting features from the actual usage scenario to obtain at least one feature to be identified, and converting the feature to be identified into a digital feature to be tested;
[0118] Obtaining a calibration digital feature from the digital feature set, wherein the relationship element of the calibration digital feature is consistent with the relationship element in the digital feature to be measured, and the value of the judgment index of the calibration digital feature is within the value range of the judgment index in the digital feature to be measured;
[0119] The feature weights of the calibrated digital features in the digital feature set are superimposed to obtain a correlation value;
[0120] When the correlation value is greater than a preset value, the digital feature set is used as a candidate digital feature set.
[0121] Since there is a corresponding relationship between digital feature sets and high-value consumables, the selection of high-value consumables depends on the correlation between the digital feature sets and the actual usage scenarios. The high-value consumables corresponding to digital feature sets with high correlation may be the high-value consumables required for the actual usage scenarios, but further verification is needed.
[0122] Reference Figure 7 As shown, performing a matching check between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set includes the following steps:
[0123] Acquire patient images in actual usage scenarios, identify organs in the patient images, and obtain target organs;
[0124] identifying a digital feature associated with the target organ from the candidate digital feature set as an associated digital feature;
[0125] Based on historical data, the repair ratio of the high-value consumables corresponding to the candidate digital feature set to the relationship elements in the associated digital features is obtained, wherein when the relationship element corresponds to non-indicator clustering information, when the non-indicator clustering information is repaired, the repair ratio of the relationship element is 100%, otherwise, the repair ratio of the relationship element is 0, and when the relationship element corresponds to the indexed clustering information, the repair ratio of the relationship element is the ratio of the judgment indicator in the indexed clustering information that is repaired;
[0126] Accumulating the repair ratios of the relationship elements in the associated digital features in the candidate digital feature set to obtain a comprehensive ratio;
[0127] When the comprehensive ratio is greater than the preset ratio, the candidate digital feature set is used as the target digital feature set.
[0128] The verification method is to estimate the use effect of high-value consumables to determine whether the alternative digital feature set is the target digital feature set. High-value consumables that match the actual usage scenario will inevitably be able to produce better repairs for the symptoms in the actual usage scenario. Therefore, the target digital feature set can be selected based on the repair situation. In this way, the target high-value consumables used in the actual usage scenario can be determined.
[0129] Reference Figure 8 As shown, based on the target high-value consumables, forming replaceable high-value consumables includes the following steps:
[0130] Removing candidate digital feature sets from at least one digital feature set to obtain at least one alternative digital feature set;
[0131] Based on the candidate digital feature sets corresponding to the target high-value consumables, select an alternative digital feature set with the greatest overlap with the candidate digital feature set as the similar digital feature set;
[0132] The calculation method of the overlap between the alternative digital feature set and the candidate digital feature set is as follows:
[0133] Count the number of identical digital features in the replaceable digital feature set and the candidate digital feature set as the target number;
[0134] The target number is divided by the total number of digital features in the alternative digital feature set and the candidate digital feature set to obtain the overlap between the alternative digital feature set and the candidate digital feature set.
[0135] When the target high-value consumables are in use, due to the restriction of the procurement period, some of the target high-value consumables are prone to insufficient inventory. Therefore, it is necessary to form replaceable high-value consumables so that when the inventory is insufficient, they can be combined and matched, thereby reducing the usage of the target high-value consumables, so as to support the procurement time and replenish the target high-value consumables.
[0136] Reference Fig. 9 As shown, the combination of target high-value consumables and replaceable high-value consumables includes the following steps:
[0137] The target mean is obtained by taking the average of the inventory of target high-value consumables and replaceable high-value consumables;
[0138] Subtract the target high-value consumables and the replaceable high-value consumables from the target mean respectively to obtain at least one allocation value;
[0139] Connect the assigned values in sequence to form a ratio to obtain a combined ratio;
[0140] When in use, the target high-value consumables and replaceable high-value consumables are allocated in usage according to the combination ratio.
[0141] A medical high-value consumables management system, used to implement the above-mentioned medical high-value consumables management method, comprising:
[0142] A feature extraction module, wherein the feature extraction module obtains at least one usage scenario of all high-value consumables based on historical data, extracts features of the usage scenario, and obtains at least one identification feature;
[0143] A feature switching module, which uses a switching mechanism to convert the identification features into digital features and form a judgment interval of the digital features;
[0144] A mapping allocation module, wherein the mapping allocation module maps digital features to high-value consumables, and the digital features mapped to the same high-value consumables are aggregated into a digital feature set;
[0145] A weighting module, wherein the weighting module assigns weights to the digital features according to the importance of the digital features;
[0146] A set selection module, wherein the set selection module obtains a current actual usage context, calculates a correlation between the actual usage context and the digital feature set, and selects at least one digital feature set as a candidate digital feature set based on the correlation;
[0147] A matching verification module, wherein the matching verification module performs a matching degree verification on the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set, and uses the high-value consumables corresponding to the target digital feature set as the target high-value consumables;
[0148] A consumable replacement module, wherein the consumable replacement module forms replaceable high-value consumables based on target high-value consumables;
[0149] A consumables usage module, which uses the target high-value consumables as medical consumables for actual usage scenarios when the inventory of the target high-value consumables is higher than a preset value, and uses the target high-value consumables and replaceable high-value consumables in combination when the inventory of the target high-value consumables does not exceed the preset value, and replenishes the inventory of the target high-value consumables.
[0150] Furthermore, the present solution also proposes a storage medium on which a computer-readable program is stored, and when the computer-readable program is called, the above-mentioned medical high-value consumables management method is executed.
[0151] It is understandable that the inventory medium can be a magnetic medium, such as a floppy disk, a hard disk, a magnetic tape; an optical medium, such as a DVD; or a semiconductor medium, such as a solid state drive (SSD).
[0152] To sum up, the advantages of the present invention are: by setting a feature switching module, a weighting module, a set selection module, a matching verification module and a consumables usage module, the usage scenario is digitally converted, which can facilitate automatic identification and judgment, thereby providing high-value consumables that match the usage scenario, and verifying the high-value consumables and the usage scenario to ensure that there will be no errors in their use, and when using them, taking into account the inventory situation, for high-value consumables with less inventory, similar high-value consumables are used for combination replacement, so that they can be supported until the next purchase, thereby avoiding errors in manual operation and temporary shortages of high-value consumables.
[0153] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.
Claims
1. A method for managing high-value medical consumables. It is characterized in that include: Based on the historical data, at least one usage scenario of all high-value consumables is obtained, and features of the usage scenario are extracted to obtain at least one identification feature; Use the switching mechanism to convert the identification features into digital features and form the judgment interval of the digital features; Mapping digital features to high-value consumables, and summarizing digital features mapped to the same high-value consumables into a digital feature set; Assign weights to numerical features based on their importance; Acquire a current actual usage scenario, calculate a correlation between the actual usage scenario and the digital feature set, and select at least one digital feature set as a candidate digital feature set based on the correlation; Performing a matching check between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set; The high-value consumables corresponding to the target digital feature set are used as target high-value consumables; Based on the target high-value consumables, replaceable high-value consumables are formed; When the inventory of the target high-value consumables is higher than the preset value, the target high-value consumables are used as medical consumables for actual usage scenarios; When the inventory of the target high-value consumables does not exceed the preset value, the target high-value consumables and the replaceable high-value consumables are used in combination, and the inventory of the target high-value consumables is replenished; The method of using the switching mechanism to convert the identification feature into a digital feature and forming a judgment interval of the digital feature includes the following steps: Summarize and remove duplicates of the cluster information and the usage of patient consumables in the identification features, generate at least one relationship element after removing duplicates, and pair the relationship element with the cluster information or the usage of patient consumables, wherein at least one relationship element is a number with different values; Replace the clustering information in the identification feature with the relationship element corresponding to the use of the patient's consumables to obtain a digital feature; The clustering information in the identification features is classified to obtain non-indicator clustering information and indicator clustering information, wherein the indicator clustering information is information with indicators, and the non-indicator clustering information is judgment information without indicators; Extracting indicators from the indexed clustering information in the identification features to obtain judgment indicators, and pairing the judgment indicators with the indexed clustering information; Identify indexed clustering information corresponding to the same judgment index in different identification features, and obtain a value range of the judgment index based on at least one indexed clustering information corresponding to the same judgment index; The value range of the judgment index corresponds to the indexed clustering information, and the value range of at least one judgment index is used as the judgment interval of the digital feature; The matching degree verification between the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set comprises the following steps: Acquire patient images in actual usage scenarios, identify organs in the patient images, and obtain target organs; identifying a digital feature associated with the target organ from the candidate digital feature set as an associated digital feature; Based on historical data, the repair ratio of the high-value consumables corresponding to the candidate digital feature set to the relationship elements in the associated digital features is obtained, wherein when the relationship element corresponds to non-indicator clustering information, when the non-indicator clustering information is repaired, the repair ratio of the relationship element is 100%, otherwise, the repair ratio of the relationship element is 0, and when the relationship element corresponds to the indexed clustering information, the repair ratio of the relationship element is the ratio of the judgment indicator in the indexed clustering information that is repaired; Accumulating the repair ratios of the relationship elements in the associated digital features in the candidate digital feature set to obtain a comprehensive ratio; When the comprehensive ratio is greater than the preset ratio, the candidate digital feature set is used as the target digital feature set.
2. A method for managing high-value medical consumables according to claim 1, It is characterized in that The feature extraction of the usage scenario to obtain at least one identification feature comprises the following steps: Obtain usage context, including patient medical records and patient consumables usage; Extract relevant information about the use of patient consumables from the patient medical records of the use scenario, classify and identify the relevant information, and divide the relevant information into basic information elements during the classification and identification; Based on big data, the proportion of the cases where consumables and basic information elements appear at the same time to the cases where consumables appear is counted as the discrimination ratio; When the discrimination ratio of the basic information element is greater than a preset value, the basic information element is necessary information, otherwise, the basic information element is non-essential information; Clustering the necessary information to obtain at least one clustering information; The clustering information is paired and summarized with the patient consumable usage to obtain identification features.
3. A method for managing high-value medical consumables according to claim 2, It is characterized in that Mapping digital features to high-value consumables comprises the following steps: Based on at least one usage of the high-value consumable, pairing the patient consumable usage with the high-value consumable that includes the patient consumable usage; The digital features are mapped to high-value consumables corresponding to the patient consumable usage purposes in the digital features.
4. A method for managing high-value medical consumables according to claim 3, It is characterized in that The step of assigning weights to digital features according to their importance comprises the following steps: Obtain high-value consumables corresponding to the digital features as target high-value consumables; Based on historical data, the probability of the target high-value consumables appearing when clustering information appears is counted as the weighted probability; The clustering information in the identification feature corresponding to the digital feature is obtained, the weight probability of the clustering information corresponding to the digital feature is averaged to obtain the feature weight, and the feature weight is assigned to the digital feature.
5. A method for managing high-value medical consumables according to claim 4, It is characterized in that Calculating the correlation between the actual usage scenario and the digital feature set, and selecting at least one digital feature set based on the correlation comprises the following steps: Extracting features from the actual usage scenario to obtain at least one feature to be identified, and converting the feature to be identified into a digital feature to be tested; Obtaining a calibration digital feature from the digital feature set, wherein the relationship element of the calibration digital feature is consistent with the relationship element in the digital feature to be measured, and the value of the judgment index of the calibration digital feature is within the value range of the judgment index in the digital feature to be measured; The feature weights of the calibrated digital features in the digital feature set are superimposed to obtain a correlation value; When the correlation value is greater than a preset value, the digital feature set is used as a candidate digital feature set.
6. A method for managing high-value medical consumables according to claim 5, It is characterized in that The forming of replaceable high-value consumables based on the target high-value consumables comprises the following steps: Removing candidate digital feature sets from at least one digital feature set to obtain at least one alternative digital feature set; Based on the candidate digital feature sets corresponding to the target high-value consumables, select an alternative digital feature set with the greatest overlap with the candidate digital feature set as the similar digital feature set; The calculation method of the overlap between the alternative digital feature set and the candidate digital feature set is as follows: Count the number of identical digital features in the replaceable digital feature set and the candidate digital feature set as the target number; The target number is divided by the total number of digital features in the alternative digital feature set and the candidate digital feature set to obtain the overlap between the alternative digital feature set and the candidate digital feature set.
7. A method for managing high-value medical consumables according to claim 6, It is characterized in that The combined use of the target high-value consumables and the replaceable high-value consumables comprises the following steps: The target mean is obtained by taking the average of the inventory of target high-value consumables and replaceable high-value consumables; Subtract the target high-value consumables and the replaceable high-value consumables from the target mean respectively to obtain at least one allocation value; Connect the assigned values in sequence to form a ratio to obtain a combined ratio; When in use, the target high-value consumables and replaceable high-value consumables are allocated in usage according to the combination ratio.
8. A medical high-value consumables management system, used to implement the medical high-value consumables management method according to any one of claims 1 to 7, It is characterized in that include: A feature extraction module, wherein the feature extraction module obtains at least one usage scenario of all high-value consumables based on historical data, extracts features of the usage scenario, and obtains at least one identification feature; A feature switching module, which uses a switching mechanism to convert the identification features into digital features and form a judgment interval of the digital features; A mapping allocation module, wherein the mapping allocation module maps digital features to high-value consumables, and the digital features mapped to the same high-value consumables are aggregated into a digital feature set; A weighting module, wherein the weighting module assigns weights to the digital features according to the importance of the digital features; A set selection module, wherein the set selection module obtains a current actual usage context, calculates a correlation between the actual usage context and the digital feature set, and selects at least one digital feature set as a candidate digital feature set based on the correlation; A matching verification module, wherein the matching verification module performs a matching degree verification on the candidate digital feature set and the actual usage scenario to obtain at least one target digital feature set, and uses the high-value consumables corresponding to the target digital feature set as the target high-value consumables; A consumable replacement module, wherein the consumable replacement module forms replaceable high-value consumables based on target high-value consumables; A consumables usage module, which uses the target high-value consumables as medical consumables for actual usage scenarios when the inventory of the target high-value consumables is higher than a preset value, and uses the target high-value consumables and replaceable high-value consumables in combination when the inventory of the target high-value consumables does not exceed the preset value, and replenishes the inventory of the target high-value consumables.
