Intelligent triage classification collaborative scheduling method and device based on fusion three-layer collaboration, and medium

Abstract

The embodiment of the invention provides an intelligent triage classification collaborative scheduling method and equipment based on fusion three-layer collaboration and a medium, and relates to the technical field of data processing, and the method comprises the steps: carrying out the identity binding of an intelligent RFID identification bracelet of a target user through an intelligent triage device of a terminal layer, carrying out the multi-modal data collection, and obtaining a multi-modal data set; for the obtained target user level, dynamically encrypting the target user level, writing the target user level into the intelligent RFID identification bracelet, and transmitting the target user level to the cloud layer; obtaining a resource scheduling scheme; and dynamically encrypting the resource scheduling scheme and the target user level, and pushing the encrypted resource scheduling scheme and the target user level to an HIS system of a target hospital of an application layer to complete resource collaborative scheduling. The problems of manual dependence, static identification, data isolated island and response lag in traditional triage are solved. The technical effects of automatic triage, reversible identification, closed-loop data and resource collaboration are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of data processing, and in particular to an intelligent triage classification collaborative scheduling method, device and medium based on fusion of three-layer collaboration. BACKGROUND

[0002] Currently, disaster site treatment still adopts traditional manual triage classification. The traditional manual triage data shows that the average time for handwriting classification is 62±8 seconds per person, and the misjudgment rate is 18%-25% (according to the Research on Quality Evaluation of Triage Classification in Disaster Site in Chinese Journal of Emergency Medicine, 2024). Moreover, the current triage classification has the following problems:

[0003] Lagging identification medium: Traditional classification identification mostly uses paper wristbands or plastic tags (such as the mechanical classification card disclosed in CN2145490Y), which records the classification level by handwriting. This method is inefficient, the handwriting is easy to blur, and it is easy to be tampered with or replaced accidentally, leading to misdiagnosis.

[0004] Low information carrying capacity: Due to the limitations of physical size and medium, traditional wristbands cannot record key information such as the history of vital signs, allergy history, and medication of the wounded, forming an information island.

[0005] Static and irreversible: As disclosed by Nanjing Medical University, a "wounded triage classification identification wristband" (application number CN201920288592.2), it uses a mechanical color-changing ink capsule, which can quickly display color, but once the color is switched, it cannot be reversed. This means that when the condition of the wounded worsens or improves, the identification cannot be dynamically updated, losing its guiding significance for subsequent treatment, and even possibly misleading the rescue.

[0006] No real-time monitoring capability: There is a time difference from when the wounded wear the wristband to when they are transferred for treatment. During this period, the physiological state of the wounded may deteriorate rapidly. Traditional technology completely lacks the ability to automatically and continuously monitor the vital signs of the wounded during this period and provide early warnings, leaving a large safety blind spot. The average time for transferring wounded in disaster sites is 47 minutes, and 23% of grade II wounded deteriorate to grade I during this period (according to the Analysis of Disaster Wounded Transfer Risk in Trauma Surgery Journal, 2023). Traditional technology cannot provide early warnings.

[0007] It should be noted that the information disclosed in this background section is only intended to increase the understanding of the overall background of the present application, and should not be considered as acknowledging or implying in any form that this information constitutes prior art known to those skilled in the art. SUMMARY

[0008] To address the aforementioned shortcomings or improvement needs of existing technologies, this invention provides an intelligent triage and collaborative scheduling method, device, and medium based on a fusion of three layers. This solves the problems of manual dependence, static labeling, data silos, and delayed response in traditional triage. Through three-layer collaboration, it achieves the technical effects of automated triage, reversible labeling, closed-loop data processing, and resource collaboration. The specific technical solution is as follows:

[0009] According to a first aspect of the present invention, an intelligent triage and classification collaborative scheduling method based on a fusion three-layer collaboration is provided. The method is applied to a triage and collaborative scheduling system comprising a terminal layer, a cloud layer, and an application layer. The method includes: binding the target user's intelligent RFID tag wristband to the target user's identity through an intelligent triage device at the terminal layer, and collecting multimodal data to obtain a multimodal data set; performing embedded state association integration and identification on the multimodal data set to obtain the target user's level; dynamically encrypting the target user's level and writing it into the intelligent RFID tag wristband and transmitting it to the cloud layer, wherein the intelligent RFID tag wristband displays the corresponding level color indicator through an E-Ink flexible display screen; identifying schedulable resources based on the cloud layer, and performing resource scheduling analysis in conjunction with the target user's level to obtain a resource scheduling scheme, wherein the resource scheduling scheme is synchronously transmitted to the intelligent RFID tag wristband; dynamically encrypting the resource scheduling scheme and the target user's level and pushing it to the target hospital's HIS system at the application layer to complete resource collaborative scheduling.

[0010] In one implementation, an intelligent triage device binds the target user's smart RFID tag wristband to their identity and collects multimodal data to obtain a multimodal data set. The following processing is also performed: an integrated medical-grade multi-parameter module is connected via an interface to a finger-clip pulse oximeter, a non-invasive blood pressure cuff, and a forehead thermometer to obtain a first data set; wound images are captured using a high-definition RGB camera, and the location of the injured person is captured using an infrared thermal imaging module to obtain a second data set; voice is collected using a microphone array and commands are recognized using a voiceprint recognition module to obtain a third data set; the first, second, and third data sets are then combined to obtain the multimodal data set.

[0011] In one implementation, the multimodal data set is subjected to embedded state association integration and identification to obtain a target user level, and the following processing is also performed: the multimodal data set is integrated and scored according to a preset weight to obtain an embedded state score; the embedded state score is identified according to a preset scoring rule to obtain the target user level.

[0012] In one implementation, the resource scheduling scheme and the target user level are dynamically encrypted and pushed to the target hospital's HIS system to complete resource collaborative scheduling. The following processing is also performed: an initial encryption analysis is conducted on the resource scheduling scheme and the target user level using a dynamic encryption module to obtain an initial encryption key; the initial encryption key is randomly adjusted multiple times to obtain multiple adjusted encryption keys; key directionality identification is performed based on the multiple adjusted encryption keys and the initial encryption key to determine a target encryption key; the resource scheduling scheme and the target user level are encrypted based on the target encryption key, and the encrypted data is pushed to the target hospital's HIS system.

[0013] In one implementation, based on the plurality of adjusted encryption keys and the initial encryption key, key directionality identification is performed to determine the target encryption key, and the following processes are further performed: traversing the initial encryption key and the plurality of adjusted encryption keys, and performing key sparsity evaluation in conjunction with a historical encryption key database to obtain initial sparsity coefficients and a plurality of adjusted sparsity coefficients; based on the initial sparsity coefficients and the plurality of adjusted sparsity coefficients, performing directionality identification on the initial encryption key and the plurality of adjusted encryption keys to obtain a directional encryption key; and based on the directional encryption key, the initial encryption key, and the plurality of adjusted encryption keys, key directionality identification is performed to obtain the target encryption key.

[0014] In one implementation, based on the directional encryption key, the initial encryption key, and the plurality of adjustment encryption keys, key directionality identification is performed to obtain a target encryption key. The following processing is also performed: using the directional encryption key as the adjustment direction, the initial encryption key and the plurality of adjustment encryption keys are adjusted according to a preset adjustment range to obtain an updated initial encryption key and a plurality of updated adjustment encryption keys; it is determined whether the sparsity coefficients of the updated initial encryption key and the plurality of updated adjustment encryption keys are greater than or equal to the sparsity coefficient corresponding to the directional encryption key. If so, the key corresponding to the maximum sparsity coefficient among the updated initial encryption key and the plurality of updated adjustment encryption keys is taken as the target encryption key; otherwise, the directional encryption key is taken as the target encryption key.

[0015] In one implementation, the multimodal data set is embedded with state association integration and identification to obtain the target user level. The target user level is dynamically encrypted and written into a smart RFID tag bracelet. The following processing is also performed: the smart RFID tag bracelet periodically measures preset indicators and sends them back to the smart triage device for re-examination. If the re-examination result meets the preset deterioration condition, a re-grading calculation is triggered to obtain the re-grading result. The re-grading result is reversibly written into the smart RFID tag bracelet.

[0016] According to a second aspect of the present invention, an electronic device is provided, including a memory and a processor, the memory storing executable instructions, wherein when the processor executes the executable instructions stored in the memory, it implements any step of the first aspect of the present invention.

[0017] A third aspect of the present invention discloses a computer-readable storage medium storing a computer program for performing any step of the first aspect of the present invention.

[0018] Beneficial effects of the embodiments of the present invention:

[0019] In the solution provided by this invention, the intelligent triage device at the terminal layer binds the target user's intelligent RFID tag wristband to their identity and collects multimodal data to obtain a multimodal data set. Then, embedded state association and integration identification are performed on the multimodal data set to obtain the target user's level. The target user level is dynamically encrypted and written into the intelligent RFID tag wristband and transmitted to the cloud layer. The intelligent RFID tag wristband displays the corresponding level color indicator on an E-Ink flexible display screen. Based on the cloud layer, schedulable resources are identified, and resource scheduling analysis is performed in conjunction with the target user level to obtain a resource scheduling plan. The resource scheduling plan is synchronously transmitted to the intelligent RFID tag wristband. Then, the resource scheduling plan and the target user level are dynamically encrypted and pushed to the target hospital's HIS system at the application layer to complete resource collaborative scheduling. This achieves the technical effects of automated triage, reversible tagging, closed-loop data, and resource collaboration. Of course, implementing any product or method of this invention does not necessarily require achieving all of the above advantages simultaneously. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A schematic diagram of the intelligent damage detection classification collaborative scheduling method based on fusion three-layer collaboration provided by the present invention is shown;

[0022] Figure 2 An internal structural diagram of the electronic device provided by the present invention is shown.

[0023] Explanation of reference numerals in the attached diagram: Bus 500, Receiver 501, Processor 502, Transmitter 503, Memory 504, Bus Interface 505. Detailed Implementation

[0024] To facilitate understanding of the present invention, a more complete description of the invention will be given below with reference to the accompanying drawings, which illustrate preferred embodiments of the invention. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein; rather, these embodiments are provided to enable a more thorough and complete understanding of the disclosure of the invention.

[0025] Furthermore, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0026] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0027] Unless otherwise expressly stated, throughout the specification and claims, the term "comprising" or its variations such as "including" or "comprises" shall be understood to include the stated elements or components without excluding other elements or other components.

[0028] The present invention provides an intelligent triage and collaborative scheduling method, device and medium based on three-layer fusion collaboration, which is used to solve the problems of manual dependence, static labeling, data silos and response lag in traditional triage.

[0029] Example 1: See Figure 1 The flowchart of the intelligent damage detection classification and collaborative scheduling method based on fusion three-layer collaboration provided by the embodiments of the present invention is shown. The method is applied to a classification and collaborative scheduling system, which includes a terminal layer, a cloud layer, and an application layer. The method includes:

[0030] A1: The target user's smart RFID tag wristband is bound to the intelligent triage device at the terminal layer, and multimodal data is collected to obtain a multimodal data set;

[0031] In one implementation, the target user's smart RFID tag wristband is bound to an intelligent triage device for identification, and multimodal data is collected to obtain a multimodal data set. Step A1 may further include:

[0032] It integrates a medical-grade multi-parameter module and connects to a finger-clip pulse oximeter, non-invasive blood pressure cuff, and forehead thermometer via an interface to obtain the first set of data;

[0033] Wound images were captured using a high-definition RGB camera, and the location of the injured person was captured using an infrared thermal imaging module to obtain a second dataset.

[0034] Voice is collected using a microphone array and commands are recognized using a voiceprint recognition module to obtain a third dataset.

[0035] The first data set, the second data set, and the third data set are combined to obtain the multimodal data set.

[0036] It should be noted that this invention establishes a three-tiered collaborative architecture of "terminal (on-site) - cloud - application (hospital)," achieving closed-loop management of patient information from collection, transmission, verification to resource scheduling. This breaks down information silos, provides accurate pre-notification information to hospitals, and enables full traceability of injuries. Specifically, the terminal layer completes patient triage and information entry through intelligent triage devices, transmitting the data wirelessly to intelligent classification wristbands and the cloud system. The cloud layer uploads the classification results and coordinates hospital resource scheduling (such as ICU bed allocation and ambulance route planning). Finally, the application layer encrypts and pushes the data to the designated hospital system, breaking down information silos and enabling injury traceability.

[0037] The intelligent RFID tagging wristband uses an NXP MIFARE DESFire EV2 chip (4KB storage capacity, 13.56MHz communication frequency), a flexible display (1.54-inch, E-Ink material), and is powered by a button battery (with a battery life of ≥72 hours). It uses the SM4 algorithm for data encryption. The intelligent triage device is a comprehensive terminal with medical sensor interfaces, a visual acquisition module, a voice interaction module, and embedded AI processing capabilities, used to quickly complete identification and multimodal data acquisition at disaster sites. The medical-grade multi-parameter module refers to a modular device that meets medical device standards and can simultaneously measure multiple physiological parameters, such as blood oxygen, blood pressure, and body temperature. A high-definition RGB camera and an infrared thermal imaging module are used to acquire visible light images and temperature distribution based on thermal radiation, respectively, to analyze the extent of injuries and personnel location. The microphone array is an array of multiple microphones used to pick up sound and support spatial positioning and voice recognition. The voiceprint recognition module is used to confirm the identity of the injured person or recognize verbal commands through voice characteristics.

[0038] The system uses a triage device to scan a smart RFID wristband, binding a unique ID to the patient's personal file. Then, the system sequentially activates multi-parameter acquisition modules—a pulse oximeter, a non-invasive blood pressure cuff, and a forehead thermometer—to acquire basic vital sign data, forming the first dataset. Next, a high-definition RGB camera captures images of the wound, while an infrared thermal imaging module simultaneously locates the patient's surface temperature distribution and position, generating the second dataset. Subsequently, a microphone array collects voice information, and a voiceprint recognition module verifies the patient's identity or recognizes voice commands such as pain or difficulty breathing, generating the third dataset. These three datasets are then integrated with the patient's identity information in chronological order into a multimodal dataset, which is cached locally or uploaded to the cloud for subsequent embedded status recognition and severity assessment.

[0039] By unifying and integrating heterogeneous physiological, visual, and acoustic signals from multiple sources, a dataset with temporal consistency and unique identification is formed. Compared to traditional manual triage, complete data acquisition and identification can be completed on-site within 30 seconds, significantly shortening triage initiation time. For example, in earthquake rescue, the terminal automatically initiates the data acquisition process after detecting that the injured person is wearing a wristband. The data shows blood oxygen saturation of 87%, heart rate of 122 beats / min, forehead temperature of 38.2℃, RGB image detection of bleeding in the left leg, infrared thermal imaging showing an abnormal temperature area on the body surface, and the voiceprint command "difficulty breathing" being recognized. The terminal generates a complete multimodal data set in real time, providing reliable input for subsequent intelligent triage.

[0040] A2: The multimodal data set is embedded with state association integration and identification to obtain the target user level. The target user level is dynamically encrypted and written into the smart RFID tag wristband and transmitted to the cloud layer. The smart RFID tag wristband displays the corresponding level color code through an E-Ink flexible display screen.

[0041] In one implementation, the multimodal data set is subjected to embedded state association integration and identification to obtain the target user level. Step A2 may further include:

[0042] The multimodal dataset is integrated and scored according to preset weights to obtain an embedded state score;

[0043] The embedded status score is identified according to the preset scoring rules to obtain the target user level.

[0044] In one implementation, the multimodal data set is subjected to embedded state association integration and identification to obtain the target user level. The target user level is then dynamically encrypted and written into a smart RFID tag wristband. Step A2 may further include:

[0045] The smart RFID tag wristband periodically measures preset indicators and sends them back to the smart triage device for re-examination. If the re-examination result meets the preset deterioration conditions, it triggers the re-grading calculation and obtains the re-grading result.

[0046] The reclassification results are reversibly written into the smart RFID tag wristband.

[0047] In one possible embodiment, the E-Ink flexible display includes four colors, each corresponding to a different level: Level I red, Level II orange, Level III yellow, and Level IV green. It is sunlight-visible, has ultra-low power consumption, and only consumes power during refresh. The multimodal integrated scoring is a comprehensive state score obtained by weighted fusion of data from different sensor sources, including physiological parameters, image features, and speech content, to reflect the overall vital signs of the injured person. Preset weights represent the importance of each modality determined by the system during the training phase based on a clinical dataset; for example, blood oxygen saturation 40%, blood pressure 20%, image-related injury 30%, and speech-related pain level 10%.

[0048] In one possible embodiment, the preset scoring rules are different levels corresponding to different scores pre-set by those skilled in the art, such as ≥8 points being level I, 6≤S<8 being level II, 3≤S<6 being level III, and S<3 being level IV.

[0049] First, the raw data for each modality are standardized and time-aligned. After filtering, interpolation, and normalization, physiological data are used to obtain physiological state sub-scores, such as deduction for blood oxygen <90%, weighting for heart rate >120 beats / min, and penalty for abnormal body temperature. Visual data is processed by a deep convolutional network to extract indicators such as wound area, bleeding volume, and limb integrity, and mapped to injury severity scores. The speech module extracts pain cries, breathing rhythm, and frequency of conscious response through a speech recognition model and generates a consciousness state score.

[0050] Subsequently, the system performs linear or nonlinear fusion of each sub-score according to preset weights. The embedded state score = 0.5 × physiological sub-score + 0.3 × visual sub-score + 0.2 × speech sub-score. The final output embedded state score is a continuous value in the range of 0-1, representing the individual's immediate health risk index. For example, the on-site detection results of an earthquake victim are: blood oxygen 86%, heart rate 126, blood pressure 90 / 58, body temperature 38.6℃. The visual module identifies a bleeding area of ​​approximately 4.3cm² in the lower limbs, suggesting a possible fracture. The speech module identifies slow speech and delayed response. After fusion with preset weights, the embedded state score is calculated to be 8.2 points, corresponding to Level I. The terminal immediately encapsulates and encrypts this score and the classification result, writing it into the wristband to complete an intelligent and traceable classification closed loop.

[0051] In one embodiment, the preset deterioration condition is a pre-set constraint for reclassification by those skilled in the art, such as a heart rate consistently >120 beats / min for more than 2 minutes or a fall detected by an accelerometer. The periodic measurement preset index refers to the smart RFID tag wristband automatically collecting key vital signs or motion parameters, such as heart rate, blood oxygen saturation, body temperature, acceleration, and body surface temperature, at system-set time intervals, such as 30 seconds, 1 minute, or 5 minutes, during disaster relief or transport.

[0052] At disaster sites or during transport, smart RFID wristbands continuously sample vital signs and cache them locally. When the detection cycle reaches a set time or a significant change is detected, the wristband automatically transmits the latest data to a nearby smart triage device. The triage device compares this data with the previously collected data. If the indicators meet preset deterioration conditions, such as a heart rate >120 beats / min and blood oxygen <90% for two consecutive minutes, a reclassification algorithm is immediately triggered to recalculate the embedded status score and target level. If the new level is more severe than the original level, the system dynamically encrypts and generates an update data packet, and then securely writes the reclassification result to the wristband's storage area, while simultaneously refreshing the color of the E-Ink flexible display (e.g., from orange to red). This achieves the goal of dynamic, reversible, and adaptive updating of the classification.

[0053] A3: Based on the cloud layer, schedulable resources are identified, and resource scheduling analysis is performed in conjunction with the target user level to obtain a resource scheduling scheme, wherein the resource scheduling scheme is synchronously transmitted to the smart RFID tag wristband;

[0054] In one embodiment, after the terminal layer uploads the encrypted target user level and geographic location data to the cloud, the cloud resource scheduling engine immediately executes the schedulable resource identification process. This engine obtains real-time data on bed availability, ICU occupancy rates, and ambulance locations from the hospital's HIS system via API interfaces, and calls the GIS module to calculate the distance and route accessibility between each ambulance and the injured person. Subsequently, the system determines the weight priority based on the target user level; for example, Level I critically injured patients have a weight of 0.9, Level II 0.7, Level III 0.4, and Level IV 0.2. These parameters are then input into a scheduling model based on linear programming or multi-objective optimization, with the objective function being to minimize the overall response time and resource conflict rate. The cloud generates a resource scheduling scheme containing information such as ambulance number, receiving hospital, estimated arrival time, and priority level. This scheme is encrypted and synchronously sent to the corresponding injured person's RFID wristband. The specific encryption method is consistent with the encryption described below. The wristband then displays a brief status prompt on its E-Ink screen, such as "Ambulance dispatched—estimated arrival in 4 minutes," and allows the rescue terminal to query complete instructions via near-field communication.

[0055] A4: Dynamically encrypt the resource scheduling scheme and the target user level and push them to the target hospital's HIS system at the application layer to complete resource collaborative scheduling.

[0056] In one implementation, the resource scheduling scheme and the target user level are dynamically encrypted and pushed to the target hospital's HIS system to complete resource collaborative scheduling. Step A4 may further include:

[0057] The resource scheduling scheme and the target user level are initialized and encrypted using a dynamic encryption module to obtain an initial encryption key.

[0058] The initial encryption key is randomly adjusted multiple times to obtain multiple adjusted encryption keys;

[0059] Based on the multiple adjusted encryption keys and the initial encryption key, key directionality identification is performed to determine the target encryption key;

[0060] The resource scheduling scheme and the target user level are encrypted based on the target encryption key, and the encrypted data is pushed to the target hospital's HIS system.

[0061] In one embodiment, an initial encryption key is obtained using the TLS 1.3 protocol + SM4 encryption resource scheduling scheme and the target user level. Then, the encryption key is randomly and non-linearly adjusted, such as through bit shifting, bit flipping, or hash perturbation, to generate multiple adjusted encryption keys. A key directionality identification module calculates the security score of each candidate key through sparsity and similarity evaluation, selecting the target encryption key with the optimal direction distribution. Finally, this target encryption key is used to perform final encryption, encapsulating the data into a secure data packet with a digital signature. This packet is then pushed to the target hospital's HIS system to complete resource collaborative scheduling.

[0062] In one implementation, key directionality identification is performed based on the plurality of adjusted encryption keys and the initial encryption key to determine the target encryption key. Step A4 may further include:

[0063] The initial encryption key and the plurality of adjusted encryption keys are traversed, and the sparsity of the keys is evaluated in combination with the historical encryption key library to obtain the initial sparsity coefficient and the plurality of adjusted sparsity coefficients.

[0064] Based on the initial sparsity coefficients and multiple adjusted sparsity coefficients, the directionality of the initial encryption key and the multiple adjusted encryption keys is identified to obtain a directional encryption key;

[0065] Based on the directional encryption key, the initial encryption key, and the plurality of adjusted encryption keys, the key directionality is identified to obtain the target encryption key.

[0066] In one implementation, the target encryption key is obtained by identifying the directionality of the key based on the directional encryption key, the initial encryption key, and the plurality of adjustment encryption keys. Step A4 may further include:

[0067] Using the directional encryption key as the adjustment direction, the initial encryption key and the plurality of adjustment encryption keys are adjusted according to a preset adjustment range to obtain an updated initial encryption key and a plurality of updated adjustment encryption keys;

[0068] Determine whether the sparsity coefficients of the updated initial encryption key and the multiple updated adjustment encryption keys are greater than or equal to the sparsity coefficients corresponding to the directional encryption key. If so, use the key corresponding to the maximum sparsity coefficient among the updated initial encryption key and the multiple updated adjustment encryption keys as the target encryption key.

[0069] If not, then the direction encryption key will be used as the target encryption key.

[0070] In one embodiment, the mean cosine similarity between the initial encryption key and the plurality of adjusted encryption keys and the historical encryption keys in the historical encryption key database is calculated. The difference between the calculated result and 1 is used as the sparsity coefficient to obtain the initial sparsity coefficient and the plurality of adjusted sparsity coefficients. A higher sparsity coefficient indicates a lower probability of the key appearing in the historical key database and a more dispersed distribution, thus indicating higher security. The initial sparsity coefficient and the plurality of adjusted sparsity coefficients are compared, and the encryption key corresponding to the maximum value among the initial sparsity coefficient and the plurality of adjusted sparsity coefficients is used as the directional encryption key, which is the currently most secure encryption key.

[0071] Using the directional encryption key as a reference vector, the initial key and adjustment key are fine-tuned within a preset range defined by those skilled in the art to generate an updated key set. The updated key set includes an updated initial encryption key and multiple updated adjustment encryption keys. The sparsity coefficient of the updated key set is calculated again. If any key has a higher sparsity than the directional encryption key, the one with the highest sparsity is selected as the final target key. If the condition is not met, the directional encryption key is retained as the target key. The target key is used to dynamically encrypt the resource scheduling scheme and the target user level, and is destroyed after transmission to ensure single-use and unreproducible results.

[0072] By continuously sparsely distributing the key in the historical key space, the system can significantly improve key unpredictability and reuse security, preventing data packets between the cloud and hospital systems from being replayed or brute-forced.

[0073] Example 2: Figure 2 The diagram shown is a schematic representation of the structure of an exemplary electronic device of the present invention. Figure 2In this document, the bus architecture is represented by bus 500. Bus 500 may include any number of interconnected buses and bridges, and bus 500 connects various circuits including one or more processors represented by processor 502 and memory represented by memory 504. Bus 500 may also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 505 provides an interface between bus 500 and receiver 501 and transmitter 503. Receiver 501 and transmitter 503 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium.

[0074] The memory 504, as a computer-readable storage medium, can be used to store software programs, computer-executable programs, and modules, such as the program instructions / modules corresponding to the intelligent fault detection classification and collaborative scheduling method based on fusion three-layer collaboration in this embodiment of the invention. The processor 502 executes various functional applications and data processing of the computer device by running the software programs, instructions, and modules stored in the memory 504, thereby realizing the aforementioned intelligent fault detection classification and collaborative scheduling method based on fusion three-layer collaboration.

[0075] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

[0076] The foregoing description of specific exemplary embodiments of the invention is for illustrative and explanatory purposes. These descriptions are not intended to limit the invention to the precise forms disclosed, and it will be apparent that many changes and variations can be made in accordance with the foregoing teachings. The exemplary embodiments were chosen and described in order to explain the specific principles of the invention and its practical application, thereby enabling those skilled in the art to implement and utilize various different exemplary embodiments of the invention, as well as various different choices and variations. The scope of the invention is intended to be defined by the claims and their equivalents.

Claims

1. A three-layer collaborative intelligent damage detection and classification collaborative scheduling method, characterized in that, The method is applied to a classification-based collaborative scheduling system, which includes a terminal layer, a cloud layer, and an application layer. The method includes: The intelligent triage device at the terminal layer binds the target user's smart RFID tag wristband to their identity and collects multimodal data to obtain a multimodal data set. The multimodal data set is embedded with state association integration and identification to obtain the target user level. The target user level is dynamically encrypted and written into the smart RFID tag wristband and transmitted to the cloud layer. The smart RFID tag wristband displays the corresponding level color code through an E-Ink flexible display screen. Based on the cloud layer, schedulable resources are identified, and resource scheduling analysis is performed in conjunction with the target user level to obtain a resource scheduling scheme. The resource scheduling scheme is synchronously transmitted to the smart RFID tag wristband. The resource scheduling scheme and the target user level are dynamically encrypted and pushed to the target hospital's HIS system at the application layer to complete resource collaborative scheduling.

2. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 1, characterized in that, The intelligent triage device binds the target user's smart RFID tag wristband to their identity and collects multimodal data to obtain a multimodal data set, including: It integrates a medical-grade multi-parameter module and connects to a finger-clip pulse oximeter, non-invasive blood pressure cuff, and forehead thermometer via an interface to obtain the first set of data; Wound images were captured using a high-definition RGB camera, and the location of the injured person was captured using an infrared thermal imaging module to obtain a second dataset. Voice is collected using a microphone array and commands are recognized using a voiceprint recognition module to obtain a third dataset. The first data set, the second data set, and the third data set are combined to obtain the multimodal data set.

3. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 1, characterized in that, Embedded state association integration and identification are performed on the multimodal dataset to obtain the target user level, including: The multimodal dataset is integrated and scored according to preset weights to obtain an embedded state score; The embedded status score is identified according to the preset scoring rules to obtain the target user level.

4. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 1, characterized in that, The resource scheduling scheme and the target user level are dynamically encrypted and pushed to the target hospital's HIS system to complete resource collaborative scheduling, including: The resource scheduling scheme and the target user level are initialized and encrypted using a dynamic encryption module to obtain an initial encryption key. The initial encryption key is randomly adjusted multiple times to obtain multiple adjusted encryption keys; Based on the multiple adjusted encryption keys and the initial encryption key, key directionality identification is performed to determine the target encryption key; The resource scheduling scheme and the target user level are encrypted based on the target encryption key, and the encrypted data is pushed to the target hospital's HIS system.

5. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 4, characterized in that, Based on the multiple adjusted encryption keys and the initial encryption key, key directionality identification is performed to determine the target encryption key, including: The initial encryption key and the plurality of adjusted encryption keys are traversed, and the sparsity of the keys is evaluated in combination with the historical encryption key library to obtain the initial sparsity coefficient and the plurality of adjusted sparsity coefficients. Based on the initial sparsity coefficients and multiple adjusted sparsity coefficients, the directionality of the initial encryption key and the multiple adjusted encryption keys is identified to obtain a directional encryption key; Based on the directional encryption key, the initial encryption key, and the plurality of adjusted encryption keys, the key directionality is identified to obtain the target encryption key.

6. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 5, characterized in that, Based on the directional encryption key, the initial encryption key, and the plurality of adjusted encryption keys, key directionality identification is performed to obtain the target encryption key, including: Using the directional encryption key as the adjustment direction, the initial encryption key and the plurality of adjustment encryption keys are adjusted according to a preset adjustment range to obtain an updated initial encryption key and a plurality of updated adjustment encryption keys; Determine whether the sparsity coefficients of the updated initial encryption key and the multiple updated adjustment encryption keys are greater than or equal to the sparsity coefficients corresponding to the directional encryption key. If so, use the key corresponding to the maximum sparsity coefficient among the updated initial encryption key and the multiple updated adjustment encryption keys as the target encryption key. If not, then the direction encryption key will be used as the target encryption key.

7. The intelligent damage detection and classification collaborative scheduling method integrating three-layer collaboration as described in claim 1, characterized in that, The system further includes: performing embedded state association integration and identification on the multimodal data set to obtain the target user level; writing the target user level into the smart RFID tag wristband after dynamic encryption; and also includes: The smart RFID tag wristband periodically measures preset indicators and sends them back to the smart triage device for re-examination. If the re-examination result meets the preset deterioration conditions, it triggers the re-grading calculation and obtains the re-grading result. The reclassification results are reversibly written into the smart RFID tag wristband.

8. An electronic device, characterized in that, The electronic device includes: Memory, used to store executable instructions; The processor, when executing the executable instructions stored in the memory, implements the intelligent fault detection and classification collaborative scheduling method of any one of claims 1-7, which integrates three-layer collaboration.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the intelligent fault detection and classification collaborative scheduling method that integrates three layers as described in any one of claims 1-7.

Images