Emergency rescue wounded personnel injury condition assessment method and system

Abstract

The invention provides an emergency rescue wounded person injury condition assessment method and system, and the method comprises the steps: collecting the real-time data of a wounded person at an emergency site, and carrying out the multi-dimensional data fusion of the real-time data through a constructed multi-modal data analysis model, and the data comprise physiological data, injury condition data and language data; according to the fused multi-dimensional feature data, performing triage classification on the injury condition of the wounded person by adopting a machine learning algorithm; and inputting the triage classification result into a pre-established injury condition evaluation model, outputting and obtaining an injury condition result of the wounded person to assist in predicting the illness state of the wounded person, and using the injury condition result to represent the severity of the wounded person. According to the invention, rapid triage and classification, intelligent auxiliary diagnosis and prediction of disease development of patients can be realized in a major disaster accident site, and the purpose of assisting efficient rescue is achieved.

Description

Technical Field

[0001] The present application belongs to the field of emergency rescue technology, and in particular relates to a method and system for assessing the injury of emergency rescue casualties. Background Art

[0002] Trauma, a global health issue, places a significant economic burden on health systems. Effectively reducing early mortality among trauma patients plays a crucial role in alleviating this economic burden. Trauma scoring systems quantify the severity of a patient's injuries and are crucial for accurate diagnosis, emergency department triage, and prognosis prediction.

[0003] Since major disasters and accidents can instantly cause large numbers of critically ill casualties, the systems carried by most rescue equipment are only single-purpose and do not have autonomous triage functions. In the complex and changing disaster relief environment, their role is relatively limited, and they cannot make accurate judgments and thus perform reasonable triage. As a result, the treatment process has shortcomings such as long time consumption, poor accuracy, and lack of quantitative standards, and it is impossible to achieve the purpose of rapid triage, intelligent auxiliary diagnosis, and prediction of patient disease progression. Summary of the Invention

[0004] In view of this, the present application aims to propose a method and system for assessing the injury condition of emergency rescue casualties to solve at least one of the above problems.

[0005] To achieve the above objectives, the technical solution of this application is implemented as follows: In a first aspect, the present application provides a method for assessing the injury of a wounded person during emergency rescue, comprising: Collect real-time data of the injured at the emergency scene and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, wherein the data includes physiological data, injury data and language data; Based on the fused multi-dimensional feature data, a machine learning algorithm is used to triage the injuries of the wounded. The triage results are input into a pre-established injury assessment model, and the injury results of the injured are output to assist in predicting the injured patient's condition. The injury results are used to characterize the severity of the injured patient.

[0006] In a second aspect, based on the same inventive concept, the present application also provides an emergency rescue casualty injury assessment system, comprising: A data acquisition module is configured to collect real-time data of the injured at the emergency scene and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, wherein the data includes physiological data, injury data and language data; The triage module is configured to triage the injured based on the fused multi-dimensional feature data using a machine learning algorithm; The injury assessment module is configured to input the injury classification results into a pre-established injury assessment model and output the injury results of the injured person to assist in assessing the injured person's condition. The injury results are used to characterize the severity of the injured person.

[0007] In a third aspect, based on the same inventive concept, the present application also provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in the first aspect when executing the program.

[0008] In a fourth aspect, based on the same inventive concept, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method described in the first aspect.

[0009] Compared with the prior art, the method and system for assessing the injury status of emergency rescue casualties described in this application have the following beneficial effects: The present application describes a method and system for assessing the injury status of emergency rescue casualties. This method is based on the collection of real-time data on the casualties and can achieve rapid injury classification, intelligent auxiliary diagnosis, and prediction of the patient's condition progression at the scene of major disaster accidents, thereby achieving the purpose of assisting in efficient rescue. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of this application. The exemplary embodiments and descriptions of this application are intended to explain this application and do not constitute an improper limitation on this application. In the accompanying drawings: Figure 1 This is a flow chart of a method for assessing the injury of a wounded person in emergency rescue according to an embodiment of the present application; Figure 2 This is a schematic diagram of the structure of an emergency rescue casualty injury assessment system according to an embodiment of the present application; Figure 3 This is a schematic diagram of the hardware structure of the electronic device described in an embodiment of the present application. DETAILED DESCRIPTION

[0011] In order to make the objectives, technical solutions and advantages of this application more clear, this application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0012] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should have the usual meanings understood by people with ordinary skills in the field to which this application belongs. The "first", "second" and similar words used in the embodiments of the present application do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0013] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0014] See also Figure 1 As shown, this embodiment provides a method for assessing the injury condition of a wounded person in emergency rescue, which specifically includes the following steps: Step S101: collect real-time data of the injured at the emergency scene, and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, where the data includes physiological data, injury data and language data.

[0015] In step S101, the physiological data includes at least body temperature information, electrocardiogram information, blood oxygen concentration information and blood pressure information; Injury data at least include trauma information, burn information and bleeding information; Language data shall at least include the casualty's reaction and state of consciousness.

[0016] Specifically, at the first aid site, for the collection of physiological data, this embodiment uses portable multi-functional sensors (such as electrocardiogram (ECG) detectors, body temperature sensors, blood oximeters, blood pressure monitors, etc.) to monitor the physiological parameters of the injured in real time. These data include key physiological parameters such as heart rate, blood oxygen saturation, respiratory rate, and body temperature, which directly affect the health status of the injured. These sensor data can be uploaded to the central cloud system via wireless networks (such as 5G or Wi-Fi).

[0017] For the collection of injury data, this embodiment uses high-resolution cameras and infrared thermal imaging equipment to capture and record the external injuries of the injured, and combines image processing technology (such as computer vision) to analyze the injured's trauma, burns, bleeding and other information to identify the location and severity of the injured person's injuries.

[0018] If the injured person is able to respond, simple communication can also be carried out through language recognition technology. The language input will be used to assess the injured person's state of consciousness and converted into structured data using natural language processing (NLP) technology.

[0019] In some embodiments, data alignment and preprocessing are performed on the physiological data, injury data, and language data, wherein the preprocessing is normalization processing; Perform feature extraction on pre-processed physiological data, injury data, and language data; The extracted features are input into a multimodal deep neural network model. The multimodal deep neural network architecture consists of multiple branches, each of which is used to learn the features of each feature. Among them, in the multimodal deep neural network model, the extracted physiological features, injury features and language features are spliced ​​into a unified feature vector through the input layer and input into the neural network layer. After each branch learns the features separately, the output results of each branch are fused and the final multi-dimensional feature data is generated through the fully connected layer.

[0020] Specifically, in this embodiment, sensor data, image data, and language data are integrated to build a multimodal data analysis model to more comprehensively understand the patient's condition. The deep learning model can automatically process and fuse data from these different sources. This step specifically includes the following: 1.1 Data alignment and preprocessing To perform data fusion, we must first ensure that the data from different sources are aligned and that the preprocessing is done to remove noise and inconsistencies: Timing alignment: Align multi-source sensor data along the time axis to ensure that each data point corresponds to the same moment; for example, electrocardiogram (ECG) data, blood oxygen data, and the patient's verbal response all need to be synchronized at the same timestamp.

[0021] Standardization and normalization: For data of different dimensions (for example, heart rate and image data), normalization is required to eliminate scale differences and ensure that each data source contributes equally to the final analysis results.

[0022] 1.2 Feature Extraction The feature extraction methods for different data sources vary, and their purpose is to convert raw data into high-dimensional feature representations that can be used for modeling.

[0023] Physiological signal feature extraction: Extract features from physiological signals using signal processing technology (this embodiment uses Fast Fourier Transform (FFT) technology). For example, extract important features such as R waves and heart rate variability from ECG data.

[0024] Image feature extraction: Use deep learning models such as convolutional neural networks (CNNs) to extract features from images, such as the area of ​​the wound, the depth of the burn, and the type of trauma.

[0025] Language feature extraction: Extract the features of language data through language recognition and sentiment analysis technology, such as language clarity, intonation changes, language response, etc.

[0026] 1.3 Model Architecture The multimodal deep neural network (MDNN) architecture adopted in this embodiment consists of multiple branches, each of which processes the feature extraction and learning of a data modality. The outputs of all branches will be fused at a certain layer, and finally the prediction results will be output through a shared fully connected layer.

[0027] The MDNN model architecture is designed as follows: 1.3.1 Physiological Data Branch: This branch processes physiological signal data such as heart rate and blood oxygen, and uses a fully connected layer to learn the characteristics of physiological signals.

[0028] Input: Physiological signals after time domain and frequency domain feature extraction.

[0029] Output: High-level feature representation of the extracted physiological signal.

[0030] 1.3.2 Image Data Branch: This branch processes image data using a convolutional neural network (CNN) to automatically learn injury features in images.

[0031] Input: Image data of the injured (such as trauma photos, infrared thermal images).

[0032] Output: high-level feature representation of the image, such as injury location, wound area, etc.

[0033] 1.3.3 Language Data Branch: This branch processes language data through a recurrent neural network (RNN) to analyze the state of consciousness or emotion in the language.

[0034] Input: Language data of the casualty.

[0035] Output: Emotional or state information of the language response.

[0036] 1.4 Feature Fusion The key to MDNN is how to fuse data from different modalities. The specific fusion method is as follows: Early fusion: Features from different modalities are concatenated directly at the network's input layer to form a long vector, which is then fed into the model for training. This method is simple and suitable for situations where there are strong correlations between features. The fusion method combines physiological signals, image features, and language features into a unified feature vector and feeds it into a shared neural network layer.

[0037] Late fusion: After each branch has learned its features, the outputs of each branch are fused, typically using methods such as weighted averaging or concatenation. Finally, the final prediction output is generated through a fully connected layer. This fusion method involves concatenating or weighted fusion the output features of each branch, then passing them through a fully connected layer to generate the final prediction result, which is then output through the output layer.

[0038] This application uses a pre-built multimodal data analysis model to fuse different types of data. Through data fusion, it can make up for the shortcomings of a single data source and achieve more accurate and comprehensive injury analysis. The model can comprehensively consider multiple data sources and capture all aspects of the injury of the injured.

[0039] Step S102: Based on the fused multi-dimensional feature data, a machine learning algorithm is used to perform injury classification on the injured.

[0040] In some embodiments, based on the multi-dimensional feature data and using a random forest algorithm, the injury classification of the injured person is performed according to the injury classification rules, and the injury classification includes at least trauma, burn, hemorrhage, poisoning, and physiological categories; Trauma classification rules are determined based on the location, type, and severity of the trauma; The burn classification rules are determined based on the burn area and burn depth; The classification rules for bleeding categories are determined based on the amount of bleeding and the location of bleeding; The rules for classifying poisoning are based on the state of consciousness and the degree of delayed response; Physiological classification rules are determined by body temperature, heart rate, blood oxygen concentration and blood pressure.

[0041] Specifically, in this embodiment, in the injury classification, the classification rules are based on the clinical manifestations, injury characteristics, and analysis results of various physiological data of the injured person to judge and determine the severity and type of the injured person's injury. These rules are set based on the evaluation criteria of the injured person's physiological data, image data, and language data.

[0042] Based on multimodal analysis of the above data and using the random forest algorithm to classify the injuries of the injured (it should be noted that the random forest algorithm here is a conventional technical means in this field, and the specific algorithm will not be further described), it can effectively guide emergency personnel to formulate subsequent treatment plans, improve the survival probability and treatment efficiency of the injured. The classification results are as follows: Trauma monitoring is based on sensor data, image data and language data to monitor the injuries and consciousness of the injured. Trauma types are mainly divided into two categories: external trauma (such as contusions, fractures, lacerations, etc.) and internal trauma (such as crush injuries, visceral injuries, brain trauma, etc.).

[0043] Burns are classified according to the depth of skin damage and the area of ​​burn, and are divided into first-degree burns, second-degree burns and third-degree burns.

[0044] Bleeding is divided into mild bleeding, moderate bleeding and severe bleeding according to the amount of bleeding and the location of bleeding.

[0045] Poisoning is determined based on the effect of the drug type (such as sedatives, drugs, etc.) on respiratory rate and heart rate, and is divided into mild poisoning and severe poisoning.

[0046] Physiological injury is determined based on the injured person's body temperature, heart rate, blood oxygen concentration and blood pressure, and is divided into first-degree injury, second-degree injury and third-degree injury.

[0047] This application uses machine learning algorithms (such as decision trees, support vector machines, neural networks, etc.) to train and predict the fused multi-dimensional feature data, automatically identify the injury type of the injured and classify them. This process does not require excessive reliance on manual judgment, and can classify injuries faster and more accurately, thereby improving the efficiency and accuracy of emergency decision-making.

[0048] Step S103: Input the injury classification results into a pre-established injury assessment model, and output the injury results of the injured person to assist in predicting the injured person's condition. The injury results are used to characterize the severity of the injured person.

[0049] Specifically, in this embodiment, the trauma assessment model formula is: Where, represents the comprehensive injury score, represents the trauma index score, represents the burn index score, Indicates the poisoning index score, represents the physiological index score, represents the bleeding index score; 、 、 、 、 They represent the weight coefficients of trauma, burns, poisoning, physiological and bleeding indicators respectively, to indicate the contribution of each factor to the final injury score.

[0050] Among them, the calculation of the trauma score is based on the injury classification results (such as fracture, soft tissue contusion, etc.) as well as the trauma location and severity of the injury.

[0051] Where, Indicates the trauma classification results, Indicates the internal injury classification results, 、 Represent the weight coefficients of external injury and internal injury respectively.

[0052] In some embodiments, grading is performed according to the comprehensive injury score to characterize the severity of the injury.

[0053] Specifically, in this embodiment, the injury classification is performed based on the comprehensive score S. This embodiment establishes four levels of injury classification, and the specific injury conditions of the injured are judged by the score level, as follows: 0-20 points: Mild injury (such as small abrasion, first-degree burn) 21-50 points: moderate injury (such as simple fracture, second-degree burn) 51-80 points: Severe injuries (such as multiple trauma, third-degree burns) 81-100 points: Critical injury (such as massive bleeding with shock, severe poisoning) This application uses the triage results generated by the machine learning algorithm as input into a pre-established injury assessment model to further quantify the patient's injury severity. The injury assessment model will generate a specific injury score through a comprehensive analysis of the patient's injury severity, providing a basis for subsequent rescue decisions.

[0054] The method described in this application collects multi-dimensional data of the injured, performs triage based on multi-dimensional data fusion, combines the injury scores and weight coefficients of different classifications, and finally obtains a comprehensive injury score. The severity of the injured person's injury can be quickly and accurately judged by the score result level. It can realize rapid triage, intelligent auxiliary diagnosis, and prediction of the patient's disease progression at the scene of major disaster accidents, so as to achieve the purpose of facilitating efficient rescue. It should be noted that the above description is limited to some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims may be performed in an order different from that described in the above embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0055] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the embodiments of the present application also provide an emergency rescue casualty injury assessment system.

[0056] like Figure 2 As shown, the emergency rescue casualty injury assessment system includes: The data collection module 11 is configured to collect real-time data of the injured at the emergency scene and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, wherein the data includes physiological data, injury data and language data; The triage module 12 is configured to triage the injured patient's injuries using a machine learning algorithm based on the fused multi-dimensional feature data; The injury assessment module 13 is configured to input the injury classification results into a pre-established injury assessment model and output the injury results of the injured person to assist in assessing the condition of the injured person. The injury results are used to characterize the severity of the injured person.

[0057] For the convenience of description, the above system is described as being divided into various modules according to their functions. Of course, when implementing the embodiments of the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0058] The system of the above embodiment is used to implement the corresponding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0059] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein when the processor executes the program, the method described in any of the above embodiments is implemented.

[0060] Figure 310 is a schematic diagram showing a more specific hardware structure of an electronic device provided in this embodiment. The device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are communicatively connected to each other within the device via the bus 1050.

[0061] The processor 1010 can be implemented using a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0062] The memory 1020 can be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage devices, dynamic storage devices, etc. The memory 1020 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented through software or firmware, the relevant program code is stored in the memory 1020 and is called and executed by the processor 1010.

[0063] The input / output interface 1030 is used to connect to an input / output module to enable information input and output. The input / output module can be configured as a component within the device (not shown) or can be externally connected to the device to provide corresponding functions. Input devices may include a keyboard, mouse, touch screen, microphone, various sensors, etc. Output devices may include a display, speaker, vibrator, indicator light, etc.

[0064] The communication interface 1040 is used to connect to a communication module (not shown) to enable communication between the device and other devices. The communication module can communicate via wired means (such as USB, network cable, etc.) or wireless means (such as mobile network, Wi-Fi, Bluetooth, etc.).

[0065] The bus 1050 comprises a pathway for transmitting information between various components of the device, such as the processor 1010 , the memory 1020 , the input / output interface 1030 , and the communication interface 1040 .

[0066] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040, and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, it will be understood by those skilled in the art that the above device may only include the components necessary to implement the embodiments of this specification, and does not necessarily include all the components shown in the figure.

[0067] The electronic device of the above embodiment is used to implement the corresponding method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be described in detail here.

[0068] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments and methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the method described in any of the above embodiments.

[0069] The computer-readable media of this embodiment includes permanent and non-permanent, removable and non-removable media that can be used to store information by any method or technology. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassettes, tape disk storage or other magnetic storage devices, or any other non-transmission media that can be used to store information that can be accessed by a computing device.

[0070] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0071] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. Within the scope of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0072] Additionally, to simplify the description and discussion, and to avoid obscuring the understanding of the embodiments of the present application, commonly known power / ground connections to integrated circuit (IC) chips and other components may or may not be shown in the provided figures. Furthermore, devices may be shown in block diagram form to avoid obscuring the understanding of the embodiments of the present application, and this also takes into account the fact that the implementation details of these block diagram devices are highly dependent on the platform on which the embodiments of the present application will be implemented (i.e., these details should be fully understood by those skilled in the art). Where specific details (e.g., circuits) are set forth to describe the exemplary embodiments of the present application, it will be apparent to those skilled in the art that the embodiments of the present application can be implemented without these specific details or with variations therefrom. Therefore, these descriptions should be considered illustrative rather than restrictive.

[0073] Although the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may utilize the discussed embodiments.

[0074] The embodiments of the present application are intended to cover all such substitutions, modifications, and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of this application.

Claims

1. A method for assessing the injury of an emergency rescue victim, characterized in that: include: Collect real-time data of the injured at the emergency scene and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, wherein the data includes physiological data, injury data and language data; Based on the fused multi-dimensional feature data, a machine learning algorithm is used to triage the injuries of the wounded. The triage results are input into a pre-established injury assessment model, and the injury results of the injured are output to assist in predicting the injured patient's condition. The injury results are used to characterize the severity of the injured patient.

2. The method according to claim 1, wherein: The physiological data includes at least body temperature information, electrocardiogram information, blood oxygen concentration information and blood pressure information; The injury data at least includes trauma information, burn information and bleeding information; The language data at least includes the reaction and consciousness state of the injured person.

3. The method according to claim 1, wherein: Performing data alignment and preprocessing on the physiological data, injury data, and language data, wherein the preprocessing is normalization processing; performing feature extraction on the pre-processed physiological data, injury data, and language data; Inputting the extracted features into a multimodal deep neural network model, wherein the multimodal deep neural network architecture consists of multiple branches, each branch being used to perform feature learning on each feature; Among them, in the multimodal deep neural network model, the extracted physiological features, injury features and language features are spliced ​​into a unified feature vector through the input layer and input into the neural network layer. After each branch learns the features respectively, the output results of each branch are fused and the final multi-dimensional feature data is generated through the fully connected layer.

4. The method according to claim 1, wherein: Based on the multi-dimensional feature data, the casualty's injury is classified according to the injury classification rules using a random forest algorithm, wherein the injury classification includes at least trauma, burn, hemorrhage, poisoning, and physiological categories; The trauma classification rules are determined based on the location, type, and severity of the trauma; The burn classification rules are determined based on the burn area and burn depth; The bleeding classification rules are determined based on the amount of bleeding and the bleeding site; The poisoning classification rules are determined based on the state of consciousness and the degree of delayed reaction; The physiological classification rules are determined by body temperature, heart rate, blood oxygen concentration and blood pressure.

5. The method according to claim 1, wherein The trauma assessment model formula is specifically: ； Where, represents the comprehensive injury score, represents the trauma index score, represents the burn index score, Indicates the poisoning index score, represents the physiological index score, represents the bleeding index score; 、 、 、 、 They represent the weight coefficients of trauma, burns, poisoning, physiological and bleeding indicators respectively, to indicate the contribution of each factor to the final injury score.

6. The method according to claim 5, characterized in that The trauma index score calculation formula is: Where, Indicates the trauma classification results, Indicates the internal injury classification results, 、 Represent the weight coefficients of external injury and internal injury respectively.

7. The method according to claim 5, characterized in that Also includes: The comprehensive injury score is used to grade the injury to characterize the severity of the injury.

8. An emergency rescue casualty injury assessment system, characterized in that: include: A data acquisition module is configured to collect real-time data of the injured at the emergency scene and perform multi-dimensional data fusion on the real-time data through the constructed multimodal data analysis model, wherein the data includes physiological data, injury data and language data; The triage module is configured to triage the injured based on the fused multi-dimensional feature data using a machine learning algorithm; The injury assessment module is configured to input the injury classification results into a pre-established injury assessment model and output the injury results of the injured person to assist in assessing the injured person's condition. The injury results are used to characterize the severity of the injured person.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 7 when executing the program.

10. A non-transitory computer-readable storage medium, characterized in that in, The non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a computer to execute the method according to any one of claims 1 to 7.

Images