Artificial intelligence-assisted limb junction war wound emergency treatment system and artificial intelligence-assisted limb junction war wound emergency treatment method

Abstract

The invention belongs to the technical field of artificial intelligence auxiliary treatment, and particularly relates to an artificial intelligence auxiliary limb junction war wound emergency treatment system and method. According to the method, through analysis of a multi-dimensional wound information set and a potential blood loss diffusion path, accurate identification of latent blood loss at the junctions of the four limbs is achieved, the defect that potential risks are prone to being ignored due to the fact that a traditional method only depends on dominant hemorrhage features is overcome, and through real-time monitoring of pressure sensing data and video streams and combination of spatial positions and diffusion intensity information, the potential blood loss of the junctions of the four limbs is accurately identified. The compression point and the pressure range can be dynamically adjusted, the effectiveness and safety of hemostasis operation are improved, the compression hemostasis effect is evaluated in real time, auxiliary operation of the next stage is generated, dynamic optimization of the treatment process is achieved, and treatment delay caused by complex wounds or hidden bleeding is reduced.

Description

Technical Field

[0001] This invention belongs to the field of artificial intelligence-assisted rescue technology, specifically relating to an artificial intelligence-assisted emergency rescue system and method for combat trauma at the junction of limbs. Background Technology

[0002] With the increasing complexity of modern battlefield environments and the rise in high-intensity combat missions, the types of injuries faced by battlefield casualties are becoming more diverse and dangerous. Among these, injuries to the limb junctions—those occurring at the junction of the limb and trunk, including the shoulder joint area, armpit, hip joint area, and groin—have become one of the most dangerous and complex types of battlefield injuries. The anatomical structure of this area is extremely complex, with a dense network of blood vessels, intricate arterial and venous systems, and a complex distribution of soft tissues, muscle groups, and nerves. This makes it easy for injuries to cause rapid and significant blood loss, accompanied by severe circulatory dysfunction and potential nerve damage.

[0003] According to clinical and battlefield first aid statistics, the mortality rate of injuries to the limb junction is significantly higher than that of injuries to other parts of the body. This is primarily due to the severity of the injury itself. Injuries to this area are highly concealed; many blood vessels and deep tissue damage are difficult to observe directly on the surface. The injured person may not show obvious signs of bleeding, but internal blood loss may already be severe. This makes traditional first aid assessments inaccurate in reflecting the injury's condition. The blood loss pathways are complex; after injury, blood may spread along multiple deep tissue pathways, creating multiple potential blood loss routes. Simple compression or bandaging is insufficient to effectively stop the bleeding. First aid for injuries to this area is challenging. Under on-site conditions, it is difficult to quickly determine the optimal compression point, control the bleeding range, and dynamically adjust treatment procedures. Especially in the high-pressure, low-resource battlefield environment, this can easily lead to delays in treatment or operational errors, thereby increasing the risk of death for the injured person. Summary of the Invention

[0004] The purpose of this invention is to provide an artificial intelligence-assisted emergency treatment system and method for combat trauma to the limb junction, which can monitor the trauma status in real time, identify occult blood loss, accurately guide compression hemostasis operations, dynamically adjust treatment measures, and improve the treatment effect and survival rate of combat trauma to the limb junction.

[0005] The specific technical solution adopted by this invention is as follows: An artificial intelligence-assisted emergency treatment method for combat trauma at the limb junction includes: Obtain the raw trauma data of the injured person, determine the location of the injury based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the injury is located at the junction of the limbs, filter the raw trauma data and generate a multidimensional trauma information set. Based on a multidimensional trauma information set, and combined with pre-established vascular convergence data and tissue connectivity data at the junctions of the limbs, information on the potential blood loss diffusion path and corresponding diffusion intensity in the body is obtained. Based on the original trauma data, the degree of deviation between overt bleeding characteristics and diffusion intensity information is obtained, and risk indication information is generated to characterize the risk of occult blood loss at the junction of the limbs. Based on potential blood loss diffusion path information and risk indication information, combined with a pre-established surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point are obtained, and the corresponding recommended pressure range is obtained according to the diffusion intensity information, and compression hemostasis operation is performed. During the compression hemostasis procedure, pressure sensor data and current video stream data are acquired, and the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted in conjunction with information on potential blood loss diffusion paths. After completing the compression hemostasis procedure, raw trauma data and potential blood loss diffusion path information are continuously collected, and the effect of the current compression hemostasis treatment is obtained. Based on the effect of the current compression hemostasis treatment, auxiliary operations for the next stage of emergency treatment are generated. Based on a multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints.

[0006] In a preferred embodiment, raw trauma data of the injured person is obtained, the location of the trauma is determined based on the raw trauma data, and it is determined whether the trauma is located at the junction of the limbs. If the trauma is located at the junction of the limbs, the raw trauma data is filtered to generate a multidimensional trauma information set, including: Acquire raw trauma data from the injured, including multispectral image data, real-time video stream data, and time-series data of vital signs; The location of the injury is extracted based on the original trauma data, and the location of the injury is compared with the preset anatomical location range of the junction of the human limbs to determine whether the location of the injury is located at the junction of the limbs. If the location of the injury is determined to be at the junction of the limbs, the original trauma data is filtered and reconstructed according to the screening rules corresponding to the trauma risk perception target at the junction of the limbs, generating a multidimensional trauma information set for the junction of the limbs.

[0007] If it is determined that the injury did not occur at the junction of the limbs, treatment for the junction of the limbs will not be performed.

[0008] In a preferred embodiment, based on a multidimensional trauma information set and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, information on the potential blood loss diffusion pathway and corresponding diffusion intensity within the body is obtained, including: Multispectral image data, real-time video stream data, and vital sign time-series data were extracted based on a multidimensional trauma information set; Based on multispectral image data, the diffusion direction and range of potential blood loss within the tissue are obtained, and spatial diffusion constraint information is generated by combining pre-established vascular convergence data and tissue connectivity data at the limb junctions. The rate of change of the bleeding status on the body surface over time is obtained based on real-time video stream data, and combined with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs, time evolution constraint information is generated. Based on the time series data of vital signs, the impact of potential blood loss on the overall circulatory status of the injured person is obtained, and combined with the pre-established vascular convergence data and tissue connectivity data of the limb junction, physiological response constraint information is generated. Based on spatial diffusion constraint information, temporal evolution constraint information, and physiological response constraint information, and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, the diffusion process of potential blood loss in the body is constrained and combined to obtain the diffusion sequence and coverage of potential blood loss in the body, and to generate potential blood loss diffusion path information. Based on the spatial diffusion constraints, temporal evolution constraints, and physiological response constraints corresponding to the potential blood loss diffusion path information in different diffusion segments, diffusion intensity information corresponding to the potential blood loss diffusion path information is obtained.

[0009] In a preferred embodiment, based on raw trauma data, the degree of deviation between overt bleeding characteristics and diffusion intensity information is obtained to generate risk indication information for characterizing the risk of occult blood loss at the limb junction, including: Based on the raw trauma data, time-series data of vital signs are extracted, and the trajectory of changes in vital signs of the patient's overall circulatory status over time is obtained based on the time-series data of vital signs. Based on the raw trauma data, the overt bleeding characteristics of the trauma site are extracted, and combined with the potential blood loss diffusion path information and the corresponding diffusion intensity information, the expected range of change of vital signs trajectory under the action of the potential blood loss diffusion path is obtained. Determine whether the actual changes in the trajectory of vital signs are within the expected range; If the actual changes in the trajectory of vital signs are not within the expected range, the deviation of the actual changes from the expected range is obtained and marked as the boundary occult blood loss deviation interval. If the actual changes in the trajectory of vital signs are within the expected range, it is determined that there is no occult blood loss deviating from the range at the junction of the limbs; Obtain the duration and magnitude of deviation from the occult blood loss range at the junction, and generate risk indication information.

[0010] In a preferred embodiment, based on potential blood loss diffusion path information and risk indication information, combined with a pre-established surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point are obtained, and according to the diffusion intensity information, the corresponding recommended pressure range is obtained, and compression hemostasis is performed, including: Obtain a surface compression blood loss inhibition table, which includes information on multiple potential blood loss diffusion paths and a set of candidate surface compression locations corresponding to each potential blood loss diffusion path. Based on the potential blood loss diffusion path information, obtain the corresponding set of candidate body surface compression locations from the body surface compression blood loss inhibition table; Obtain the degree of inhibition matching between each surface compression location in the candidate set of surface compression locations and the potential blood loss diffusion path; Based on the degree of inhibition matching and combined with risk indication information, each pressure point in the candidate body surface pressure location set is screened to obtain the three-dimensional coordinates of the pressure point for applying pressure to stop bleeding on the body surface. The potential blood loss diffusion path information is mapped to the patient's body surface coordinate system, and the body surface action area covering the preset spatial radius is constructed with the three-dimensional coordinates of the compression point as the center. Based on the spatial distance information between the potential blood loss diffusion path information and the three-dimensional coordinates of the compression point, the spatial distance information is mapped to a preset set of compression pressure level intervals to obtain the pressure level interval corresponding to the three-dimensional coordinates of the compression point. Based on the pressure level range, obtain the recommended pressure range corresponding to the three-dimensional coordinates of the compression point, and perform compression hemostasis based on the three-dimensional coordinates of the compression point and the recommended pressure range.

[0011] In a preferred embodiment, during the compression hemostasis procedure, pressure sensor data and current video stream data are acquired, and combined with information on potential blood loss diffusion paths, the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted, including: During the compression hemostasis procedure, pressure sensor data and current video stream data are acquired. The actual pressure distribution characteristics of the current body surface compression state at the three-dimensional coordinates of the compression point are obtained based on pressure sensor data, and the hemostasis response characteristics are obtained based on the current video stream data. The hemostasis response characteristics include the body surface color change state, bleeding pattern and local tissue collapse state. The degree of blood loss inhibition coverage under the current body surface compression state is obtained based on the actual pressure distribution characteristics and hemostasis response characteristics. Determine whether the blood loss inhibition coverage is lower than the preset blood loss inhibition threshold; If the coverage of blood loss inhibition is lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is deemed insufficient. If the coverage of blood loss inhibition is not lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is considered sufficient. The value of blood loss inhibition is obtained based on the blood loss inhibition coverage being lower than a preset blood loss inhibition threshold; Obtain spatial distance information between potential blood loss diffusion paths and the three-dimensional coordinates of the compression point; Based on spatial distance information and blood loss inhibition values, the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted.

[0012] In a preferred embodiment, after applying pressure to stop the bleeding, raw trauma data and information on potential blood loss pathways are continuously collected, and the effectiveness of the current pressure hemostasis is obtained. Based on the effectiveness of the current pressure hemostasis, auxiliary procedures for the next stage of emergency treatment are generated, including: After completing the compression hemostasis operation at the target compression point, continuously acquire the patient's original trauma data and potential blood loss diffusion path information, extract vital sign time series data from the original trauma data and mark it as updated vital sign time series data. At the same time, continuously acquire the patient's potential blood loss diffusion path information and mark it as updated potential blood loss diffusion path information. The degree of hemostatic inhibition of the current compression hemostasis operation is obtained based on the updated vital signs time series data and the updated potential blood loss diffusion path information; The effectiveness of the current compression hemostasis treatment is determined based on the degree of hemostasis inhibition. Based on the current effect of compression hemostasis, auxiliary procedures for the next stage of emergency treatment are generated. These auxiliary procedures include compression adjustment methods, supplementary hemostasis measures, and emergency treatment procedures.

[0013] In a preferred embodiment, based on a multidimensional trauma information set, risk indication information, the effectiveness of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints, including: Based on the multidimensional trauma information set, risk indication information, the effect of compression hemostasis and corresponding auxiliary operations, comprehensive status information is generated. The comprehensive status information includes patient status update information, on-site treatment information, medical supply demand information and evacuation demand information. Under preset resource constraints, the comprehensive disposal requirements for each disposal operation are generated based on the comprehensive status information. Based on the comprehensive handling needs corresponding to each handling operation, a collaborative handling plan is generated by sorting them according to preset priorities and on-site conditions.

[0014] The present invention also provides an artificial intelligence-assisted emergency treatment system for combat trauma to the limb junction, used in the aforementioned artificial intelligence-assisted emergency treatment method for combat trauma to the limb junction, comprising: The multidimensional trauma module is used to acquire the raw trauma data of the injured person, obtain the location of the trauma based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the trauma is located at the junction of the limbs, the raw trauma data is filtered to generate a multidimensional trauma information set. The potential blood loss module, based on a multidimensional trauma information set and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, obtains information on the potential blood loss diffusion path and corresponding diffusion intensity within the body. The overt bleeding module is used to obtain the degree of deviation between overt bleeding characteristics and diffusion intensity information based on the original trauma data, and generate risk indication information to characterize the risk of occult blood loss at the junction of the limbs; The compression hemostasis module, based on potential blood loss diffusion path information and risk indication information, combined with a pre-established body surface compression blood loss inhibition table, obtains the three-dimensional coordinates of the compression point, and obtains the corresponding recommended pressure range according to the diffusion intensity information, and performs compression hemostasis operation. The hemostasis adjustment module is used to acquire pressure sensor data and current video stream data during the compression hemostasis operation, and adjust the three-dimensional coordinates of the compression point and the recommended pressure range in combination with potential blood loss diffusion path information. The auxiliary operation module is used to continuously collect raw trauma data and potential blood loss diffusion path information after the compression hemostasis operation is completed, and to obtain the effect of the current compression hemostasis treatment. Based on the effect of the current compression hemostasis treatment, it generates auxiliary operations for the next stage of emergency treatment. The collaborative treatment module generates a collaborative treatment plan based on a multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, under preset resource constraints.

[0015] And, an artificial intelligence-assisted terminal for emergency treatment of combat trauma at the limb junction, comprising: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors enable artificial intelligence-assisted emergency treatment methods for traumatic injuries at the limb junction.

[0016] The technical effects achieved by this invention are as follows: This invention achieves accurate identification of occult blood loss at the limb junctions by using a multidimensional trauma information set and potential blood loss diffusion path analysis. It overcomes the shortcomings of traditional methods that rely solely on overt bleeding characteristics and easily overlook potential risks. Through real-time monitoring of pressure sensor data and video streams, combined with spatial location and diffusion intensity information, it can dynamically adjust the compression point and pressure range, improving the effectiveness and safety of hemostasis. It also provides real-time evaluation of the compression hemostasis effect and generates auxiliary operations for the next stage, achieving dynamic optimization of the treatment process. This reduces treatment delays caused by complex trauma or occult bleeding. Furthermore, it comprehensively analyzes the patient's condition, on-site operations, material needs, and evacuation requirements, generating collaborative treatment plans under resource constraints. This achieves overall optimization of treatment operations and resource allocation, improving emergency treatment efficiency. Based on data-driven multidimensional information analysis and dynamic adjustment mechanisms, it can adapt to different types of limb junction injuries and is compatible with different treatment resource conditions and on-site environments, demonstrating good repeatability and potential for widespread application. Attached Figure Description

[0017] Figure 1 This is a flowchart of the method provided by the present invention; Figure 2 This is a system module diagram provided by the present invention. Detailed Implementation

[0018] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0019] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0020] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in a preferred embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that mutually excludes other embodiments.

[0021] Furthermore, the present invention will be described in detail with reference to the schematic diagrams. When describing the embodiments of the present invention in detail, the schematic diagrams are merely examples for ease of explanation and should not limit the scope of protection of the present invention.

[0022] Please see the appendix Figure 1 As shown, an artificial intelligence-assisted emergency treatment method for combat trauma at the limb junction is provided, including: S1. Obtain the raw trauma data of the injured person, determine the location of the injury based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the injury is located at the junction of the limbs, filter the raw trauma data and generate a multidimensional trauma information set. S2. Based on a multidimensional trauma information set, and combined with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs, information on the potential blood loss diffusion path and corresponding diffusion intensity information in the body is obtained. S3. Based on the original trauma data, obtain the degree of deviation between the overt bleeding characteristics and the diffusion intensity information, and generate risk indication information to characterize the risk of occult blood loss at the junction of the limbs; S4. Based on potential blood loss diffusion path information and risk indication information, combined with the pre-established body surface compression blood loss inhibition table, obtain the three-dimensional coordinates of the compression point, and according to the diffusion intensity information, obtain the corresponding recommended pressure range, and implement compression hemostasis operation. S5. During the compression hemostasis operation, acquire pressure sensor data and current video stream data, and combine them with potential blood loss diffusion path information to adjust the three-dimensional coordinates of the compression point and the recommended pressure range. S6. After completing the compression hemostasis operation, continuously collect raw trauma data and potential blood loss diffusion path information, and obtain the current compression hemostasis effect. Based on the current compression hemostasis effect, generate auxiliary operations for the next stage of emergency treatment. S7. Based on the multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints.

[0023] As described in steps S1 to S6 above, raw trauma data of the injured person is collected, including multispectral images of the injured area, real-time video streams, and time-series data of vital signs. By analyzing the location of the trauma and determining whether it is located at the junction of the limbs, traumas at the junction are screened and reconstructed to generate a multidimensional trauma information set reflecting blood status, tissue connectivity characteristics, and changes in vital signs. Based on the multidimensional trauma information set, combined with vascular convergence data and tissue connectivity data at the junction of the limbs, the diffusion paths of potential blood loss in the injured person and the diffusion intensity information of each path are obtained. The overt bleeding characteristics are compared with the diffusion intensity of potential blood loss to obtain the degree of deviation and generate risk indication information. Information is used to characterize the occult blood loss risk range and its severity in the injured person. Combined with potential blood loss diffusion paths, risk indicator information, and a surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point and the corresponding recommended pressure range are obtained to achieve precise compression hemostasis in the target area. During the operation, pressure sensor data and video stream data are collected in real time to dynamically monitor the hemostasis effect. The compression point and pressure range are adjusted according to the blood loss inhibition coverage and spatial relationship to ensure maximum hemostasis. After the initial compression hemostasis operation is completed, trauma data and potential blood loss diffusion information are continuously collected to obtain the compression hemostasis treatment effect. Based on the real-time treatment effect, the next... The auxiliary operations during this phase include pressure adjustment, supplementary hemostasis, or other emergency treatments. By integrating multidimensional trauma information, risk indications, the effectiveness of pressure hemostasis, and auxiliary operations, a collaborative treatment plan is generated under resource constraints (such as the quantity of available medical supplies, the capabilities of emergency personnel, and evacuation conditions). Through multidimensional trauma information and potential blood loss diffusion path analysis, it achieves accurate identification of occult blood loss at the limb junctions, overcoming the shortcomings of traditional methods that rely solely on overt bleeding characteristics and easily overlook potential risks. Through real-time monitoring of pressure sensor data and video streams, combined with spatial location and diffusion intensity information, the pressure point and pressure range can be dynamically adjusted. This technology improves the effectiveness and safety of hemostasis, assesses the effect of compression hemostasis in real time, and generates auxiliary operations for the next stage, achieving dynamic optimization of the treatment process. It reduces treatment delays caused by complex trauma or hidden bleeding, comprehensively analyzes the patient's condition, on-site operations, material needs, and evacuation requirements, and generates collaborative treatment plans under resource constraints. This achieves overall optimization of treatment operations and resource allocation, improving emergency treatment efficiency. Based on a data-driven multidimensional information analysis and dynamic adjustment mechanism, it can adapt to different types of limb junction injuries and is compatible with different treatment resource conditions and on-site environments, showing good repeatability and potential for widespread application.

[0024] It should be noted that, in order to enhance the practical application capabilities in complex battlefield environments, the data acquisition equipment and artificial intelligence terminals have been optimized for battlefield adaptability. For environments with vibration, low temperatures, and electromagnetic interference, the multispectral image acquisition equipment and pressure sensing equipment adopt seismic-resistant structural design, low-temperature operation management, and anti-interference communication mechanisms to ensure the stability and reliability of data acquisition. Simultaneously, in terms of data processing, a local lightweight artificial intelligence model is used for real-time injury identification and risk assessment, combined with a hierarchical calculation strategy to reduce real-time decision-making latency. It can operate independently even under communication constraints. Furthermore, to meet the needs of rapid deployment of battlefield first aid, the equipment adopts a modular and portable structure and a one-button start-up automatic calibration mechanism, and reduces operational complexity through graphical prompts, enabling first aid personnel to quickly complete trauma assessment and treatment decision support in high-intensity combat environments.

[0025] In a preferred embodiment, raw trauma data of the injured person is acquired, the location of the trauma is determined based on the raw trauma data, and it is determined whether the trauma is located at the junction of the limbs. If the trauma is located at the junction of the limbs, the raw trauma data is filtered to generate a multidimensional trauma information set, including: S101. Obtain the raw trauma data of the injured person, including multispectral image data, real-time video stream data and time-series data of vital signs; S102. Extract the location of the injury based on the original trauma data, and compare the location of the injury with the preset anatomical location range of the junction of the human limbs to determine whether the location of the injury is located at the junction of the limbs. If the location of the injury is determined to be at the junction of the limbs, the original trauma data is filtered and reconstructed according to the screening rules corresponding to the trauma risk perception target at the junction of the limbs, generating a multidimensional trauma information set for the junction of the limbs.

[0026] If it is determined that the injury did not occur at the junction of the limbs, treatment for the junction of the limbs will not be performed.

[0027] As described in steps S101 to S102 above, the raw trauma data of the injured person is acquired, including multispectral image data (used to characterize the blood infiltration status and bleeding spread range of tissues at different depths of the trauma site, reflecting spatial and tissue-level information of the trauma), real-time video stream data (used to characterize the temporal changes in the bleeding status of the trauma site and the relative changes in the pressure position on the body surface, providing a basis for dynamic monitoring of hemostasis operations), and vital sign time series data (used to characterize the pattern of changes in the overall circulatory status of the injured person over time, providing a physiological basis for potential blood loss and hidden risk analysis). Based on the raw trauma data, the location of the injury is extracted and compared with the preset anatomical location range of the junction of the limbs to determine whether the injury is located at the junction of the limbs. When the injury is determined to be... When the trauma is located at the junction of the limbs, the original trauma data is screened and reconstructed based on the trauma risk perception target at the junction. The resulting multidimensional trauma information set can simultaneously characterize the spatial distribution of the trauma, the depth of tissue damage, the intensity and spread trend of bleeding, and the changing patterns of the patient's vital signs. If the trauma is not located at the junction of the limbs, this processing step is skipped to avoid interference from invalid information and improve processing efficiency. By simultaneously acquiring multispectral images, real-time video streams, and time-series data of vital signs, the trauma can be comprehensively characterized in spatial, temporal, and physiological dimensions. This addresses the shortcomings of traditional single data sources, which are unable to reflect the complexity of trauma. The multidimensional information set is generated only when the trauma is located at the junction of the limbs, accurately locating high-risk areas, improving data processing efficiency, and providing a targeted basis for the risk assessment of occult blood loss.

[0028] In a preferred embodiment, based on a multidimensional trauma information set and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, information on the potential blood loss diffusion pathway and corresponding diffusion intensity within the body is obtained, including: S201. Extract multispectral image data, real-time video stream data, and vital sign time series data based on a multidimensional trauma information set; S202. Based on multispectral image data, obtain the diffusion direction and diffusion range of potential blood loss within the tissue, and combine it with pre-established vascular convergence data and tissue connectivity data at the limb junctions to generate spatial diffusion constraint information. S203. Based on real-time video stream data, obtain the rate of change of the bleeding status of the body surface over time, and combine it with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs to generate time evolution constraint information. S204. Based on the time series data of vital signs, obtain the degree of impact of potential blood loss on the overall circulatory status of the injured, and combine it with the pre-established vascular convergence data and tissue connectivity data of the limb junctions to generate physiological response constraint information. S205. Based on spatial diffusion constraint information, temporal evolution constraint information and physiological response constraint information, and combined with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs, the diffusion process of potential blood loss in the body is constrained and combined to obtain the diffusion sequence and diffusion coverage of potential blood loss in the body, and to generate potential blood loss diffusion path information. S206. Based on the spatial diffusion constraint information, temporal evolution constraint information, and physiological response constraint information corresponding to the potential blood loss diffusion path information in different diffusion segments, obtain the diffusion intensity information corresponding to the potential blood loss diffusion path information.

[0029] As described in steps S201 to S206 above, multispectral image data, real-time video stream data, and vital sign time-series data are extracted from the multidimensional trauma information set. Based on the multispectral image data, the diffusion direction and range of potential blood loss within the tissue are analyzed. Combined with vascular convergence and tissue connectivity data at the limb junctions, spatial diffusion constraint information is generated to limit the spatial diffusion characteristics of potential blood loss within the body. Based on the real-time video stream data, the rate of change of the bleeding state on the body surface over time is obtained. Combined with vascular convergence and tissue connectivity data, temporal evolution constraint information is generated to limit the temporal evolution characteristics of potential blood loss. Based on the vital sign time-series data, the impact of potential blood loss on the overall circulatory status of the injured person is assessed. Combined with vascular convergence and tissue connectivity data, physiological response constraint information is generated to limit the physiological response characteristics of potential blood loss. The spatial diffusion constraint information, temporal evolution constraint information, and physiological response constraint information are then combined. By combining constrained information with vascular convergence and tissue connectivity data, the diffusion sequence and coverage of potential blood loss within the body are derived, and potential blood loss diffusion path information is generated. The potential blood loss diffusion path is divided into several continuous spatial segments. By combining spatial diffusion constraint information, temporal evolution constraint information, and physiological response constraint information of different segments, diffusion intensity information of each segment is obtained. Through multi-dimensional information fusion (spatial, temporal, and vital signs), accurate modeling of potential blood loss diffusion paths is achieved, solving the problem that traditional methods relying solely on observation of surface bleeding cannot accurately predict blood loss within the body. The combined analysis of spatial diffusion constraints, temporal evolution constraints, and physiological response constraints allows potential blood loss prediction to consider not only tissue structure but also temporal changes and physiological effects, improving the scientific rigor and reliability of the prediction. The spatial distribution and intensity of potential blood loss within the body are quantified, enabling real-time assessment and adjustment of the treatment process.

[0030] It is worth mentioning that the data on vascular convergence and tissue connectivity at the junctions of the limbs were generated through the collation and statistical analysis of multi-source medical data. Data sources included a standard human anatomy database, medical imaging data (including CT angiography images, MRI soft tissue images, and 3D reconstructed images), and historical trauma treatment case records. Preferably, the medical imaging data came from multi-center clinical data, with a sample size of no less than several hundred cases, covering different genders, age ranges, and body types to improve the representativeness and applicability of the statistical data. During the data construction process, the medical imaging data were first subjected to 3D reconstruction and position alignment under a unified coordinate system, and the vascular structure, number of vascular branches, vascular diameter range, and muscle and tendon data were analyzed. The membrane connections were labeled and statistically analyzed. Next, an anatomical connectivity table was established based on the spatial distribution of blood vessels and tissue connections. The degree of vascular convergence was calculated based on the number of blood vessels per unit space, branch density, and connection complexity. Subsequently, reference data on potential blood loss propagation paths at the limb junctions was compiled based on the statistical results. This data describes the direction of blood loss propagation and the preferred propagation areas in different anatomical regions. To reduce the impact of individual anatomical differences on the application effect, the standard anatomical reference data was proportionally adjusted according to the patient's body shape parameters during actual use. Furthermore, the local anatomical relationships were corrected by combining real-time trauma site location information, thereby improving the specificity and accuracy of potential blood loss propagation path identification.

[0031] In a preferred embodiment, based on the original trauma data, the degree of deviation between overt bleeding characteristics and diffusion intensity information is obtained to generate risk indication information for characterizing the risk of occult blood loss at the limb junction, including: S301. Extract vital sign time series data based on raw trauma data, and obtain the trajectory of vital sign changes in the overall circulatory status of the injured person over time based on the vital sign time series data. S302. Based on the original trauma data, extract the overt bleeding characteristics of the trauma site, and combine the potential blood loss diffusion path information and the corresponding diffusion intensity information to obtain the expected range of change of vital signs trajectory under the action of the potential blood loss diffusion path. S303. Determine whether the actual changes in the trajectory of vital signs are within the expected range; If the actual changes in the trajectory of vital signs are not within the expected range, the deviation of the actual changes from the expected range is obtained and marked as the boundary occult blood loss deviation interval. If the actual changes in the trajectory of vital signs are within the expected range, it is determined that there is no occult blood loss deviating from the range at the junction of the limbs; S304. Obtain the duration and magnitude of the deviation from the boundary of occult blood loss, and generate risk indication information.

[0032] As described in steps S301 to S304 above, vital sign time-series data (such as heart rate, blood pressure, blood oxygen saturation, etc.) are extracted from the original trauma data. Based on the time series, a trajectory of vital sign changes in the overall circulatory status of the injured person over time is established. This trajectory reflects the current blood circulation status of the injured person and the impact of potential blood loss on the circulatory system. Overt bleeding characteristics (such as bleeding volume, oozing pattern, and color changes) at the site of trauma are extracted. Combined with information on the potential blood loss diffusion path and intensity, an expected range of vital sign changes is established. This expected range describes the normal fluctuation range of the injured person's vital signs under potential blood loss. As a benchmark, the actual vital sign change trajectory is compared with the expected range to determine if there is a deviation. If the actual change exceeds the expected range, the excess portion is marked as the boundary of hidden blood loss deviation range. If the actual changes are within the expected range, it is determined that there is no risk of occult blood loss at the junction of the limbs. The duration and magnitude of the deviation from the occult blood loss range are quantified to generate risk indication information, which is used to characterize the severity of occult blood loss at the junction of the limbs and the priority of emergency treatment. Occult blood loss is quantified through the deviation of vital signs, so that the blood loss situation in the body, which is originally difficult to observe directly, can be judged through data analysis, which improves the accuracy of treatment. Through continuous monitoring of vital sign time series data, potential occult blood loss ranges can be dynamically identified, abnormal blood flow changes can be detected in time, and on-site personnel can be guided to make dynamic interventions. The vascular structure at the junction of the limbs is complex, and traditional treatments are difficult to accurately judge the degree of blood loss in the body. By analyzing the deviation between overt bleeding characteristics and potential blood loss diffusion information, the risk of occult blood loss can be identified early, and the effectiveness of treatment can be improved.

[0033] In a preferred embodiment, based on potential blood loss diffusion path information and risk indication information, combined with a pre-established surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point are obtained, and according to the diffusion intensity information, the corresponding recommended pressure range is obtained, and compression hemostasis is performed, including: S401. Obtain the body surface compression blood loss inhibition table, wherein the body surface compression blood loss inhibition table includes information on multiple potential blood loss diffusion paths and a set of candidate body surface compression locations corresponding to each potential blood loss diffusion path. S402. Obtain the corresponding set of candidate body surface compression locations from the body surface compression blood loss inhibition table based on the potential blood loss diffusion path information; S403. Obtain the degree of inhibition matching between each body surface compression location and the potential blood loss diffusion path in the candidate body surface compression location set; S404. Based on the degree of inhibition matching and combined with risk indication information, each body surface compression location in the candidate body surface compression location set is screened to obtain the three-dimensional coordinates of the compression point for applying compression hemostasis on the body surface. S405. Map the potential blood loss diffusion path information to the patient's body surface coordinate system, and construct a body surface action area covering a preset spatial radius with the three-dimensional coordinates of the compression point as the center. S406. Based on the spatial distance information between the potential blood loss diffusion path information and the three-dimensional coordinates of the compression point, map the spatial distance information to a preset set of compression pressure level intervals to obtain the pressure level interval corresponding to the three-dimensional coordinates of the compression point. S407. Based on the pressure level range, obtain the recommended pressure range corresponding to the three-dimensional coordinates of the compression point, and perform compression hemostasis based on the three-dimensional coordinates of the compression point and the recommended pressure range.

[0034] As described in steps S401 to S407 above, a surface compression hemorrhage inhibition table is pre-established, relating the location of surface compression to the potential blood loss diffusion path. This table is compiled from clinical hemostasis records, combat trauma first aid data, and human tissue biomechanics experimental data. Specifically, it includes: statistically analyzing the correspondence between different compression locations, compression intensities, and compression durations with hemostatic effects; classifying and recording bleeding changes under different compression conditions; and forming a data table showing the correspondence between compression location, pressure range, and hemostatic effect level. The pressure range can be obtained through statistical analysis of pressure sensor test data and simulated bleeding experiment results. Multiple batches of experimental data were used to determine the recommended pressure range. This involved recording each potential blood loss diffusion path and its corresponding set of candidate compression locations. Each potential blood loss diffusion path corresponded to multiple candidate compression points, which were generated based on the anatomical structures of the limb junctions (such as vascular convergence, tissue connections, and bone location) and historical trauma hemostasis data. Each candidate point recorded its spatial coordinates, compression direction, and estimated compression effect. For example, assuming a potential blood loss path at a limb junction extends from the upper arm to the shoulder, candidate compression points might include the proximal brachial artery and lateral shoulder compression points. The inhibitory effect of each candidate point was correlated with the potential blood loss diffusion path. The degree of matching of blood loss paths is pre-recorded in a table. Based on the current potential blood loss path of the injured person, a set of candidate compression locations is extracted from the compression blood loss inhibition table. For each candidate compression location, the degree of inhibition matching with the potential blood loss path is obtained, and the effectiveness of compression at this location in inhibiting potential blood loss is measured. The degree of inhibition matching is combined with the risk indication information of occult blood loss at the junction to screen the candidate compression locations. The screening results determine the three-dimensional coordinates of the compression point, ensuring that the compression point can effectively inhibit potential blood loss while taking into account the real-time risk status of the injured person. The potential blood loss path is mapped to the coordinate system of the injured person's body surface, and a space covering the pre-set space is constructed with the selected compression point as the center. The system maps the pressure level range of the compression point to a preset pressure level range based on the spatial distance between the compression point and the potential blood loss path. This yields the pressure level range corresponding to the compression point. Based on the pressure level range, a recommended pressure range for the compression point is generated. The compression point and the recommended pressure range are then superimposed on the actual injury area of ​​the patient using augmented reality (AR) devices or other projection devices in a spatially registered manner. This guides emergency responders to perform compression hemostasis, achieving visualized and operable precise compression. By using potential blood loss path and inhibition matching analysis, the system achieves three-dimensional spatial positioning of the compression point, improving hemostasis accuracy and avoiding deviations from traditional experience-based operations.

[0035] In a preferred embodiment, during the compression hemostasis procedure, pressure sensor data and current video stream data are acquired, and combined with information on potential blood loss diffusion paths, the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted, including: S501. During the execution of the compression hemostasis operation, acquire pressure sensor data and current video stream data; S502. Based on pressure sensor data, obtain the actual pressure distribution characteristics of the current body surface compression state at the three-dimensional coordinates of the compression point, and obtain the hemostasis response characteristics based on the current video stream data. The hemostasis response characteristics include the body surface color change state, bleeding pattern and local tissue collapse state. S503. Obtain the blood loss inhibition coverage degree of the current body surface compression state based on the actual pressure distribution characteristics and hemostasis response characteristics; S504. Determine whether the blood loss inhibition coverage is lower than the preset blood loss inhibition threshold; If the coverage of blood loss inhibition is lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is deemed insufficient. If the coverage of blood loss inhibition is not lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is considered sufficient. S505. Obtain the blood loss inhibition lower value based on the blood loss inhibition coverage being lower than the preset blood loss inhibition threshold; S506. Obtain the spatial distance information between the potential blood loss diffusion path information and the three-dimensional coordinates of the compression point; S507. Based on spatial distance information and blood loss inhibition values, adjust the three-dimensional coordinates of the compression point and the recommended pressure range.

[0036] As described in steps S501 to S507 above, during the compression hemostasis operation, real-time pressure data is acquired by pressure sensors placed in the compression area to characterize the actual implementation status of the compression location and intensity on the body surface. Simultaneously, real-time images of the compression area are acquired through the current video stream to monitor hemostasis response characteristics, such as changes in body surface color, bleeding patterns, and local tissue collapse. Based on the pressure sensor data, the actual pressure distribution characteristics at the three-dimensional coordinates of the compression point are obtained, and combined with the hemostasis response characteristics in the video stream, the inhibitory effect of the current compression state on potential blood loss pathways is evaluated. Specifically, this can be achieved by mapping the actual pressure distribution to the coverage area of ​​potential blood loss pathways and combining this with changes in the hemostasis response characteristics (…). The degree of hemostasis suppression coverage is quantified by changes in color (e.g., lighter color, reduced bleeding, improved tissue stability). A preset hemostasis suppression threshold is used to determine whether the compression hemostasis operation has achieved effective hemostasis. This threshold can be determined based on clinical emergency experience data, historical hemostasis records, and statistical analysis of human tissue compression experiments. The degree of hemostasis suppression can be comprehensively judged by the pressure coverage ratio, changes in bleeding, and changes in local tissue condition. For example, when the effective pressure coverage area reaches 60%–80% or more, and the bleeding rate decreases by 50%–70% or more, and this continues for more than a preset duration (e.g., 15–30 seconds), then the degree of hemostasis suppression can be considered to have reached the effective hemostasis threshold. If it is below this range, it is considered not to have achieved effective hemostasis. A preset hemorrhage inhibition threshold is used. To improve parameter adaptability, the threshold range can be appropriately adjusted in practical applications based on the patient's body size, wound location, and on-site conditions. However, the adjustment range is preferably limited to ±10% to 20% of the preset baseline range to ensure consistency of judgment criteria and repeatability. The hemorrhage inhibition coverage is compared with the preset hemorrhage inhibition threshold. If it is lower than the threshold, the current compression hemostasis effect is determined to be insufficient; if it is not lower than the threshold, the compression hemostasis effect is determined to be sufficient. When the compression hemostasis effect is insufficient, the lower value of hemorrhage inhibition (i.e., the threshold minus the current coverage) is obtained. Combined with the potential hemorrhage diffusion path information and the spatial distance information between the three-dimensional coordinates of the compression point, the spatial distribution of the insufficient area is obtained. The correspondence of potential blood loss paths is dynamically adjusted based on spatial distribution information and insufficient blood loss inhibition, combined with a preset adjustment strategy table, to adjust the three-dimensional coordinates of the compression point and the recommended pressure range. For example, if a certain area is not adequately covered, the compression pressure can be increased or the position of the compression point can be fine-tuned. For insufficient areas along potential blood loss paths, auxiliary compression points can be added to expand the coverage. Through this closed-loop adjustment mechanism, compression hemostasis can respond in real time to the dynamic changes in trauma and blood loss, achieving precise control. It can dynamically optimize the compression plan according to the actual situation at the trauma site and the patient's physiological response, ensuring stable and reliable hemostasis, quantifying the degree of blood loss inhibition coverage and dynamic pressure adjustment, and reducing the risk of tissue damage caused by human judgment errors or improper operation.

[0037] In a preferred embodiment, after completing the compression hemostasis procedure, raw trauma data and potential blood loss propagation pathway information are continuously collected, and the current compression hemostasis effect is obtained. Based on the current compression hemostasis effect, auxiliary procedures for the next stage of emergency treatment are generated, including: S601. After completing the compression hemostasis operation at the target compression point, continuously acquire the patient's original trauma data and potential blood loss diffusion path information, extract vital sign time series data from the original trauma data and mark it as updated vital sign time series data. At the same time, continuously acquire the patient's potential blood loss diffusion path information and mark it as updated potential blood loss diffusion path information. S602. Obtain the degree of hemostasis inhibition of the current compression hemostasis operation based on the updated vital signs time series data and the updated potential blood loss diffusion path information; S603. Obtain the effect of the current compression hemostasis treatment based on the degree of hemostasis inhibition; S604. Based on the current effect of compression hemostasis, generate auxiliary operations for the next stage of emergency treatment. The auxiliary operations include compression adjustment method, supplementary hemostasis measures and emergency treatment operations.

[0038] As described in steps S601 to S604 above, after completing the compression hemostasis operation at the target compression point, the patient's raw trauma data (including multispectral image data, real-time video stream data, and vital sign time-series data) and potential blood loss diffusion path information are continuously collected. Vital sign time-series data is used to characterize the changes in the patient's overall circulatory status over time and is marked as updated. Potential blood loss diffusion path information is used to reflect the dynamic diffusion trend of blood loss within the body and is marked as updated. Based on the updated vital sign time-series data and potential... In the information on the blood loss diffusion path, assess the degree of hemostatic inhibition of the current compression hemostasis operation. The degree of hemostatic inhibition can be obtained by comparing the actual changes in vital signs with the expected range of changes in circulatory status, such as whether the trends in heart rate, blood pressure, and blood oxygen saturation match the expectations. Simultaneously, combine the flow simulation data of potential blood loss diffusion paths to quantify the proportion or coverage of blood loss inhibition. For example, if a potential blood loss path is expected to cause a 10 mmHg drop in blood pressure, but the actual blood pressure drop is only 3 mmHg, then the degree of hemostatic inhibition is 70%. Based on the degree of hemostatic inhibition, combined with the preset hemostatic... The system utilizes an effectiveness database to obtain the current results of compression hemostasis. Hemostasis effectiveness is graded, for example, adequate hemostasis (inhibition level ≥90%), partial hemostasis (inhibition level 50-90%), and insufficient hemostasis (inhibition level <50%). Based on the current effectiveness of compression hemostasis, the system generates a supportive procedure plan for the next stage. Supportive procedures include fine-tuning the compression point location, adjusting the recommended pressure range, adding auxiliary compression points or applying other supplementary hemostasis measures (such as tourniquets or tourniquets), and other emergency procedures (such as intravenous infusion, medication administration, or transport plans). The generated plan can be accessed through... Matched with a pre-set emergency response strategy and dynamically optimized using real-time monitoring data, it ensures that emergency procedures are scientific and feasible. After initial compression hemostasis, it can continuously collect trauma and vital sign data to achieve real-time monitoring of the injured person's condition, ensuring the sustainability of hemostasis. It can adapt to complex trauma environments, enabling continuous and closed-loop hemostasis management, allowing emergency personnel to adjust compression procedures in a timely manner, minimizing the risk of occult blood loss. By dynamically adjusting hemostasis procedures and auxiliary measures, it reduces blood loss and circulatory load fluctuations, improving the overall physiological stability of the injured person.

[0039] In a preferred embodiment, based on a multidimensional trauma information set, risk indication information, the effectiveness of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints, including: S701. Based on the multidimensional trauma information set, risk indication information, the effect of compression hemostasis and corresponding auxiliary operations, generate comprehensive status information, which includes patient status update information, on-site treatment information, medical supply demand information and evacuation demand information. S702. Under preset resource constraints, generate comprehensive disposal requirements for each disposal operation based on comprehensive status information; S703. Based on the comprehensive handling requirements corresponding to each handling operation, sort them according to preset priority and on-site conditions to generate a collaborative handling plan.

[0040] As described in steps S701 to S703 above, based on the multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, comprehensive status information is generated to comprehensively describe the patient's status and the demand for on-site medical resources. This includes patient status update information (by mapping the vital sign time-series data and risk indication information in the multidimensional trauma information set, the patient's physiological stability, trends, and potential deterioration risk are quantified; for example, based on real-time data of blood pressure, heart rate, and potential blood loss diffusion paths, the patient's circulatory system stability index and the risk level of occult blood loss are obtained), and on-site treatment information (by mapping the effect of compression hemostasis and auxiliary operations, the content and sequence of on-site medical operations implemented and to be implemented, and the relationship between each operation and the injury). The system includes the following information: the correspondence between the patient's status and the patient's condition (e.g., marking which pressure points have been operated on, which require pressure adjustment, and which auxiliary hemostasis operations are pending); medical supply demand information (by mapping on-site treatment information and multi-dimensional trauma information to obtain the demand and urgency of hemostasis, fixation, and support supplies for each operation; for example, if the patient currently needs 2 tourniquets and 3 sets of sterile dressings, and marking the supplies as scarce or in place); and evacuation demand information (by mapping patient status update information and on-site treatment information to assess the urgency level, evacuation conditions, and estimated evacuation risks; for example, if the patient's blood pressure is declining significantly and the potential risk of blood loss is high, the urgency level of evacuation is marked as "high," and immediate preparation for transfer is required). Based on comprehensive status information and under preset resource constraints, a comprehensive treatment requirement is generated for each treatment operation. This requirement includes the necessary medical resources, personnel capabilities, priority, and expected outcome. For example, for a compression operation targeting a potential blood loss pathway, the requirement is "1 compression strap, 1 first responder, high priority." Based on the comprehensive treatment requirements for each operation, combined with preset priorities, on-site resource availability, and the order of operations, a collaborative treatment plan is generated. This plan ensures that the operation matches the patient's trauma status, resource allocation, and time sequence, enabling parallel multitasking and resource optimization. For instance, if there are two injured patients on-site, one requiring high-priority compression hemostasis, and the other requiring assisted hemostasis, but resources are limited... The system automatically prioritizes high-risk patients, while simultaneously allocating resources and personnel to achieve coordinated treatment. By comprehensively analyzing patient trauma information, risk indicators, treatment effects, and auxiliary operations, it generates comprehensive status information for each patient, enabling a comprehensive assessment of their condition and treatment needs. Under preset resource constraints, it generates comprehensive requirements for each treatment operation and prioritizes them to achieve optimal allocation of limited personnel, resources, and time. The coordinated treatment plan ensures that different treatment operations do not conflict and are ordered logically, reducing blind operations and repetitive work on-site, and improving on-site treatment efficiency and success rate. The comprehensive status information can be updated with real-time monitoring data, allowing the coordinated treatment plan to be dynamically adjusted and optimized as the patient's condition changes.

[0041] Please see the appendix Figure 2 As shown, the present invention also provides an artificial intelligence-assisted emergency treatment system for combat trauma at the limb junction, used in the aforementioned artificial intelligence-assisted emergency treatment method for combat trauma at the limb junction, comprising: The multidimensional trauma module is used to acquire the raw trauma data of the injured person, obtain the location of the trauma based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the trauma is located at the junction of the limbs, the raw trauma data is filtered to generate a multidimensional trauma information set. The potential blood loss module, based on a multidimensional trauma information set and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, obtains information on the potential blood loss diffusion path and corresponding diffusion intensity within the body. The overt bleeding module is used to obtain the degree of deviation between overt bleeding characteristics and diffusion intensity information based on the original trauma data, and generate risk indication information to characterize the risk of occult blood loss at the junction of the limbs; The compression hemostasis module, based on potential blood loss diffusion path information and risk indication information, combined with a pre-established body surface compression blood loss inhibition table, obtains the three-dimensional coordinates of the compression point, and obtains the corresponding recommended pressure range according to the diffusion intensity information, and performs compression hemostasis operation. The hemostasis adjustment module is used to acquire pressure sensor data and current video stream data during the compression hemostasis operation, and adjust the three-dimensional coordinates of the compression point and the recommended pressure range in combination with potential blood loss diffusion path information. The auxiliary operation module is used to continuously collect raw trauma data and potential blood loss diffusion path information after the compression hemostasis operation is completed, and to obtain the effect of the current compression hemostasis treatment. Based on the effect of the current compression hemostasis treatment, it generates auxiliary operations for the next stage of emergency treatment. The collaborative treatment module generates a collaborative treatment plan based on a multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, under preset resource constraints.

[0042] The aforementioned multidimensional trauma module acquires raw trauma data from the injured person, including multispectral image data, real-time video stream data, and vital sign time-series data. Based on this raw data, it extracts the location of the trauma and compares it with a pre-defined anatomical range at the limb junction to determine if the trauma is located there. If so, the raw trauma data is filtered and reconstructed to generate a multidimensional trauma information set. The potential blood loss module, based on this multidimensional trauma information set and combined with pre-established vascular convergence and tissue connectivity data at the limb junction, infers the potential blood loss diffusion path and corresponding diffusion intensity within the body. Constraints such as spatial diffusion, temporal evolution, and physiological responses work together to generate potential blood loss diffusion path information. The overt bleeding module extracts overt bleeding features from raw trauma data and compares them with potential blood loss diffusion path information to assess the degree of deviation. Based on this deviation, it generates risk indication information to characterize the risk of occult blood loss at the limb junctions. The compression hemostasis module combines potential blood loss diffusion path information, risk indication information, and a pre-established surface compression blood loss inhibition table to obtain the three-dimensional coordinates of candidate compression points. It then determines the recommended pressure range based on the diffusion intensity and performs compression hemostasis via an augmented reality display device. The hemostasis adjustment module collects pressure sensor data and video stream data in real time during the compression hemostasis process, monitors the actual compression status and hemostasis effect, and adjusts the adjustment based on the potential blood loss diffusion path information and blood loss inhibition table. The system dynamically adjusts the 3D coordinates of the compression point and the recommended pressure range to control the coverage. An auxiliary operation module continuously collects raw trauma data and potential blood loss diffusion paths after compression hemostasis is achieved, obtaining the current compression hemostasis effect. Based on the effect, it generates auxiliary operations for the next stage of emergency treatment, including compression adjustment, supplementary hemostasis, and other emergency treatment operations. A collaborative treatment module integrates multi-dimensional trauma information, risk indication information, compression hemostasis effect, and auxiliary operations to generate comprehensive status information, including patient status updates, on-site treatment information, medical supply needs, and evacuation needs. Under preset resource constraints, it generates collaborative treatment plans based on operation priorities and on-site conditions, achieving dynamic optimization of treatment operations. In coordination with resources, a multidimensional trauma information set and potential blood loss analysis enable early identification of both occult and overt blood loss, providing a basis for precise hemostasis decisions. The hemostasis adjustment module and auxiliary operation module, combined with real-time sensor data, achieve dynamic closed-loop control of compression hemostasis operations, improving the success rate of hemostasis. Under resource constraints, the collaborative treatment module rationally arranges compression operations, auxiliary operations, and material usage to avoid resource waste and improve treatment efficiency. By integrating risk indication information and collaborative treatment plans, the module scientifically assesses the urgency level and risk of patient evacuation, prioritizing the treatment of high-risk patients and improving the probability of survival. Augmented reality overlay displays the location of compression points and recommended pressure ranges, providing intuitive and operable guidance for emergency personnel and reducing operational errors.

[0043] And, an artificial intelligence-assisted terminal for emergency treatment of combat trauma at the limb junction, comprising: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors enable artificial intelligence-assisted emergency treatment methods for traumatic injuries at the limb junction.

[0044] The above description is merely a preferred embodiment of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention. Structures, devices, and operating methods not specifically described or explained in this invention are implemented according to conventional methods in the art unless otherwise specified or limited.

Claims

1. A method for emergency treatment of combat injuries at the junction of limbs assisted by artificial intelligence, characterized in that, include: Obtain the raw trauma data of the injured person, determine the location of the injury based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the injury is located at the junction of the limbs, filter the raw trauma data and generate a multidimensional trauma information set. Based on a multidimensional trauma information set, and combined with pre-established vascular convergence data and tissue connectivity data at the junctions of the limbs, information on the potential blood loss diffusion path and corresponding diffusion intensity in the body is obtained. Based on the original trauma data, the degree of deviation between overt bleeding characteristics and diffusion intensity information is obtained, and risk indication information is generated to characterize the risk of occult blood loss at the junction of the limbs. Based on potential blood loss diffusion path information and risk indication information, combined with a pre-established surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point are obtained, and the corresponding recommended pressure range is obtained according to the diffusion intensity information, and compression hemostasis operation is performed. During the compression hemostasis procedure, pressure sensor data and current video stream data are acquired, and the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted in conjunction with information on potential blood loss diffusion paths. After completing the compression hemostasis procedure, raw trauma data and potential blood loss diffusion path information are continuously collected, and the effect of the current compression hemostasis treatment is obtained. Based on the effect of the current compression hemostasis treatment, auxiliary operations for the next stage of emergency treatment are generated. Based on a multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints.

2. The method for emergency treatment of combat wounds at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, Obtain the raw trauma data of the injured person, determine the location of the injury based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the injury is located at the junction of the limbs, filter the raw trauma data to generate a multidimensional trauma information set, including: Acquire raw trauma data from the injured, including multispectral image data, real-time video stream data, and time-series data of vital signs; The location of the injury is extracted based on the original trauma data, and the location of the injury is compared with the preset anatomical location range of the junction of the human limbs to determine whether the location of the injury is located at the junction of the limbs. If the location of the injury is determined to be at the junction of the limbs, the original trauma data is filtered and reconstructed according to the screening rules corresponding to the trauma risk perception target at the junction of the limbs, and a multidimensional trauma information set for the junction of the limbs is generated. If it is determined that the injury did not occur at the junction of the limbs, treatment for the junction of the limbs will not be performed.

3. The method for emergency treatment of combat trauma at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, Based on a multidimensional trauma information set, and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, information on the potential blood loss diffusion pathways and corresponding diffusion intensity within the body is obtained, including: Multispectral image data, real-time video stream data, and vital sign time-series data were extracted based on a multidimensional trauma information set; Based on multispectral image data, the diffusion direction and range of potential blood loss within the tissue are obtained, and spatial diffusion constraint information is generated by combining pre-established vascular convergence data and tissue connectivity data at the limb junctions. The rate of change of the bleeding status on the body surface over time is obtained based on real-time video stream data, and combined with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs, time evolution constraint information is generated. Based on the time series data of vital signs, the impact of potential blood loss on the overall circulatory status of the injured person is obtained, and combined with the pre-established vascular convergence data and tissue connectivity data of the limb junction, physiological response constraint information is generated. Based on spatial diffusion constraint information, temporal evolution constraint information, and physiological response constraint information, and combined with pre-established vascular convergence data and tissue connectivity data at the limb junctions, the diffusion process of potential blood loss in the body is constrained and combined to obtain the diffusion sequence and coverage of potential blood loss in the body, and to generate potential blood loss diffusion path information. Based on the spatial diffusion constraints, temporal evolution constraints, and physiological response constraints corresponding to the potential blood loss diffusion path information in different diffusion segments, diffusion intensity information corresponding to the potential blood loss diffusion path information is obtained.

4. The method for emergency treatment of combat wounds at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, Based on the raw trauma data, the degree of deviation between overt bleeding characteristics and diffusion intensity information is obtained to generate risk indication information for characterizing the risk of occult blood loss at the limb junctions, including: Based on the raw trauma data, time-series data of vital signs are extracted, and the trajectory of changes in vital signs of the patient's overall circulatory status over time is obtained based on the time-series data of vital signs. Based on the raw trauma data, the overt bleeding characteristics of the trauma site are extracted, and combined with the potential blood loss diffusion path information and the corresponding diffusion intensity information, the expected range of change of vital signs trajectory under the action of the potential blood loss diffusion path is obtained. Determine whether the actual changes in the trajectory of vital signs are within the expected range; If the actual changes in the trajectory of vital signs are not within the expected range, the deviation of the actual changes from the expected range is obtained and marked as the boundary occult blood loss deviation interval. If the actual changes in the trajectory of vital signs are within the expected range, it is determined that there is no occult blood loss deviating from the range at the junction of the limbs; Obtain the duration and magnitude of deviation from the occult blood loss range at the junction, and generate risk indication information.

5. The method for emergency treatment of combat trauma at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, Based on potential blood loss diffusion pathway information and risk indication information, combined with a pre-established surface compression blood loss inhibition table, the three-dimensional coordinates of the compression point are obtained. Based on the diffusion intensity information, the corresponding recommended pressure range is determined, and compression hemostasis is performed, including: Obtain a surface compression blood loss inhibition table, which includes information on multiple potential blood loss diffusion paths and a set of candidate surface compression locations corresponding to each potential blood loss diffusion path. Based on the potential blood loss diffusion path information, obtain the corresponding set of candidate body surface compression locations from the body surface compression blood loss inhibition table; Obtain the degree of inhibition matching between each surface compression location in the candidate set of surface compression locations and the potential blood loss diffusion path; Based on the degree of inhibition matching and combined with risk indication information, each pressure point in the candidate body surface pressure location set is screened to obtain the three-dimensional coordinates of the pressure point for applying pressure to stop bleeding on the body surface. The potential blood loss diffusion path information is mapped to the patient's body surface coordinate system, and the body surface action area covering the preset spatial radius is constructed with the three-dimensional coordinates of the compression point as the center. Based on the spatial distance information between the potential blood loss diffusion path information and the three-dimensional coordinates of the compression point, the spatial distance information is mapped to a preset set of compression pressure level intervals to obtain the pressure level interval corresponding to the three-dimensional coordinates of the compression point. Based on the pressure level range, obtain the recommended pressure range corresponding to the three-dimensional coordinates of the compression point, and perform compression hemostasis based on the three-dimensional coordinates of the compression point and the recommended pressure range.

6. The method for emergency treatment of combat wounds at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, During the compression hemostasis procedure, pressure sensor data and current video stream data are acquired, and combined with information on potential blood loss diffusion paths, the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted, including: During the compression hemostasis procedure, pressure sensor data and current video stream data are acquired. The actual pressure distribution characteristics of the current body surface compression state at the three-dimensional coordinates of the compression point are obtained based on pressure sensor data, and the hemostasis response characteristics are obtained based on the current video stream data. The hemostasis response characteristics include the body surface color change state, bleeding pattern and local tissue collapse state. The degree of blood loss inhibition coverage under the current body surface compression state is obtained based on the actual pressure distribution characteristics and hemostasis response characteristics. Determine whether the blood loss inhibition coverage is lower than the preset blood loss inhibition threshold; If the coverage of blood loss inhibition is lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is deemed insufficient. If the coverage of blood loss inhibition is not lower than the preset blood loss inhibition threshold, the current compression hemostasis effect is considered sufficient. The value of blood loss inhibition is obtained based on the blood loss inhibition coverage being lower than a preset blood loss inhibition threshold; Obtain spatial distance information between potential blood loss diffusion paths and the three-dimensional coordinates of the compression point; Based on spatial distance information and blood loss inhibition values, the three-dimensional coordinates of the compression point and the recommended pressure range are adjusted.

7. The method for emergency treatment of combat wounds at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, After completing the compression hemostasis procedure, raw trauma data and potential blood loss propagation pathways are continuously collected, and the effectiveness of the current compression hemostasis is obtained. Based on the effectiveness of the current compression hemostasis, auxiliary procedures for the next stage of emergency treatment are generated, including: After completing the compression hemostasis operation at the target compression point, continuously acquire the patient's original trauma data and potential blood loss diffusion path information, extract vital sign time series data from the original trauma data and mark it as updated vital sign time series data. At the same time, continuously acquire the patient's potential blood loss diffusion path information and mark it as updated potential blood loss diffusion path information. The degree of hemostatic inhibition of the current compression hemostasis operation is obtained based on the updated vital signs time series data and the updated potential blood loss diffusion path information; The effectiveness of the current compression hemostasis treatment is determined based on the degree of hemostasis inhibition. Based on the current effect of compression hemostasis, auxiliary procedures for the next stage of emergency treatment are generated. These auxiliary procedures include compression adjustment methods, supplementary hemostasis measures, and emergency treatment procedures.

8. The method for emergency treatment of combat trauma at the junction of limbs assisted by artificial intelligence according to claim 1, characterized in that, Based on a multidimensional trauma information set, risk indication information, the effectiveness of compression hemostasis, and auxiliary operations, a collaborative treatment plan is generated under preset resource constraints, including: Based on the multidimensional trauma information set, risk indication information, the effect of compression hemostasis and corresponding auxiliary operations, comprehensive status information is generated. The comprehensive status information includes patient status update information, on-site treatment information, medical supply demand information and evacuation demand information. Under preset resource constraints, the comprehensive disposal requirements for each disposal operation are generated based on the comprehensive status information. Based on the comprehensive handling requirements corresponding to each handling operation, a collaborative handling plan is generated by prioritizing the operations according to preset priorities and on-site conditions.

9. An artificial intelligence-assisted emergency treatment system for combat trauma at the limb junction, applied to the artificial intelligence-assisted emergency treatment method for combat trauma at the limb junction as described in any one of claims 1 to 8, characterized in that, include: The multidimensional trauma module is used to acquire the raw trauma data of the injured person, obtain the location of the trauma based on the raw trauma data, and determine whether it is located at the junction of the limbs. If the trauma is located at the junction of the limbs, the raw trauma data is filtered to generate a multidimensional trauma information set. The potential blood loss module, based on a multidimensional trauma information set and combined with pre-established vascular convergence data and tissue connectivity data at the junction of the limbs, obtains information on the potential blood loss diffusion path and corresponding diffusion intensity information in the body. The overt bleeding module is used to obtain the degree of deviation between overt bleeding characteristics and diffusion intensity information based on the original trauma data, and generate risk indication information to characterize the risk of occult blood loss at the junction of the limbs; The compression hemostasis module, based on potential blood loss diffusion path information and risk indication information, combined with a pre-established body surface compression blood loss inhibition table, obtains the three-dimensional coordinates of the compression point, and obtains the corresponding recommended pressure range according to the diffusion intensity information, and performs compression hemostasis operation. The hemostasis adjustment module is used to acquire pressure sensor data and current video stream data during the compression hemostasis operation, and adjust the three-dimensional coordinates of the compression point and the recommended pressure range in combination with the potential blood loss diffusion path information. The auxiliary operation module is used to continuously collect raw trauma data and potential blood loss diffusion path information after the compression hemostasis operation is completed, and to obtain the effect of the current compression hemostasis treatment. Based on the effect of the current compression hemostasis treatment, it generates auxiliary operations for the next stage of emergency treatment. The collaborative treatment module generates a collaborative treatment plan based on a multidimensional trauma information set, risk indication information, the effect of compression hemostasis, and auxiliary operations, under preset resource constraints.

10. An artificial intelligence-assisted emergency treatment terminal for combat trauma at the limb junction, characterized in that, include: One or more processors; A storage device on which one or more programs are stored; When one or more programs are executed by one or more processors, the one or more processors implement the artificial intelligence-assisted emergency treatment method for combat trauma at the limb junction as described in any one of claims 1 to 8.

Images