Pre-hospital triage method for emergency rescue linkage

Abstract

The invention relates to the technical field of pre-hospital triage. The invention relates to a pre-hospital triage method for emergency rescue linkage. The method comprises the following steps: S1, acquiring image data of an ambulance and detection data of a patient, and combining the image data with the detection data to perform detection body feature analysis; s2, performing identity information identification on the patient according to the detected body features, performing historical case extraction according to the identified identity information, and then combining the historical cases with the detected body features to perform definite disease and suspected disease analysis; according to the method, a personalized rescue scheme is generated according to definite diagnosis diseases and detected body characteristics, the department with the highest professional value is selected in combination with emergency department information and attendant doctors on duty, medical resources are reasonably distributed, the rescue efficiency and quality are improved, and the diagnosis and suspicion threshold values are set by calculating the similarity with standard body characteristics. The disease can be scientifically judged.

Description

Technical Field

[0001] The present invention relates to the technical field of pre-hospital triage, and in particular to a pre-hospital triage method for emergency and first aid linkage. Background Art

[0002] In the field of emergency medicine, rapid and accurate triage and efficient rescue plan formulation are crucial to saving patients' lives and improving treatment outcomes. Through preliminary judgment and classification of patients' symptoms, patients can be quickly assigned to appropriate medical resources to achieve timely and effective treatment. The purpose is to improve emergency rescue efficiency, reduce mortality and disability rates, and protect patients' lives and health as much as possible.

[0003] At present, the existing technology mainly works in the hospital emergency room or the first aid site. In the hospital emergency room scenario, a series of examinations and diagnoses are performed after the patient is sent to the hospital. There are often problems such as cumbersome medical procedures and long waiting times, which will not only delay the best time for treatment of the patient, but may also cause the condition to worsen and seriously affect the treatment effect, which runs counter to the purpose of improving the efficiency and quality of treatment. Therefore, a pre-hospital triage method for emergency rescue linkage is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a pre-hospital triage method for emergency rescue linkage to solve the problems raised in the above-mentioned background technology.

[0005] To achieve the above object, a pre-hospital triage method for emergency rescue linkage is provided, comprising the following steps: S1, collecting image data of the ambulance and test data of the patient, and combining the image data with the test data to perform body feature analysis; S2. Identify the patient's identity information based on the detected physical characteristics, extract historical cases based on the identified identity information, and then analyze confirmed and suspected diseases based on the historical cases combined with the detected physical characteristics; S3. Collect historical emergency data, obtain historical patient-related data and emergency disease reports from the historical emergency data to perform historical physical characteristics analysis, and obtain historical physical characteristics corresponding to each disease; S4, performing similarity analysis on the detected physical features in combination with the suspected symptoms, and performing concurrent similarity threshold analysis on the suspected symptoms according to the confirmed symptoms, and then performing diagnosis adjustment on the suspected symptoms by combining the concurrent similarity threshold with the similarity; S5. Obtain the emergency department information of the hospital, and use the AI ​​model to generate a rescue plan based on the confirmed symptoms and detected physical characteristics. Combine the emergency department information with the rescue plan to perform rescue analysis and selection, and then feed back the selected rescue plan to the ambulance and emergency department for plan confirmation.

[0006] As a further improvement of the technical solution, the S1 establishes an emergency rescue linkage module at the data management end of the hospital, establishes an information transmission connection between the ambulance and the emergency rescue linkage module and the emergency department, thereby obtaining the image data of the ambulance and the test data of the patient in real time; Transmit image data from a camera in an ambulance; Transmit test data based on medical equipment in the ambulance.

[0007] As a further improvement of the technical solution, the step of S1 is as follows: S1.1. Analyze the patient's trauma based on the image data and analyze the patient's physical condition based on the test data; S1.2. Combine the trauma analysis results and the body condition analysis results to generate detection body features.

[0008] As a further improvement of the technical solution, the steps of S2 are as follows: S2.1. Identify the patient's identity information based on the detected physical characteristics and obtain the patient's identity information; S2.2. Input the identity information into the hospital data management terminal to extract historical cases, obtain the hospital's diagnostic case data from the historical cases, and perform intelligent medical diagnosis analysis on the detected physical characteristics to obtain the diagnostic case data corresponding to the detected physical characteristics; S2.3. Obtain the standard physical characteristics of each disease, and then combine the diagnostic case data corresponding to the historical cases and the diagnostic case data corresponding to the detected physical characteristics with the standard physical characteristics to analyze the confirmed diseases and suspected diseases, and obtain the confirmed diseases and suspected diseases corresponding to the patient.

[0009] As a further improvement of the present technical solution, during the process of identity information recognition, when the physical features cannot be detected to accurately recognize the identity information, S2.1 will feed back the identity information with high recognition similarity to the rescue personnel for auxiliary filling, and determine the patient based on the identity information filled in by the rescue personnel.

[0010] As a further improvement of the technical solution, the steps of S3 are as follows: S3.1. The hospital data management terminal collects the relevant data of each emergency rescue and stores it as historical emergency data; S3.2. Obtain historical patient-related data from historical emergency data, and conduct historical physical characteristics analysis on the detected physical characteristics in the historical patient-related data and the emergency disease report, so as to obtain the historical physical characteristics corresponding to each disease based on the emergency diagnosis disease report and the detected physical characteristics in the historical patient-related data.

[0011] As a further improvement of the technical solution, the step of S4 is as follows: S4.1. Perform similarity analysis on the detected physical features in combination with the suspected diseases to obtain the similarity between the detected physical features and each of the suspected diseases; S4.2. Perform concurrent similarity threshold analysis on suspected symptoms based on confirmed symptoms, and set concurrent similarity thresholds based on the analysis results; S4.3. Analyze the similarity of each symptom in the suspected symptoms with the concurrent similarity threshold. When the similarity is greater than the concurrent similarity threshold, the symptom is adjusted from the suspected symptoms to the confirmed symptoms. Conversely, when the similarity is less than the concurrent similarity threshold, no adjustment is made.

[0012] As a further improvement of the technical solution, the confirmed disease in S4 is a set of diseases that the patient suffers from; Suspected diseases are a collection of diseases that the suspected patients suffer from; The minimum number of confirmed and suspected diseases can be 0.

[0013] As a further improvement of the technical solution, the step of S5 is as follows: S5.1. Obtain the emergency department information of the hospital, and extract the introduction of the doctor on duty corresponding to each emergency department according to the emergency department information; S5.2. Generate a rescue plan for the patient using an AI model based on the confirmed symptoms and detected physical characteristics. Then, conduct a professional numerical analysis of the rescue plan in combination with the on-duty doctor corresponding to each emergency department. Then, feed back the rescue plan to the ambulance information terminal and the emergency department with the highest professional value. The emergency department with the highest professional value will be selected as the department for the patient to visit.

[0014] Compared with the prior art, the present invention has the following beneficial effects: In this pre-hospital triage method for emergency and first aid linkage, a personalized first aid plan is generated according to confirmed symptoms and detected physical characteristics, and the department with the highest professional value is selected in combination with the emergency department information and the expertise of the doctor on duty, so that medical resources are reasonably allocated to improve the efficiency and quality of treatment. By calculating the similarity with standard physical characteristics and setting the confirmed and doubtful thresholds, the symptoms can be scientifically determined. At the same time, combined with historical physical characteristic analysis and similarity comparison, the doubtful symptoms are diagnosed and adjusted, and then the first aid plan generated by AI is sent to the emergency department. The doctor in the emergency department can understand the patient's condition through the first aid plan, and can refer to the first aid plan to prepare for the first aid in advance, so that the patient's sitting in the ambulance is equivalent to entering the hospital, and the doctor prepares the first aid work for the patient in advance, realizing emergency and first aid linkage. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is the overall flow chart of the present invention; Figure 2A flowchart of the present invention for analyzing trauma of a patient based on image data; Figure 3 A flowchart of the present invention for obtaining the patient's identity information; Figure 4 A flowchart of the present invention for collecting relevant data of each emergency rescue; Figure 5 A flowchart of setting a concurrent similarity threshold according to analysis results of the present invention; Figure 6 A flowchart of selecting the emergency department with the highest professional value as the department where the patient visits for treatment in the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 - Figure 6 As shown, the purpose of this embodiment is to provide a pre-hospital triage method for emergency rescue linkage, comprising the following steps: S1, collecting image data of the ambulance and test data of the patient, and combining the image data with the test data to perform body feature analysis; S1 establishes an emergency rescue linkage module on the data management side of the hospital, establishes an information transmission link between the ambulance and the emergency rescue linkage module and the emergency department, and thus obtains the image data of the ambulance and the test data of the patient in real time; Transmit image data from a camera in an ambulance; Using the camera in the ambulance, the image data is transmitted to the emergency rescue linkage module and the emergency department in real time through the established information transmission connection; Transmitting test data from medical equipment in the ambulance; Utilize the medical equipment in the ambulance (electrocardiogram monitor, blood pressure monitor, thermometer, blood oxygen saturation meter, etc.) to transmit the patient's test data (physiological indicators) to the emergency rescue linkage module and the emergency department in real time through the established information transmission connection.

[0018] The steps of S1 are as follows: S1.1. Analyze the patient's trauma based on the image data and analyze the patient's physical condition based on the test data; S1.2. Combine the trauma analysis results and the body condition analysis results to generate the body features for detection. The specific steps are as follows: Trauma identification: Use target detection algorithms to identify the trauma area and mark the location, use image segmentation algorithms to calculate the size and shape of the wound, combine medical knowledge to determine the wound depth and trauma type, extract feature information such as wound edge roughness and color changes from the trauma area, quantify and encode to form trauma analysis results; Physical status assessment: physical status is assessed based on data, medical standards and clinical experience. For example, a heart rate of 60 to 100 beats / minute is normal, which is used to judge heart function; normal blood pressure is 90 to 139 mmHg systolic and 60 to 89 mmHg diastolic, which is used to judge the circulatory system; body temperature is used to judge fever or hypothermia; blood oxygen saturation of 95% to 100% is used to assess the respiratory system, quantify and encode physical status assessment results, extract key feature information such as heart rate change trend and blood pressure fluctuation range, and form physical status analysis results; Generate detected body characteristics: Match and integrate the trauma analysis results and the physical condition analysis results, integrate information such as the severity of the trauma and the functional status of various body systems, and present the detected body characteristics in a structured manner.

[0019] S2. Identify the patient's identity information based on the detected physical characteristics, extract historical cases based on the identified identity information, and then analyze confirmed and suspected diseases based on the historical cases combined with the detected physical characteristics; The steps of S2 are as follows: S2.1. Identify the patient's identity information based on the detected physical characteristics and obtain the patient's identity information; Key biometric features (such as facial features, fingerprint features, iris features, etc.) and medical features (such as special disease markers, implanted medical device models, etc.) are extracted from the detected body features, and the extracted features are converted into digital vectors that can be processed by computers. For example, facial features can be used to extract feature vectors through convolutional neural networks (CNN), and fingerprint features can be converted into feature point coordinate vectors. The converted feature vectors are compared with the patient identity information database stored in the hospital data management end.

[0020] S2.1 During the process of identity recognition, when the detection of physical features cannot accurately identify the identity, the identity information with a high degree of similarity will be fed back to the rescue personnel for auxiliary filling, and the patient will be identified based on the identity information filled in by the rescue personnel.

[0021] Use the distance measurement algorithm to calculate the similarity between the feature vector of the patient to be identified and the feature vectors of each patient in the database, and set a similarity threshold. When there is a record whose similarity with the feature vector of the patient to be identified exceeds the threshold, the patient identity information corresponding to the record is determined as the identification result. When the similarity of all comparison results does not exceed the threshold, that is, the detection of physical characteristics cannot accurately identify the identity information, three identity information with higher similarity are screened out from high to low in terms of similarity, and these three identity information are fed back to the rescue personnel. The rescue personnel assist in filling in the information according to the actual situation (such as asking the patient, checking the personal documents, etc.), and finally determine the patient's identity information. S2.2. Input the identity information into the hospital data management terminal to extract historical cases, obtain the hospital's diagnostic case data from the historical cases, and perform intelligent medical diagnosis analysis on the detected physical characteristics to obtain the diagnostic case data corresponding to the detected physical characteristics. The specific steps are as follows: Historical case extraction: The confirmed patient identity information is used as the query keyword and input into the database query system of the hospital data management end. The database searches based on the keyword and extracts all the historical diagnosis case data of the patient from the historical case storage table, including the consultation time, diagnosis results, examination report, treatment plan and other information; Intelligent medical diagnosis and analysis: The detected physical feature data is input into a pre-trained intelligent medical diagnosis model (such as a neural network model based on deep learning). The model calculates the input feature data through internal parameters and algorithms, and outputs the predicted diagnosis results and corresponding probability values. According to the set probability threshold (such as a probability greater than 0.5), the model screens out diseases with higher probability as the diagnostic case data corresponding to the detected physical features.

[0022] S2.3. Obtain the standard physical characteristics of each disease, then combine the diagnostic case data corresponding to the historical cases and the diagnostic case data corresponding to the detected physical characteristics with the standard physical characteristics to analyze the confirmed diseases and suspected diseases, and obtain the confirmed diseases and suspected diseases corresponding to the patient. The specific steps are as follows: Standard physical characteristics acquisition is to retrieve the pre-set standard physical characteristics data for each disease from the medical knowledge base of the hospital data management end. These data cover multi-dimensional information such as symptoms, physiological indicators, and pathological characteristics; Similarity calculation: For the standard physical characteristics of each disease, the similarity between historical case data, test physical characteristics data and standard characteristics is calculated respectively. The similarity formula for numerical characteristics is as follows: ; Among them, D(S,C) is the similarity between the standard body feature vector and the case data vector. The smaller the distance, the higher the similarity. n is the number of numerical body features involved in the calculation. S is the standard body feature vector, si is the value of the i-th dimension in the standard body feature vector, C is the case data vector, c i is the value of the i-th dimension in the case data vector; The formula for non-numerical feature similarity is as follows: ; Among them, sim(A,B) is, A and B are text vectors representing two non-numerical physical characteristics (such as symptom descriptions) after conversion; Analysis of confirmed and suspected diseases: Set a confirmed threshold. If the similarity between the historical case data and the standard physical characteristics of a disease, and the similarity between the detected physical characteristics data and the standard physical characteristics, are both greater than or equal to the confirmed threshold, the disease is determined to be a confirmed disease; A suspected threshold is set, and the suspected threshold is set to be less than the confirmed threshold. If the similarity between the historical case data or the detected physical feature data of a disease and the standard physical feature is greater than or equal to the suspected threshold, but does not reach the confirmed threshold, the disease is determined to be a suspected disease.

[0023] S3. Collect historical emergency data, obtain historical patient-related data and emergency disease reports from the historical emergency data to perform historical physical characteristics analysis, and obtain historical physical characteristics corresponding to each disease; The steps for S3 are as follows: S3.1. The hospital data management terminal collects the relevant data of each emergency rescue and stores it as historical emergency data; The hospital data management terminal receives relevant data generated during each emergency rescue process in real time, including but not limited to the patient's basic information (name, age, gender, etc.), physical characteristic data (physiological indicators, trauma conditions, etc.), and emergency disease reports (preliminary diagnosis results, treatment measures, etc.), and stores them in a special historical emergency database to form historical emergency data.

[0024] S3.2. Obtain historical patient-related data from historical emergency data, and analyze the physical characteristics of the historical patient-related data with the emergency disease report, thereby obtaining the historical physical characteristics corresponding to each disease based on the emergency diagnosis disease report combined with the physical characteristics of the historical patient-related data. The specific steps are as follows: Filter out specific historical patient-related data from the historical emergency database according to needs. These data include the patient's physical characteristics data detected during the emergency and the corresponding emergency symptom report; For each disease, the physical characteristics data of all historical patients suffering from the disease are integrated, and the integrated data are statistically analyzed to extract typical characteristics that can represent the disease, namely historical physical characteristics. For example, the mean, median, standard deviation and other statistical quantities of numerical physical characteristics are calculated, and the frequency of occurrence and distribution of non-numerical physical characteristics are analyzed.

[0025] S4, performing similarity analysis on the detected physical features in combination with the suspected symptoms, and performing concurrent similarity threshold analysis on the suspected symptoms according to the confirmed symptoms, and then performing diagnosis adjustment on the suspected symptoms by combining the concurrent similarity threshold with the similarity; The steps of S4 are as follows: S4.1. Perform similarity analysis on the detected physical features and suspected diseases to obtain the similarity between the detected physical features and each of the suspected diseases. The formula is as follows: ; Among them, X is the physical feature to be detected, Y is the suspected disease, sim(X, Y) is the similarity between the physical feature to be detected and the suspected disease, m is the dimension of the feature vector, x j and j are the values ​​of the j-th dimension of X and Y respectively; The above calculation is for a single symptom. When the suspected symptom includes multiple symptoms, the multiple symptoms will be calculated separately with the detected physical characteristics; S4.2. Perform concurrent similarity threshold analysis on suspected symptoms based on confirmed symptoms, and set concurrent similarity threshold based on the analysis results. The formula is as follows: ; Where T is the concurrent similarity threshold, C is the confirmed disease, Y is the suspected disease, α and β are weight coefficients adjusted according to the actual situation, and 0<α, α<1, o(C∩Y) is the number of diseases that appear simultaneously in the confirmed disease set and the suspected disease set, and o(Y) is the number of diseases in the suspected disease set.

[0026] S4.3. Analyze the similarity of each symptom in the suspected symptom with the concurrent similarity threshold. When the similarity is greater than the concurrent similarity threshold, the corresponding symptom is adjusted from the suspected symptom to the confirmed symptom. Conversely, when the similarity is less than the concurrent similarity threshold, no adjustment is made. The similarity of each symptom in the suspected symptom is compared with the concurrent similarity threshold, and the comparison result is used to determine whether to adjust the symptom from a suspected symptom to a confirmed symptom. The formula is as follows: For a certain symptom among the suspected symptoms, when sim(X, Y)>T, the symptom is adjusted from the suspected symptom to the confirmed symptom; When sim(X, Y) ≤ T, no adjustment is performed.

[0027] The confirmed disease of S4 is the set of diseases that the patient is confirmed to have; Suspected diseases are a collection of diseases that the suspected patients suffer from; The minimum value of confirmed diseases and suspected diseases can be 0; By setting the lowest value to 0, the system can avoid being unable to make a diagnosis of the disease when some patients blindly occupy the rescue channel. When the number of diagnosed and suspected diseases is 0, the rescue personnel can independently allocate (give priority to the emergency department with the least traffic) to arrange rescue.

[0028] S5. Obtain the emergency department information of the hospital, and use the AI ​​model to generate a rescue plan based on the confirmed symptoms and detected physical characteristics. Combine the emergency department information with the rescue plan to perform rescue analysis and selection, and then feed back the selected rescue plan to the ambulance and emergency department for plan confirmation.

[0029] The steps of S5 are as follows: S5.1. Obtain the emergency department information of the hospital, and extract the introduction of the on-duty doctor corresponding to each emergency department according to the emergency department information; Retrieve emergency department information from the hospital information management system, including department name, types of diseases that can be received, equipment, etc. Based on the emergency department information, extract the introduction of the doctor on duty corresponding to each department, including the doctor's expertise, years of practice, number of successful cases, etc. S5.2. Generate a rescue plan for the patient using the AI ​​model based on the confirmed symptoms and the detected physical characteristics. Then, perform a professional numerical analysis on the rescue plan in combination with the on-duty doctor's introduction of each emergency department. Then, feedback the rescue plan to the ambulance information terminal and the emergency department with the highest professional value. The emergency department with the highest professional value is selected as the department where the patient will be treated. The specific steps are as follows: Generate rescue plan: Input the confirmed disease and detected physical characteristics data into the pre-trained AI model. The AI ​​model outputs a personalized rescue plan for the patient based on the input data, combined with medical knowledge and algorithms. The plan includes first aid measures, medication recommendations, follow-up treatment directions, etc. Professional numerical analysis: For each emergency department and the corresponding doctor on duty, the key information in the rescue plan is matched and analyzed with the doctor's expertise and department capabilities. Based on the matching results, the professional value of each department for the rescue plan is calculated. The formula is as follows: ; Among them, P k is the professional value of the rescue plan of the kth emergency department, w 1 、w 2 、w q are weight coefficients, R k1 The matching degree between the doctor's expertise and the disease, Rk2 is the matching degree between department equipment and rescue needs, R kq represents the score of the kth emergency department on the qth matching factor; Compare the professional values ​​of each emergency department, find out the department with the highest value, feed back the rescue plan to the ambulance information terminal and the emergency department with the highest professional value, and determine that department as the department where the patient will be treated.

[0030] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A pre-hospital triage method for emergency rescue linkage, characterized by: The following steps are involved: S1, collecting image data of the ambulance and test data of the patient, and combining the image data with the test data to perform body feature analysis; S2. Identify the patient's identity information based on the detected physical characteristics, extract historical cases based on the identified identity information, and then analyze confirmed and suspected diseases based on the historical cases combined with the detected physical characteristics; S3. Collect historical emergency data, obtain historical patient-related data and emergency disease reports from the historical emergency data to perform historical physical characteristics analysis, and obtain historical physical characteristics corresponding to each disease; S4, performing similarity analysis on the detected physical features in combination with the suspected symptoms, and performing concurrent similarity threshold analysis on the suspected symptoms according to the confirmed symptoms, and then performing diagnosis adjustment on the suspected symptoms by combining the concurrent similarity threshold with the similarity; S5. Obtain the emergency department information of the hospital, and use the AI ​​model to generate a rescue plan based on the confirmed symptoms and detected physical characteristics. Combine the emergency department information with the rescue plan to perform rescue analysis and selection, and then feed back the selected rescue plan to the ambulance and emergency department for plan confirmation.

2. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The S1 establishes an emergency rescue linkage module at the data management end of the hospital, establishes an information transmission connection between the ambulance and the emergency rescue linkage module and the emergency department, so as to obtain the image data of the ambulance and the test data of the patient in real time; Transmit image data from a camera in an ambulance; Transmit test data based on medical equipment in the ambulance.

3. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The steps of S1 are as follows: S1.

1. Analyze the patient's trauma based on the image data and analyze the patient's physical condition based on the test data; S1.

2. Combine the trauma analysis results and the body condition analysis results to generate detection body features.

4. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The steps of S2 are as follows: S2.

1. Identify the patient's identity information based on the detected physical characteristics and obtain the patient's identity information; S2.

2. Input the identity information into the hospital data management terminal to extract historical cases, obtain the hospital's diagnostic case data from the historical cases, and perform intelligent medical diagnosis analysis on the detected physical characteristics to obtain the diagnostic case data corresponding to the detected physical characteristics; S2.

3. Obtain the standard physical characteristics of each disease, and then combine the diagnostic case data corresponding to the historical cases and the diagnostic case data corresponding to the detected physical characteristics with the standard physical characteristics to analyze the confirmed diseases and suspected diseases, and obtain the confirmed diseases and suspected diseases corresponding to the patient.

5. The pre-hospital triage method for emergency rescue linkage according to claim 4 is characterized by: In the process of identity information recognition, when the detection of physical features cannot accurately identify the identity information, the identity information with high recognition similarity is fed back to the rescue personnel for auxiliary filling, and the patient is determined based on the identity information filled in by the rescue personnel.

6. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The steps of S3 are as follows: S3.

1. The hospital data management terminal collects the relevant data of each emergency rescue and stores it as historical emergency data; S3.

2. Obtain historical patient-related data from historical emergency data, and conduct historical physical characteristics analysis on the detected physical characteristics in the historical patient-related data and the emergency disease report, so as to obtain the historical physical characteristics corresponding to each disease based on the emergency diagnosis disease report and the detected physical characteristics in the historical patient-related data.

7. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The steps of S4 are as follows: S4.

1. Perform similarity analysis on the detected physical features in combination with the suspected diseases to obtain the similarity between the detected physical features and each of the suspected diseases; S4.

2. Perform concurrent similarity threshold analysis on suspected symptoms based on confirmed symptoms, and set concurrent similarity thresholds based on the analysis results; S4.

3. Analyze the similarity of each symptom in the suspected symptoms with the concurrent similarity threshold. When the similarity is greater than the concurrent similarity threshold, the symptom is adjusted from the suspected symptoms to the confirmed symptoms. Conversely, when the similarity is less than the concurrent similarity threshold, no adjustment is made.

8. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The confirmed disease in S4 is a set of diseases that the patient is suffering from; Suspected diseases are a collection of diseases that the suspected patients suffer from; The minimum number of confirmed and suspected diseases can be 0.

9. The pre-hospital triage method for emergency rescue linkage according to claim 1, characterized in that: The steps of S5 are as follows: S5.

1. Obtain the emergency department information of the hospital, and extract the introduction of the doctor on duty corresponding to each emergency department according to the emergency department information; S5.

2. Generate a rescue plan for the patient using an AI model based on the confirmed symptoms and detected physical characteristics. Then, conduct a professional numerical analysis of the rescue plan in combination with the on-duty doctor corresponding to each emergency department. Then, feed back the rescue plan to the ambulance information terminal and the emergency department with the highest professional value. The emergency department with the highest professional value will be selected as the department for the patient to visit.

Images