Information docking method used in emergency treatment process

Abstract

The invention relates to the field of medical first aid, and particularly discloses an information docking method used in an emergency treatment first aid process. The information docking method used in the emergency treatment process comprises the following steps: acquiring a patient information set; obtaining a probability value of the patient corresponding to each department by using a department recommendation model; selecting a plurality of preparatory departments, respectively calculating the comprehensive professional ability of each hospital, planning a route, and obtaining the predicted spent time; and calculating to obtain a final adaptation score of each hospital, and selecting a target hospital. According to the method, the key information of the patient is acquired and analyzed, the probability value of the patient corresponding to each department is generated, the preparation department is determined in advance, the comprehensive professional ability of the hospital, the prediction spent time and the emergency degree of the patient are comprehensively considered, the final adaptation score is calculated, and the target hospital is scientifically selected; the process not only reduces the time of on-site evaluation, but also ensures that the patient can be sent to the most appropriate hospital.

Description

Technical Field

[0001] The present invention relates to the field of medical emergency, and in particular to an information docking method used in an emergency first aid process. Background Art

[0002] Emergency first aid is a vital part of the medical system, aiming to provide timely and effective initial diagnosis and treatment for patients with sudden illness or accidental injury. In emergency calls, the information provided by the person in need is often incomplete or inaccurate, making it difficult for the dispatch center to accurately assess the patient's condition. In addition, ambulances usually send patients to the nearest hospital, but the hospital's professional capabilities also need to be considered. If the patient is only sent to the nearest hospital, the patient may still need to be transferred to a hospital with stronger specialist capabilities to receive more appropriate treatment, making the emergency treatment process more cumbersome and affecting the treatment effect.

[0003] In order to improve the efficiency and effect of emergency first aid, a more intelligent and flexible information docking method is needed; therefore, the present invention proposes an information docking method for emergency first aid process. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides an information docking method for emergency first aid, which can make greater use of the patient information obtained from the emergency call; based on the patient's specific condition, a department recommendation model is constructed to select the most appropriate department and hospital to improve the treatment effect; at the same time, a comprehensive assessment of the patient's vital signs and consciousness level is performed to accurately determine the patient's urgency and optimize first aid decisions; through these improvements, it can be ensured that the patient receives the most appropriate treatment in the shortest time, and the success rate of emergency first aid and the survival rate of the patient are improved. The specific scheme is as follows: The information connection method used in the emergency first aid process includes: When the emergency center receives an emergency call, it obtains a patient information set based on the caller's description. The patient information set includes the patient's location, cause of the incident, basic patient information, medical history, and symptoms. The acquired incident location, incident cause, patient medical history and symptom manifestations are converted into incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector respectively by using unique hot encoding; the patient's basic information and the acquired incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector are combined into a unified feature vector; a department recommendation model is constructed, the acquired unified feature vector is used as the input of the department recommendation model, and the probability value of each department corresponding to the current patient is output; Based on the probability values ​​of the current patient corresponding to each department, several reserve departments are selected from all departments; then, for several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated according to the selected reserve departments, and routes are planned for each hospital, and the predicted time required to reach each hospital from the patient's location is obtained; After the ambulance arrives at the patient's location, the patient's vital signs and level of consciousness are evaluated, and the patient's urgency is calculated; then, based on the comprehensive professional capabilities of each hospital, the predicted time required to reach each hospital from the patient's location, and the patient's urgency, the final adaptation score of each hospital is calculated; the hospital with the highest final adaptation score is selected as the target hospital, and the patient is sent to the target hospital according to the pre-planned route.

[0005] Preferably, based on the probability values ​​of the current patient corresponding to each department, several preparation departments are selected from all departments. The specific operation is as follows: The department with the largest probability value is classified as a preparatory department, and the maximum probability value is recorded; a difference threshold is set, and then for any of the other departments, it is determined whether the difference between the probability value of the current department and the maximum probability value is lower than the difference threshold. If so, the current department is classified as a preparatory department; if not, no operation is performed.

[0006] Preferably, for several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated according to the selected preparatory departments. The specific operations are as follows: The closest to the patient For any hospital among the hospitals, obtain the professional ability score of each selected preparatory department in the current hospital , =1, 2, …, ; =1, 2, …, ; Indicates the number of currently selected preparatory departments; based on the probability value of each preparatory department , using the formula Calculate and obtain the comprehensive professional ability score of the current hospital .

[0007] Preferably, the patient's vital signs and consciousness level are evaluated, and the urgency of obtaining the patient is calculated. The specific operations are as follows: Obtain the patient's vital signs data set, which includes heart rate, blood pressure, respiratory rate, body temperature, and blood oxygen saturation. , =1, 2, 3, 4, 5; arrive In turn, it represents heart rate, blood pressure, respiratory rate, body temperature and blood oxygen saturation; the normal range of the vital sign data is set to to , using the formula Calculate the normal median value of the vital sign data ; Then use the formula Calculate the risk score for this vital sign data ; Using the formula Calculate the patient's vital signs risk score ; The Glasgow Coma Scale was used to assess the patient's level of consciousness and obtain the patient's GCS score. , using the formula Calculate the patient's level of consciousness risk score ; Finally, using the formula Calculate the urgency of acquiring the patient .

[0008] Preferably, the final adaptation score of each hospital is calculated and obtained, and the specific operations are as follows: The closest to the patient Any hospital among the hospitals, based on the comprehensive professional ability score of the current hospital and the predicted time required to travel from the patient's location to the current hospital , using the formula Calculate the final adaptation score of the current hospital ;in, An upper limit for the comprehensive professional capability score of each hospital; Adjustment factor, used to adjust the prediction time Scoring the final fit The degree of influence, The value of is obtained through the swarm optimization algorithm.

[0009] Preferably, the acquired incident location, incident cause, patient medical history and symptom manifestation are converted into incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector respectively by using one-hot encoding, and the specific operation is as follows: For any patient information including the location of the incident, cause of the incident, patient history and symptom manifestation, based on the number N of all possible categories of the current patient information, an unknown category is added on the basis of the number N of categories, and according to the category corresponding to the current patient information, the N+1-dimensional one-hot encoding vector corresponding to the current patient information is obtained.

[0010] Preferably, the department recommendation model is established based on FNN, including an input layer, a hidden layer and an output layer; The input layer is used to receive the unified feature vector of the current patient; the hidden layer is used to extract the relationship between the various input features in the unified feature vector using the ReLU activation function; the output layer is used to convert the output of the hidden layer into the predicted probability of each department using the Softmax function. The output layer includes C neurons, where C is the total number of departments, and finally outputs the probability distribution of each department; To train the department recommendation model, the specific operations are as follows: Several historical patient samples are obtained, each of which contains a patient's basic information, an incident location vector, an incident cause vector, a patient's medical history vector, a symptom manifestation vector, and the type of department where the patient finally seeks medical treatment; all historical patient samples are divided into a training set and a validation set, and the parameter-initialized department recommendation model is trained using the training set, wherein the patient's basic information, the incident location vector, the incident cause vector, the patient's medical history vector, and the symptom manifestation vector are used as the department recommendation model, and the type of department where the patient finally seeks medical treatment is used as the target output of the department recommendation model; the department recommendation model is then verified using the validation set to obtain the verification result; the training conditions are set to determine whether the verification result meets the training conditions, and if so, the trained department recommendation model is output; if not, the department recommendation model is continued to be trained using the training set.

[0011] Preferably, obtaining The population optimization algorithm used for the value of is the particle swarm optimization algorithm, and the The specific operations of the value are as follows: Step 1: Randomly generate a swarm of particles, and the position of each particle represents a candidate value, and set the maximum number of iterations; for any particle, initialize the velocity of the particle, and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value of the particle ;in, Indicates the treatment effect of the patient in the final selected target hospital, represents the time it takes for the patient to reach the target hospital; traverse all particles and take the position corresponding to the particle with the highest fitness value as the global optimal position; Step 2: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position; Step 3: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, update the global best position to the individual best position of the particle. Step 4: Repeat steps 2 and 3 until the maximum number of iterations is reached. The global best position obtained is the best value.

[0012] The present invention has the following advantages: The present invention obtains and analyzes the key information of patients, uses unique hot coding and department recommendation models, and can quickly generate probability values ​​for patients corresponding to various departments, thereby determining the preparation department in advance. At the same time, the comprehensive professional capabilities of the hospital, the predicted time spent, and the urgency of the patient are comprehensively considered, and the target hospital is scientifically selected by calculating the final adaptation score. This process not only reduces the time of on-site evaluation, but also ensures that the patient can be sent to the most appropriate hospital, significantly improves the efficiency of emergency treatment, and buys precious treatment time for the patient.

[0013] The present invention plans the optimal route for each hospital in advance, and directly drives along the planned route after confirming the target hospital, thereby further shortening the transfer time, improving the scientificity and rationality of the overall emergency treatment process, and improving the quality of medical emergency services and patient satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The present invention is a flowchart of an information connection method used in an emergency first aid process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0015] In order to enable persons skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0016] This embodiment discloses an information connection method for emergency first aid process, such as Figure 1 As shown, including: When the emergency center receives an emergency call, it obtains a patient information set based on the caller's description. The patient information set includes the patient's location, cause, basic information, medical history and symptoms. The location refers to the specific location where the patient has an emergency. This information is important for judging the department because it can help preliminarily determine the environmental risks or specific types of accidents that the patient may face. For example, if the incident occurs at a construction site, the patient is more likely to suffer from trauma, fractures or heavy objects, which suggests that relevant departments such as surgery or orthopedics need to be considered. On the contrary, if the incident occurs at home, it may be more Many cases are related to medical diseases, such as heart attack, stroke or complications of diabetes. In this case, internal medicine or neurology may be a more appropriate department to choose; the cause of the incident involves the specific event or factor that led to the patient's emergency, which is directly related to the type of injury the patient may suffer or the cause of the disease; for example, if the injury is caused by a traffic accident, then the patient may need emergency surgical intervention, including evaluation by trauma surgery or emergency surgery; if it is an emergency caused by food poisoning, then gastroenterology or emergency department intervention may be required; the patient's basic information usually includes age, gender, etc., which also plays an auxiliary role in judging the department, because People of different ages and genders may face different health risks. For example, children may be more susceptible to infectious diseases or accidental injuries, so pediatrics or the emergency department may be appropriate departments. For the elderly, cardiovascular disease, cerebrovascular disease, or acute exacerbations of chronic diseases may be more common. In this case, internal medicine, cardiovascular department, or neurology department may be more appropriate choices. Patient history refers to the patient's previous health status and disease experience, which can help predict which previous diseases the patient's current symptoms may be related to. For example, if the patient has a history of heart disease and currently has symptoms such as chest pain and difficulty breathing, then the cardiovascular department may be the most appropriate department. If the patient has a history of diabetes, , if the patient is currently experiencing confusion, abnormal blood sugar, etc., then the endocrinology department or the emergency department may be the first choice; symptom manifestations refer to the patient's current physical discomfort or abnormal manifestations, which is the direct basis for judging the department, because different symptoms often point to different diseases and departments; for example, persistent severe headaches may point to the neurology department or the emergency department to rule out serious conditions such as cerebral hemorrhage and cerebral infarction; frequent vomiting and diarrhea may point to the gastroenterology department or the emergency department to deal with possible gastrointestinal infections or food poisoning; by recording and analyzing symptom manifestations in detail, medical staff can more accurately determine the department the patient needs to go to, ensuring that the patient receives timely and effective treatment; Use one-hot encoding to convert the location of the incident, the cause of the incident, the patient's medical history and symptoms into the location vector, the cause vector, the patient's medical history vector and the symptom vector respectively; One-hot encoding is a common method for processing discrete features. It converts the features of each category into a unique binary vector. The location of the incident, the cause of the incident, the patient's medical history and the symptom are all categorical variables. One-hot encoding can convert these categorical variables into numerical form so that they can be processed by machine learning algorithms; The basic information of the patient and the location vector, the cause vector, the patient's medical history vector and the symptom vector obtained are combined into a unified feature vector; Construct a department recommendation model, use the obtained unified feature vector as the input of the department recommendation model, and output the probability value of the current patient corresponding to each department; Based on the probability values ​​of the current patient corresponding to each department, several reserve departments are selected from all departments; then, for several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated according to the selected reserve departments, and routes are planned for each hospital, and the predicted time required to reach each hospital from the patient's location is obtained. These times need to take into account real-time traffic conditions to ensure the accuracy of the prediction; calculating the comprehensive professional capabilities of each hospital can ensure that the patient is sent to a hospital with professional capabilities that are more in line with the current patient, which not only improves the treatment effect, but also optimizes the allocation of medical resources to ensure that the patient can receive the most professional treatment; at the same time, by planning the optimal route for each hospital in advance and obtaining the predicted time, it can ensure that the patient arrives at the hospital in the shortest time; this reduces the transfer time, improves the efficiency of emergency treatment, and gains precious treatment time for the patient; After the ambulance arrives at the patient's location, the patient's vital signs and consciousness level are evaluated, and the patient's urgency is calculated; then, based on the comprehensive professional capabilities of each hospital, the predicted time required to reach each hospital from the patient's location, and the patient's urgency, the final adaptation score of each hospital is calculated; the hospital with the highest final adaptation score is used as the target hospital, and the patient is sent to the target hospital according to the pre-planned route; by calculating the final adaptation score, the present invention can comprehensively consider the comprehensive professional capabilities of the hospital, the predicted time and the patient's urgency, scientifically select the target hospital, and optimize the route planning; this not only improves the scientificity and accuracy of emergency decision-making, but also optimizes hospital selection, improves emergency efficiency, ensures that patients can receive the most professional treatment in a timely manner, and improves patient satisfaction and survival rate.

[0017] Preferably, based on the probability values ​​of the current patient corresponding to each department, several preparation departments are selected from all departments. The specific operation is as follows: The department with the largest probability value is classified as a preparatory department, and the maximum probability value is recorded; a difference threshold is set, and then for any of the other departments, whether the difference between the probability value of the current department and the maximum probability value is lower than the difference threshold is determined; if so, the current department is classified as a preparatory department; if not, no operation is performed; this ensures that the selected preparatory departments include not only the department with the highest probability, but also other departments with similar probabilities, thereby improving the flexibility and accuracy of the selection.

[0018] Preferably, for several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated according to the selected preparatory departments. The specific operations are as follows: The closest to the patient For any hospital among the hospitals, obtain the professional ability score of each selected preparatory department in the current hospital , =1, 2, …, ; =1, 2, …, ; Indicates the number of currently selected reserve departments; these scores can be based on the hospital's historical data, expert evaluations, or public medical quality reports; based on the probability value of each reserve department , using the formula Calculate and obtain the comprehensive professional ability score of the current hospital .

[0019] Preferably, the patient's vital signs and consciousness level are evaluated, and the urgency of obtaining the patient is calculated. The specific operations are as follows: Obtain the patient's vital signs data set, which includes heart rate, blood pressure, respiratory rate, body temperature, and blood oxygen saturation. , =1, 2, 3, 4, 5; arrive In turn, it represents heart rate, blood pressure, respiratory rate, body temperature and blood oxygen saturation; the normal range of the vital sign data is set to to , using the formula Calculate the normal median value of the vital sign data ; Then use the formula Calculate the risk score for this vital sign data ; Using the formula Calculate the patient's vital signs risk score ; The Glasgow Coma Scale was used to assess the patient's level of consciousness and obtain the patient's GCS score. , using the formula Calculate the patient's level of consciousness risk score ; Finally, using the formula Calculate the urgency of acquiring the patient .

[0020] Preferably, the final adaptation score of each hospital is calculated and obtained, and the specific operations are as follows: The closest to the patient Any hospital among the hospitals, based on the comprehensive professional ability score of the current hospital and the predicted time required to travel from the patient's location to the current hospital , using the formula Calculate the final adaptation score of the current hospital ;in, An upper limit for the comprehensive professional capability score of each hospital; Adjustment factor, used to adjust the prediction time Scoring the final fit The degree of influence, The value of is obtained through the swarm optimization algorithm.

[0021] Preferably, the acquired incident location, incident cause, patient medical history and symptom manifestation are converted into incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector respectively by using one-hot encoding, and the specific operation is as follows: For any patient information including the location of the incident, the cause of the incident, the patient's medical history and the symptoms, based on the number N of all possible categories of the current patient information, an unknown category is added on the basis of the number N, and according to the category corresponding to the current patient information, the N+1-dimensional unique hot encoding vector corresponding to the current patient information is obtained; the reason for adding unknown categories is that in reality, the information that the caller of the emergency call can express is unstable, and there may be situations where the dispatch center cannot accurately identify or classify certain locations of the incident, causes of the incident, patient's medical history and symptoms. If the dispatch center cannot determine the specific category, it can classify the information as "unknown", and the model can still handle this situation to avoid errors or failures caused by missing information.

[0022] Preferably, the department recommendation model is established based on FNN (feedforward neural network), including an input layer, a hidden layer and an output layer; The input layer is used to receive the unified feature vector of the current patient; the hidden layer is used to extract the relationship between the various input features in the unified feature vector using the ReLU activation function; the output layer is used to convert the output of the hidden layer into the predicted probability of each department using the Softmax function. The output layer includes C neurons, where C is the total number of departments, and finally outputs the probability distribution of each department; To train the department recommendation model, the specific operations are as follows: Several historical patient samples are obtained, each of which contains a patient's basic information, an incident location vector, an incident cause vector, a patient's medical history vector, a symptom manifestation vector, and the type of department where the patient finally seeks medical treatment; all historical patient samples are divided into a training set and a validation set, and the parameter-initialized department recommendation model is trained using the training set, wherein the patient's basic information, the incident location vector, the incident cause vector, the patient's medical history vector, and the symptom manifestation vector are used as the department recommendation model, and the type of department where the patient finally seeks medical treatment is used as the target output of the department recommendation model; the department recommendation model is then verified using the validation set to obtain the verification result; the training conditions are set to determine whether the verification result meets the training conditions, and if so, the trained department recommendation model is output; if not, the department recommendation model is continued to be trained using the training set.

[0023] Preferably, obtaining The population optimization algorithm used for the value of is the particle swarm optimization algorithm, which is an optimization algorithm based on swarm intelligence and simulates the foraging behavior of bird groups. In PSO (Particle Swarm Optimization), each particle represents a potential solution. The particle flies in the solution space and adjusts its position and speed through individual experience and group experience. Each particle records its own best position (individual best position), and the best position of all particles in the group (global best position) is also recorded. By iteratively updating the speed and position of the particles, PSO can find the global optimal solution and is widely used in various optimization problems. In the present invention, the adjustment coefficient can be scientifically determined by the particle swarm optimization algorithm. , ensure that the influence of prediction time on the final adaptation score is reasonable, improve the scientificity and accuracy of the score, and obtain The specific operations of the value are as follows: Step 1: Randomly generate a swarm of particles, and the position of each particle represents a candidate value, and set the maximum number of iterations; for any particle, initialize the velocity of the particle, and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value of the particle ;in, Indicates the treatment effect of the patient in the final selected target hospital, represents the time it takes for the patient to reach the target hospital; traverse all particles and take the position corresponding to the particle with the highest fitness value as the global optimal position; Step 2: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position; Step 3: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, update the global best position to the individual best position of the particle. Step 4: Repeat steps 2 and 3 until the maximum number of iterations is reached. The global best position obtained is the best value.

[0024] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all such improvements and changes should fall within the scope of protection of the appended claims of the present invention. Parts not described in detail in this specification belong to the prior art known to those skilled in the art.

Claims

1. An information docking method for emergency first aid, characterized in that: include: When the emergency center receives an emergency call, it obtains a patient information set based on the caller's description. The patient information set includes the patient's location, cause of the incident, basic patient information, medical history, and symptoms. The acquired incident location, incident cause, patient medical history and symptom manifestations are converted into incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector respectively by using unique hot encoding; the patient's basic information and the acquired incident location vector, incident cause vector, patient medical history vector and symptom manifestation vector are combined into a unified feature vector; a department recommendation model is constructed, the acquired unified feature vector is used as the input of the department recommendation model, and the probability value of each department corresponding to the current patient is output; Based on the probability values ​​of the current patient corresponding to each department, several preparation departments are selected from all departments; Then, for several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated based on the selected preparation departments, and routes are planned for each hospital, and the predicted time required to reach each hospital from the patient's location is obtained; After the ambulance arrives at the patient's location, the patient's vital signs and level of consciousness are evaluated, and the patient's urgency is calculated; then, based on the comprehensive professional capabilities of each hospital, the predicted time required to reach each hospital from the patient's location, and the patient's urgency, the final adaptation score of each hospital is calculated; the hospital with the highest final adaptation score is selected as the target hospital, and the patient is sent to the target hospital according to the pre-planned route.

2. The information docking method for emergency first aid according to claim 1, characterized in that: Based on the probability values ​​of the current patient corresponding to each department, several preparation departments are selected from all departments. The specific operations are as follows: The department with the largest probability value is classified as a preparatory department, and the maximum probability value is recorded; a difference threshold is set, and then for any of the other departments, it is determined whether the difference between the probability value of the current department and the maximum probability value is lower than the difference threshold. If so, the current department is classified as a preparatory department; if not, no operation is performed.

3. The information docking method for emergency first aid according to claim 2, characterized in that: For several hospitals closest to the patient, the comprehensive professional capabilities of each hospital are calculated based on the selected preparatory departments. The specific operations are as follows: The closest to the patient For any hospital among the hospitals, obtain the professional ability score of each selected preparatory department in the current hospital , =1, 2, …, ; =1, 2, …, ; Indicates the number of currently selected preparatory departments; based on the probability value of each preparatory department , using the formula Calculate and obtain the comprehensive professional ability score of the current hospital .

4. The information docking method for emergency first aid according to claim 3, characterized in that: Assess the patient's vital signs and level of consciousness and calculate the urgency of obtaining the patient. The specific steps are as follows: Obtain the patient's vital signs data set, which includes heart rate, blood pressure, respiratory rate, body temperature, and blood oxygen saturation. , =1, 2, 3, 4, 5; arrive It indicates heart rate, blood pressure, respiratory rate, body temperature and blood oxygen saturation in turn; Set the normal range of the vital sign data to to , using the formula Calculate the normal median value of the vital sign data ; Then use the formula Calculate the risk score for this vital sign data ; Using the formula Calculate the patient's vital signs risk score ; The Glasgow Coma Scale was used to assess the patient's level of consciousness and obtain the patient's GCS score. , using the formula Calculate the patient's level of consciousness risk score ; Finally, using the formula Calculate the urgency of acquiring the patient .

5. The information docking method for emergency first aid according to claim 4, characterized in that: Calculate and obtain the final adaptation score of each hospital. The specific operations are as follows: The closest to the patient Any hospital among the hospitals, based on the comprehensive professional ability score of the current hospital and the predicted time required to travel from the patient's location to the current hospital , using the formula Calculate the final adaptation score of the current hospital ;in, An upper limit for the comprehensive professional capability score of each hospital; Adjustment factor, used to adjust the prediction time Scoring the final fit The degree of influence, The value of is obtained through the swarm optimization algorithm.

6. The information docking method for emergency first aid according to claim 5, characterized in that: Use unique hot encoding to convert the location of the incident, the cause of the incident, the patient's medical history and symptoms into the location vector, the cause of the incident vector, the patient's medical history vector and the symptom vector respectively. The specific operations are as follows: For any patient information including the location of the incident, cause of the incident, patient history and symptom manifestation, based on the number N of all possible categories of the current patient information, an unknown category is added on the basis of the number N of categories, and according to the category corresponding to the current patient information, the N+1-dimensional one-hot encoding vector corresponding to the current patient information is obtained.

7. The information docking method for emergency first aid according to claim 6, characterized in that: The department recommendation model is built based on FNN, including input layer, hidden layer and output layer; The input layer is used to receive the unified feature vector of the current patient; the hidden layer is used to extract the relationship between the various input features in the unified feature vector using the ReLU activation function; the output layer is used to convert the output of the hidden layer into the predicted probability of each department using the Softmax function. The output layer includes C neurons, where C is the total number of departments, and finally outputs the probability distribution of each department; To train the department recommendation model, the specific operations are as follows: Several historical patient samples are obtained, each of which contains a patient's basic information, an incident location vector, an incident cause vector, a patient's medical history vector, a symptom manifestation vector, and the type of department where the patient finally seeks medical treatment; all historical patient samples are divided into a training set and a validation set, and the parameter-initialized department recommendation model is trained using the training set, wherein the patient's basic information, the incident location vector, the incident cause vector, the patient's medical history vector, and the symptom manifestation vector are used as the department recommendation model, and the type of department where the patient finally seeks medical treatment is used as the target output of the department recommendation model; the department recommendation model is then verified using the validation set to obtain the verification result; the training conditions are set to determine whether the verification result meets the training conditions, and if so, the trained department recommendation model is output; if not, the department recommendation model is continued to be trained using the training set.

8. The information docking method for emergency first aid according to claim 7, characterized in that: Get The population optimization algorithm used for the value of is the particle swarm optimization algorithm, and the The specific operations of the value are as follows: Step 1: Randomly generate a swarm of particles, and the position of each particle represents a candidate value, and set the maximum number of iterations; for any particle, initialize the velocity of the particle, and use the current position as the individual optimal position of the particle; use the formula Calculate the fitness value of the particle ;in, Indicates the treatment effect of the patient in the final selected target hospital, represents the time it takes for the patient to reach the target hospital; traverse all particles and take the position corresponding to the particle with the highest fitness value as the global optimal position; Step 2: For any particle, update the speed and position of the particle, and recalculate the fitness value; if the updated fitness value is higher than the fitness value of the individual best position of the particle, then update the individual best position of the particle to the current position; Step 3: Traverse all particles. If the fitness value of the individual best position of any particle is higher than the fitness value of the global best position, update the global best position to the individual best position of the particle. Step 4: Repeat steps 2 and 3 until the maximum number of iterations is reached. The global best position obtained is the best value.

Images