Critical disease inter-hospital referral patient scoring and resource allocation method, system and device and medium

Abstract

The invention relates to a critical disease inter-hospital referral patient scoring and resource allocation method, system and device, and a medium. The method comprises the following steps: determining referral demands of patients; constructing a resource matching model in combination with the illness state and the geographic position of the patient and the medical specialist information of the hospital to generate a recommended hospital list, and determining a transfer-in hospital and a referral time window; calculating an optimal referral path and the shortest time required by referral in the referral time window according to the transfer-in hospital and the referral time window in combination with the geographic information and the real-time road condition data; and allocating the ambulances to the patients within the available range of the ambulances based on the globally optimized resource allocation model. The advanced scoring rule and the resource allocation algorithm are introduced, the treatment efficiency of critical patients is improved, more scientific and objective decision support is provided, and unnecessary referral treatment and repeated examination are reduced, so that the medical cost is effectively reduced, medical errors are reduced, and the satisfaction degree of the patients is greatly improved.

Description

Technical Field

[0001] The present invention relates to the field of medical information technology, and in particular to a method, system, device and medium for scoring and allocating resources for critically ill inter-hospital referral patients. Background Art

[0002] With the continuous advancement of medical technology and the public's expectations for the quality of medical services, emergency treatment of critically ill patients has become crucial. The conditions of these patients are serious and change rapidly, so timely and accurate medical intervention plays a decisive role in protecting the lives of patients. In the existing medical system, inter-hospital referral is a key link in the treatment of critically ill patients, and its efficiency directly affects the treatment effect of patients.

[0003] The existing referral system has several significant shortcomings:

[0004] (1) Lack of intelligent decision support: The existing system is unable to conduct intelligent analysis of the patient's condition and therefore cannot provide scientific data support for referral decisions, resulting in the decision-making process relying more on the subjective judgment and experience of medical staff.

[0005] (2) Difficulty in resource allocation: Before patients arrive at the hospital, hospitals usually do not pre-dispatch and allocate medical equipment and resources. This often leads to irrational resource allocation and even the lack of key equipment, which seriously affects the treatment effect.

[0006] (3) The referral process is complex and repetitive: The existing referral process involves multiple steps and links, including information collection, transmission, input and subsequent processing. These links often involve repetitive work, which not only reduces work efficiency but also increases the possibility of errors.

[0007] In summary, the current inter-hospital referral system for critically ill patients urgently needs to be improved in terms of providing intelligent decision-making support, optimizing resource allocation, and simplifying the referral process, so as to ultimately improve the success rate of treatment for critically ill patients. Summary of the invention

[0008] 1. Technical issues to be resolved

[0009] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method, system, device and medium for scoring and allocating resources for critically ill inter-hospital referral patients, which solves the technical problems of insufficient intelligent decision support, optimized resource allocation and simplified referral process.

[0010] (II) Technical solution

[0011] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0012] In a first aspect, an embodiment of the present invention provides a method for scoring and allocating resources for critically ill inter-hospital referral patients, comprising:

[0013] Determine the patient's referral needs using preset condition score calculation rules and / or machine learning-based classification algorithms;

[0014] For patients who are determined to need referral, a recommendation model is built based on the patient's condition, geographic location, and the hospital's medical expertise to generate a list of recommended hospitals that can accept the corresponding condition, and the transferee hospital and referral time window are determined in the recommended hospital list;

[0015] According to the receiving hospital and referral time window, combined with geographic information and real-time traffic data, the best referral route and the shortest time required for referral within the referral time window are calculated;

[0016] Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. Within the available range of ambulances, ambulances are automatically assigned to patients according to the priority score.

[0017] Optionally, using a preset condition score calculation rule and / or a classification algorithm based on machine learning, determining the patient's referral needs includes:

[0018] Obtain the patient's disease type, vital signs, laboratory test results and required equipment information;

[0019] The preset disease score calculation rules are used to output the disease score in combination with the patient's vital signs and laboratory tests;

[0020] The pre-built logistic regression model is trained using the patient's disease type and vital signs, and the optimal coefficients are solved by minimizing the loss function;

[0021] Substitute the patient's characteristic values ​​into the trained logistic regression model formula to calculate the probability of emergency referral, and output the referral judgment result based on the set threshold;

[0022] Determine the patient's referral needs based on the disease score and / or referral judgment results, combined with the required device information;

[0023] in,

[0024] The preset calculation rules of the disease score include: vital signs scoring items and laboratory examination scoring items. For each indicator in the vital signs scoring items and laboratory examination scoring items, the score increases by 1 point every time a preset increase or decrease value unit is reached until the preset maximum score limit is reached. The total score of the disease score is obtained by the sum of the vital signs score and the laboratory examination score;

[0025] The formula for the logistic regression model is:

[0026] P(y=1|X)=1 / (1+exp(-(β0+β1x1+β2x2+...+β n *x n )));

[0027] Where P(y=1|X) represents the probability that a patient needs emergency referral given the patient’s characteristic X. The optimal coefficients include β0…β n , β0 is the intercept term, β1...β n are the coefficients corresponding to the characteristics of each patient, x1…x n are the characteristic values ​​of the patient, and exp is the natural exponential function.

[0028] Optionally, for patients who are determined to need referral, a recommendation model is constructed to generate a list of recommended hospitals that can receive the corresponding symptoms in combination with the patient's condition, geographical location, and medical expertise of the hospital, and the transferee hospital and referral time window are determined in the recommended hospital list, including:

[0029] Obtain information on the patient's condition identified for referral, the patient's geographic location, the hospital's geographic location, and the hospital's medical expertise;

[0030] According to the preset weight coefficients, the recommendation model is constructed by combining the patient's condition, geographical location and the hospital's medical expertise;

[0031] Determine the first feature item by calculating the similarity between the patient's condition and the hospital's medical expertise information;

[0032] Determine the second feature item by measuring the geographical location similarity between the patient and the hospital or calculating the geographical coordinate distance;

[0033] determining a third feature item based on at least one of a rating, a patient evaluation, and a professional certification in the medical expertise information of the hospital;

[0034] Substituting the weight coefficient and the first feature item, the second feature item, and the third feature item into the pre-built recommendation model, the matching degree between the patient and each hospital is calculated;

[0035] Sort each hospital according to the matching degree to generate a list of recommended hospitals that can receive the corresponding disease and determine the transfer-in hospital and referral time window in the recommended hospital list;

[0036] Among them, the recommended model is:

[0037] M = α*P(patient, hospital) + β*L(patient, hospital) + γ*H(patient, hospital);

[0038] Wherein, M is the matching degree, α, β, and γ are the first weight coefficient, the second weight coefficient, and the third weight coefficient, respectively, P (patient, hospital) is the first feature item, L (patient, hospital) is the second feature item, and H (patient, hospital) is the third feature item.

[0039] Optionally, according to the receiving hospital and the referral time window, combined with geographic information and real-time traffic data, the optimal referral path and the shortest time required for referral within the referral time window are calculated, including:

[0040] Determine the referral time constraint according to the location information of the receiving hospital and the given referral time window;

[0041] Based on the real-time traffic data including road congestion, traffic flow, and accident information, evaluate the factors affecting the referral route selection;

[0042] According to the transferring-out hospital and the receiving hospital, a path planning algorithm is used to calculate multiple possible referral paths from the transferring-out hospital to the receiving hospital;

[0043] For multiple possible referral paths, the estimated driving time of the ambulance on each possible referral path is obtained according to the factors affecting the selection of the referral path under the current road conditions;

[0044] According to the referral time constraint and the estimated driving time of the ambulance on each possible referral path, multiple paths that meet the referral time window requirements are screened out;

[0045] From multiple paths that meet the referral time window requirements, the best referral path is determined based on road condition factors including road stability information, total path length, and estimated driving time, and the shortest time required for referral within the referral time window along the best referral path is determined.

[0046] Optionally, after calculating the best referral route and the shortest time required for referral within the referral time window based on the referral recipient hospital and the referral time window, in combination with geographic information and real-time traffic data, the following further includes:

[0047] Upon receipt of manual intervention information including the expected referral time, ambulance dispatch is immediately triggered to ensure that the ambulance can depart according to the predetermined referral time window and the optimal referral route.

[0048] Optionally, based on a pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated in combination with the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. Within the available range of the ambulance, an ambulance is automatically allocated to the patient according to the priority score, including:

[0049] Obtain data on the patient's acuity score, minimum time required for referral within the referral time window, optimal referral route, and ambulance availability and status;

[0050] Based on a pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance;

[0051] Automatically assign ambulances to patients based on their real-time location, number, and priority score to ensure that resources are maximized and ambulance availability is not exceeded;

[0052] Among them, the global optimization resource allocation model is:

[0053] P = δ*S(patient) / D(patient, target hospital)+ε*A(ambulance);

[0054] Wherein, δ and ε are the first allocation ratio and the second allocation ratio respectively, S(patient) represents the patient's condition score, D(patient, target hospital) represents the shortest time and the best referral path required for the patient to be transferred to the transfer hospital within the referral time window, A(ambulance) represents the availability and status of the ambulance, and the availability and status of the ambulance include the number, location and whether the ambulance is idle; and the global optimization resource allocation model is provided with a feedback mechanism for dynamically adjusting the weight coefficient in the global optimization resource allocation model according to the actual treatment effect, so as to optimize the resource allocation algorithm.

[0055] Optionally, based on a pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated in combination with the patient's condition, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. After the ambulance is automatically allocated to the patient according to the priority score within the available range of the ambulance, it also includes:

[0056] After the ambulance dispatch is completed, a message notification is automatically triggered, and the relevant information is sent to the receiving hospital in the form of voice, text message, or WeChat, and an application form containing the referral patient information is submitted for review and handover.

[0057] In a second aspect, an embodiment of the present invention provides a critically ill inter-hospital referral patient scoring and resource allocation system, comprising:

[0058] A referral demand determination module, used to determine the patient's referral demand by using a preset condition score calculation rule and / or a classification algorithm based on machine learning;

[0059] The hospital recommendation module is used to build a recommendation model for patients who need to be referred, combining the patient's condition, geographical location and the hospital's medical expertise information to generate a list of recommended hospitals that can accept the corresponding symptoms, and determine the transfer-in hospital and referral time window in the recommended hospital list;

[0060] The path and time calculation module is used to calculate the best referral path and the shortest time required for referral within the referral time window based on the referral hospital and referral time window, combined with geographic information and real-time traffic data;

[0061] The ambulance allocation module is used to dynamically calculate the patient's priority score based on a pre-built global optimization resource allocation model, combining the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance, and automatically allocate an ambulance to the patient according to the priority score within the available range of the ambulance.

[0062] In a third aspect, an embodiment of the present invention provides a critically ill inter-hospital referral patient scoring and resource allocation device, comprising: at least one database; and a memory communicatively connected to the at least one database; wherein the memory stores instructions executable by the at least one database, and the instructions are executed by the at least one database so that the at least one database can execute the critically ill inter-hospital referral patient scoring and resource allocation method as described above.

[0063] In a fourth aspect, an embodiment of the present invention provides a computer-readable medium having computer-executable instructions stored thereon, which, when executed by a processor, implement the critically ill inter-hospital referral patient scoring and resource allocation method as described above.

[0064] (III) Beneficial effects

[0065] The beneficial effects of the present invention are:

[0066] By introducing advanced scoring rules, the present invention can quickly and accurately assess the urgency and referral priority of critically ill patients. This innovative function significantly improves the decision-making speed of doctors and ensures a rapid response in emergency situations. At the same time, the introduction of resource allocation algorithms achieves the optimal allocation of ambulances and medical resources. This not only reduces the idle driving time and waiting time of ambulances, improves the efficiency of ambulance use, but also ensures that patients are promptly sent to the most suitable medical institution for their condition. This improvement not only improves the efficiency of treatment, but also greatly optimizes the utilization of medical resources.

[0067] At the same time, the scoring rules and resource allocation algorithms of the present invention are based on a large amount of historical data and real-time data, providing doctors with more scientific and objective decision-making support. This data-driven approach not only improves the accuracy and effectiveness of decision-making, but also provides medical institutions with valuable data insights, which helps to continuously improve and optimize medical services.

[0068] Furthermore, by optimizing the referral process, the present invention reduces unnecessary referrals and repeated examinations, thereby effectively reducing medical costs. In addition, the rapid response and efficient referral process significantly reduces patients' waiting time and improves their overall satisfaction with medical services. It is worth mentioning that the introduction of scoring rules and resource allocation algorithms also reduces errors in human judgment. Through data-driven decision support, the risk of medical errors is greatly reduced, further enhancing patients' trust in the medical system.

[0069] In summary, the present invention significantly improves the efficiency of treating critically ill patients by introducing advanced scoring rules and resource allocation algorithms, provides more scientific and objective decision support, reduces unnecessary referrals and repeated examinations, thereby effectively reducing medical costs, reducing medical errors, and greatly improving patient satisfaction. These excellent effects together reflect the innovative value and practical value of the present invention in the medical field. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A flowchart of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0071] Figure 2 A specific flow chart of step S1 of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0072] Figure 3 A specific flow chart of step S2 of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0073] Figure 4 A specific flow chart of step S3 of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0074] Figure 5 A specific flow chart of step S4 of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0075] Figure 6 A schematic diagram of the overall process of a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention;

[0076] Figure 7 A schematic diagram of a specific implementation unit of a critically ill inter-hospital referral patient scoring and resource allocation system proposed in an embodiment of the present invention. DETAILED DESCRIPTION

[0077] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0078] like Figure 1 As shown, a method for scoring and allocating resources for critically ill inter-hospital referral patients proposed in an embodiment of the present invention includes: determining the patient's referral needs using a preset condition score calculation rule and / or a classification algorithm based on machine learning; for patients determined to need referral, combining the patient's condition, geographical location, and medical expertise information of the hospital, building a recommendation model to generate a list of recommended hospitals that can receive corresponding conditions, and determining the transferee hospital and referral time window in the recommended hospital list; calculating the optimal referral path and the shortest time required for referral within the referral time window based on the transferee hospital and the referral time window, combined with geographical information and real-time traffic data; based on a pre-constructed global optimization resource allocation model, dynamically calculating the patient's priority score in combination with the patient's condition, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance, and automatically allocating an ambulance to the patient according to the priority score within the available range of the ambulance.

[0079] By introducing advanced scoring rules, the present invention can quickly and accurately assess the urgency and referral priority of critically ill patients. This innovative function significantly improves the decision-making speed of doctors and ensures a rapid response in emergency situations. At the same time, the introduction of resource allocation algorithms achieves the optimal allocation of ambulances and medical resources. This not only reduces the idle driving time and waiting time of ambulances, improves the efficiency of ambulance use, but also ensures that patients are promptly sent to the most suitable medical institution for their condition. This improvement not only improves the efficiency of treatment, but also greatly optimizes the utilization of medical resources.

[0080] At the same time, the scoring rules and resource allocation algorithms of the present invention are based on a large amount of historical data and real-time data, providing doctors with more scientific and objective decision-making support. This data-driven approach not only improves the accuracy and effectiveness of decision-making, but also provides medical institutions with valuable data insights, which helps to continuously improve and optimize medical services.

[0081] Furthermore, by optimizing the referral process, the present invention reduces unnecessary referrals and repeated examinations, thereby effectively reducing medical costs. In addition, the rapid response and efficient referral process significantly reduces patients' waiting time and improves their overall satisfaction with medical services. It is worth mentioning that the introduction of scoring rules and resource allocation algorithms also reduces errors in human judgment. Through data-driven decision support, the risk of medical errors is greatly reduced, further enhancing patients' trust in the medical system.

[0082] In summary, the present invention significantly improves the efficiency of treatment of critically ill patients by introducing advanced scoring rules and resource allocation algorithms, provides more scientific and objective decision support, reduces unnecessary referrals and repeated examinations, thereby effectively reducing medical costs, reducing medical errors, and greatly improving patient satisfaction. These excellent effects together reflect the innovative value and practical value of the present invention in the medical field.

[0083] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0084] like Figure 1 As shown, an embodiment of the present invention provides a method for scoring and allocating resources for critically ill inter-hospital referral patients, comprising:

[0085] S1. Determine the patient's referral needs using preset condition score calculation rules and / or machine learning-based classification algorithms.

[0086] Before referral, a score is given based on the patient's vital signs and laboratory test results. The higher the score, the more serious the condition and the more urgent the referral. This scoring method ensures that the referral decision for each patient is based on objective and quantifiable data, thereby reducing the bias caused by subjective judgment. At the same time, in the process of inter-hospital referral of critically ill patients, a classification algorithm based on machine learning is used to assess whether the patient needs referral treatment, so as to assist doctors in making more accurate and timely decisions.

[0087] Therefore, based on the condition score and the results of the machine learning-based classification algorithm, the patient's referral needs can be evaluated more scientifically, objectively and accurately, thereby ensuring that every patient can receive timely and appropriate medical treatment.

[0088] Furthermore, if Figure 2 As shown, step S1 includes:

[0089] S11. Obtain the patient's disease type, vital signs, laboratory test results and required equipment information. This information is the basis for subsequent disease assessment and referral decision-making.

[0090] S12. Output the condition score by using the preset condition score calculation rules in combination with the patient's vital signs and laboratory tests.

[0091] According to the pre-set disease score calculation rules, the patient's vital signs and laboratory test results are comprehensively analyzed to obtain a score that reflects the severity of the patient's current disease. This score is a quantitative representation of the patient's condition, which helps medical staff to understand the patient's condition more intuitively.

[0092] S13. Use the patient's disease type and vital signs to train the pre-built logistic regression model, and solve the optimal coefficients by minimizing the loss function.

[0093] S14. Substitute the patient's characteristic values ​​into the trained logistic regression model formula, calculate the probability of requiring emergency referral, and output the referral judgment result according to the set threshold.

[0094] After the logistic regression model is trained, the patient's feature values ​​are substituted into the model to calculate the probability of emergency referral. This probability value indicates the possibility that the patient needs emergency referral under the current condition. According to the set threshold, this probability value is judged and the referral judgment result is output. If the probability value exceeds the set threshold, it is judged that the patient needs emergency referral; otherwise, it is judged that the patient does not need emergency referral. This step is an important link in converting the model output into actual decision-making.

[0095] S15. Determine the patient's referral needs based on the disease score and / or referral judgment results and the required equipment information.

[0096] The patient's condition score and referral judgment result are comprehensively considered to determine whether referral is needed. If the patient's condition score exceeds a certain value and the referral judgment result also shows that urgent referral is required, then the patient will be determined to have a referral need and the medical staff will be prompted to take appropriate referral measures. If the condition score is low or the referral judgment result is that no referral is required, then it is considered that the patient's current condition can be properly treated in the current medical institution and no referral is required.

[0097] Among them, the preset condition score calculation rules include: vital signs scoring items and laboratory examination scoring items. For each indicator in the vital signs scoring items and laboratory examination scoring items, the score increases by 1 point for each set increase or decrease numerical unit reached until the preset maximum score limit is reached. The total score of the condition score is obtained by the sum of the vital signs score and the laboratory examination score.

[0098] In a specific embodiment, the preset disease score calculation rule is:

[0099] Vital sign scores (e.g., each with a maximum score of 10);

[0100] Heart rate: every 10 beats / minute increase, add 1 point;

[0101] Blood pressure: 1 point for every 10 mmHg decrease;

[0102] Respiratory rate: 1 point for every 5 breaths / minute increase;

[0103] Laboratory test scores (e.g., each with a maximum score of 10);

[0104] Hemoglobin: 1 point for every 1 g / dL decrease;

[0105] Platelet count: add 1 point for every 50,000 / μL decrease;

[0106] Creatinine: 1 point for every 0.5 mg / dL increase;

[0107] Total score calculation: vital signs score + laboratory test score;

[0108] The specific calculation steps are as follows:

[0109] Assume a patient has the following vital signs and laboratory test results:

[0110] Heart rate: 120 beats / minute;

[0111] Blood pressure: 90 / 60 mmHg;

[0112] Respiratory rate: 24 times / minute;

[0113] Hemoglobin: 9 g / dL;

[0114] Platelet count: 100,000 / μL;

[0115] Creatinine: 1.5 mg / dL;

[0116] According to the above-mentioned disease score calculation rules, we can calculate:

[0117] Vital signs score: (120-60) / 10+(90-90) / 10+(24-16) / 5=6+0+1.6=7.6 points;

[0118] Laboratory test score: (9-12) / 1+(100-150) / 5+(1.5-1) / 0.5=-3+-10+3=-10 points;

[0119] Total score: 7.6+(-10)=-2.4 points;

[0120] In this example, the patient's total score is negative, which may indicate that the score needs to be adjusted or that the patient's condition may not be as severe as initially assessed.

[0121] Next, suppose there is a set of patient data, which includes multiple physiological indicators of the patient (such as heart rate, blood pressure, respiratory rate, etc.) and medical history, as well as the corresponding label of whether emergency referral is needed (for example, 1 for referral and 0 for no referral). The goal is to build a logistic regression model that can predict whether the patient needs emergency referral based on his or her characteristics. The formula of the logistic regression model is:

[0122] P(y=1|X)=1 / (1+exp(-(β0+β1x1+β2x2+...+β n *x n )));

[0123] Where P(y=1|X) represents the probability that a patient needs emergency referral given the patient’s characteristic X. The optimal coefficients include β0…β n , β0 is the intercept term, β1...β n are the coefficients corresponding to the characteristics of each patient, x1…x n are the characteristic values ​​of the patient, and exp is the natural exponential function.

[0124] Based on the above formula of the logistic regression model, in practical applications, we first need to collect a large amount of historical data, including patient characteristics and corresponding labels. Then, we use these data to train the logistic regression model and solve the optimal coefficients β0…β by minimizing the loss function (such as cross entropy loss). n Once the model is trained, it can be used to predict new cases. For each new patient, its feature value is substituted into the formula of the logistic regression model to calculate the probability of needing urgent referral. Based on this probability value, a threshold (for example, 0.5) can be set. When the probability exceeds the threshold, the patient is considered to need urgent referral; otherwise, the patient is considered not to need urgent referral.

[0125] S2. For patients who are determined to need referral, a recommendation model is constructed based on the patient's condition, geographical location, and the hospital's medical expertise to generate a list of recommended hospitals that can receive the corresponding symptoms, and the transferee hospital and referral time window are determined in the recommended hospital list.

[0126] Furthermore, if Figure 3 As shown, step S2 includes:

[0127] S21. Obtain the medical condition of the patient who needs to be referred, the patient's geographical location, the geographical location of the hospital, and the medical expertise of the hospital.

[0128] First of all, the present invention will comprehensively collect and organize relevant information of patients who need to be referred. The disease characteristics cover key information such as the type of disease, severity, and treatment stage of the patient, which are the basis for evaluating the patient's needs and whether the hospital can effectively treat the disease. Geographical location characteristics include the patient's current specific location and the hospital's geographical location, which helps to determine the physical distance between the patient and the hospital, thereby evaluating the feasibility and timeliness of the referral. The hospital's expertise reflects the hospital's professional capabilities and treatment levels in different disease areas, and is an important basis for determining the transfer hospital.

[0129] S22. Construct a recommendation model based on the preset weight coefficients, combined with the patient's condition, geographical location and the hospital's medical expertise.

[0130] After collecting enough feature information, a recommendation model is constructed based on the preset weight coefficients. The weight coefficients are set based on expert experience, historical data, or obtained through machine learning algorithms, aiming to accurately reflect the relative importance of each feature in the referral decision. Combining the patient's condition characteristics, geographical location characteristics, and hospital expertise characteristics, the model can comprehensively consider multiple factors and more accurately recommend the appropriate transfer hospital.

[0131] S23. Determine the first feature item by calculating the similarity between the patient's condition and the hospital's medical expertise information.

[0132] This step uses technologies such as text mining, natural language processing or machine learning to quantitatively assess the degree of match between the patient's description of the condition and the hospital's expertise. The higher the similarity, the more capable and experienced the hospital is in treating the patient's disease, making it more likely to become a suitable transfer hospital.

[0133] S24. Determine the second feature item by measuring the geographical location similarity between the patient and the hospital or calculating the geographical coordinate distance.

[0134] In addition to the matching degree between the condition and the expertise, the geographical location is also an important factor in the referral decision. According to the patient's geographical location and the hospital's geographical location, the geographical coordinate distance between the two is calculated or the geographical location similarity measurement method is used to determine the second feature item. This step helps to screen out hospitals that are located near the patient and convenient for transfer, thereby improving the efficiency and success rate of referral.

[0135] S25. Determine a third feature item based on at least one of the rating, patient evaluation, and professional certification in the hospital's medical expertise information.

[0136] In addition to the characteristic items that are directly related to patient needs and hospital capabilities, the overall quality of the hospital is also an important factor that needs to be considered in the recommendation model. By collecting hospital rating information (such as star rating, professional ranking, etc.), patient evaluation (such as satisfaction surveys, online evaluations, etc.), and professional certification (such as international certification, industry qualifications, etc.), the comprehensive strength and service level of the hospital can be comprehensively evaluated. This information will be included in the calculation of the recommendation model as the third characteristic item.

[0137] S26. Substitute the weight coefficient and the first characteristic item, the second characteristic item, and the third characteristic item into the pre-built recommendation model to calculate the matching degree between the patient and each hospital. The matching degree here is a result of comprehensive consideration, which not only reflects the professional ability of the hospital in treating the patient's disease, but also takes into account the convenience of the geographical location and the comprehensive strength of the hospital.

[0138] S27. Sort the hospitals according to the matching degree to generate a list of recommended hospitals that can receive the corresponding symptoms and determine the transfer-to hospital and referral time window in the recommended hospital list.

[0139] The hospitals are sorted according to the calculated matching degree to generate a list of recommended hospitals. This list is arranged from high to low matching degree, so that the hospitals at the top are more likely to become suitable transfer hospitals. At the same time, a reasonable referral time window is determined based on factors such as the patient's condition, transfer conditions, and the hospital's available resources. This time window will serve as an important basis for referral decisions to ensure that patients can be transferred to the appropriate hospital for treatment at the best time.

[0140] In the above steps, the patients to be referred are determined based on their condition, geographic location, hospital expertise, etc. These factors can be used as features to build a recommendation model. Assume the following features:

[0141] Patient condition characteristics (P): describes the severity of the patient's condition, type of treatment required, etc.

[0142] Geographic location characteristics (L): describe the patient's location or desired referral area.

[0143] Hospital Specialty Characteristics (H): describes the hospital's expertise in specific diseases or treatments.

[0144] Based on these features, a formula for a personalized recommendation algorithm can be constructed to calculate the matching degree between patients and hospitals. A simple example formula is as follows:

[0145] The recommended models are:

[0146] M = α*P(patient, hospital) + β*L(patient, hospital) + γ*H(patient, hospital);

[0147] Wherein, M is the matching degree, α, β, and γ are the first weight coefficient, the second weight coefficient, and the third weight coefficient, respectively, P(patient, hospital) is the first feature item indicating the similarity between the patient's condition and the hospital's medical expertise information, L(patient, hospital) is the second feature item indicating the matching degree between the patient's geographical location and the distance or location of the hospital, and H(patient, hospital) is the third feature item indicating the hospital's expertise in the treatment required by the patient.

[0148] Once these characteristics and calculation methods are available, the patient's specific information and hospital data can be substituted into the formula to calculate the patient's match with each hospital. Then, the hospitals can be sorted according to the match, and the hospitals with the highest match can be recommended to the patient to ensure that the patient is referred to the most appropriate hospital. By comprehensively considering multiple factors such as the hospital's strength, expertise, and treatment experience, it is possible to provide patients with a high-quality range of hospitals to choose from, and select the most appropriate hospital for treatment based on the patient's actual situation and needs.

[0149] S3. Based on the receiving hospital and referral time window, combined with geographic information and real-time traffic data, calculate the best referral route and the shortest time required for referral within the referral time window.

[0150] Furthermore, if Figure 4 As shown, step S3 includes:

[0151] S31. Determine the referral time constraint according to the location information of the transferee hospital and a given referral time window.

[0152] This step mainly determines the time limit for when the ambulance must depart and arrive at the receiving hospital based on the specific location of the receiving hospital and the acceptable referral time range (i.e., referral time window) given by the medical institution or the patient's family. If the receiving hospital is far away and the referral time window is narrow, there will be stricter requirements for the departure and arrival times of the ambulance. For example, if the referral time window is from 8 am to 12 pm, then the referral must start before 8 am to ensure that the referral is completed before 12 pm.

[0153] S32. Evaluate the influencing factors of the current road conditions on the referral path selection based on the acquired real-time road condition data including road congestion (such as the number of waiting traffic lights, etc.), traffic flow, and accident information.

[0154] Obtain real-time traffic data through various sensors, traffic monitoring systems, or third-party services. Analyze this data to understand which sections of the road are currently congested, which sections have heavy traffic, and whether there are any accidents. This information is crucial for the subsequent selection of the best referral route, as it directly affects the speed and arrival time of the ambulance.

[0155] For each calculated path, the travel time is evaluated according to the real-time traffic data obtained in S32. For example, if there is a serious traffic jam on a certain path, the travel time of the ambulance on the path may be significantly increased.

[0156] S33. Based on the transferring-out hospital and the receiving hospital, a path planning algorithm is used to calculate multiple possible referral paths from the transferring-out hospital to the receiving hospital.

[0157] A path planning algorithm (such as annealing algorithm, Dijkstra algorithm, A* algorithm, etc.) is used to calculate multiple possible referral paths from the transferring hospital to the receiving hospital. Preferably, these paths are calculated based on factors such as road networks, traffic rules, and possible turning points.

[0158] S34. For multiple possible referral paths, the estimated driving time of the ambulance on each possible referral path is obtained according to the influencing factors of the current road conditions on the referral path selection.

[0159] S35. Filter out multiple paths that meet the referral time window requirement based on the referral time constraint and the estimated driving time of the ambulance on each possible referral path.

[0160] S36. Determine the best referral path from among multiple paths that meet the referral time window requirements based on acquired road condition factors including road condition stability information, total path length, and estimated driving time, and determine the shortest time required for referral within the referral time window along the best referral path.

[0161] Among the paths that meet the referral time window, other factors are further considered to determine the best path. Road stability information is an important factor, including: flatness, skid resistance, and road surface damage, because stable road conditions mean fewer delays and higher reliability. The total length of the path is also a consideration, because shorter paths usually mean less travel time and lower costs. Combining these factors, the best referral path and the shortest time required for referral within the referral time window are selected.

[0162] Furthermore, after step S3, it also includes: upon receiving the manual intervention information including the expected referral time, the ambulance dispatch is immediately triggered to ensure that the ambulance can be dispatched in time according to the predetermined referral time window and the best referral path. This trigger mechanism improves the flexibility and response speed of the entire referral process.

[0163] S4. Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. Within the available range of ambulances, an ambulance is automatically assigned to the patient according to the priority score.

[0164] Furthermore, if Figure 5 As shown, step S4 includes:

[0165] S41. Obtain the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral route, and the availability and status data of the ambulance.

[0166] S42. Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated by combining the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance.

[0167] S43. Automatically assign ambulances to patients based on the real-time location, number of ambulances, and the priority scores of patients to ensure that resource utilization is maximized and the available range of ambulances is not exceeded.

[0168] In the above steps, patients with the highest priority scores are given priority and assigned the ambulance with the best condition and the closest location. Such an allocation strategy is designed to minimize the waiting time of patients while maximizing the effective utilization of ambulances. During the allocation process, the status and location of ambulances are continuously monitored to ensure that each ambulance can operate within its available range and respond to new referral requests or emergencies in a timely manner. This dynamic ambulance allocation method can maintain efficiency and flexibility in changing situations, thereby maximizing the satisfaction of patients' emergency medical needs and ensuring that patients can be referred to appropriate medical institutions in a timely and efficient manner.

[0169] The following is a simplified example showing how to build and use a resource allocation algorithm.

[0170] In another specific embodiment, in order to realize how to allocate limited ambulance resources to multiple critically ill patients who need to be transferred, so as to maximize the overall treatment effect, the following key factors can be considered:

[0171] The patient's condition score (S): This can be measured based on the patient's vital signs, diagnosis results, etc. The more serious the condition, the higher the demand and priority for resources.

[0172] Distance or time from the patient to the destination hospital (D): A longer distance means more transportation time and resources are required.

[0173] Availability and status of ambulances (A): including the number, location, and availability of ambulances.

[0174] Based on these factors, a global optimization resource allocation model can be constructed to calculate the priority of allocating ambulances to each patient:

[0175] P = δ*S(patient) / D(patient, target hospital)+ε*A(ambulance);

[0176] Wherein, δ and ε are the first allocation ratio and the second allocation ratio respectively, S(patient) represents the patient's condition score, D(patient, target hospital) represents the shortest time and the best referral path required for the patient to be transferred to the transfer hospital within the referral time window, A(ambulance) represents the availability and status of the ambulance, and the availability and status of the ambulance include the number, location and whether the ambulance is idle; and the global optimization resource allocation model is provided with a feedback mechanism for dynamically adjusting the weight coefficient in the global optimization resource allocation model according to the actual treatment effect, so as to optimize the resource allocation algorithm.

[0177] In practical applications, a priority score can be calculated for each patient and sorted according to the score. Patients with high priority scores will be given priority in the allocation of ambulances. At the same time, the number and location of ambulances need to be considered to ensure that the available range of ambulances is not exceeded during the allocation process.

[0178] In a specific embodiment, in order to more specifically illustrate the application of the formula, there are the following data:

[0179] Patient A: condition score S = 8 (full score is 10), distance to the target hospital D = 30 kilometers, ambulance availability A = 1 (indicating available).

[0180] Patient B: condition score S=6, distance to target hospital D=10 km, ambulance availability A=1.

[0181] Patient C: condition score S=9, distance to target hospital D=50 km, ambulance availability A=1.

[0182] The priority score is calculated according to the formula:

[0183] Patient A: P = δ*8 / 30+ε*1;

[0184] Patient B: P = δ*6 / 10+ε*1;

[0185] Patient C: P = δ*9 / 50+ε*1;

[0186] Assuming that both δ and ε are set to the same value (for example, δ = ε = 0.5, indicating that the severity score and distance are equally important), the priority score of each patient can be calculated and sorted according to the score. In this example, patient C may get the highest priority because of the most serious condition, even though he is the farthest from the target hospital. However, the actual decision may also need to consider other factors, such as the location and status of other available ambulances, and whether there are other patients who need to be transferred at the same time.

[0187] At the same time, after step S4, after the ambulance dispatch is completed, a message notification is automatically triggered, and the relevant information is sent to the receiving hospital in the form of voice, text message, or WeChat, and an application form containing the referral patient information is submitted for review and handover.

[0188] refer to Figure 6 It can be seen that the critically ill inter-hospital referral patient scoring and resource allocation method of the present invention includes the following process:

[0189] (1) Evaluation and decision making: This part is implemented by steps S1-S3 of the present invention.

[0190] (2) Contacting the receiving hospital: After the decision to refer the patient is made, the emergency department or specialist department of the receiving hospital will be contacted to confirm the intention to accept the patient and prepare for the reception.

[0191] (3) Prepare referral information: organize the patient's medical records, examination reports, imaging data, etc. to ensure the completeness and accuracy of information during the referral process.

[0192] (4) Review acceptance conditions.

[0193] (5) Implement the rescue resource matching work of step S4 of the present invention.

[0194] (6) Arrange referral: Arrange ambulance or medical transport services according to the arrangements of the receiving hospital to ensure that the patient arrives at the receiving hospital safely and in a timely manner.

[0195] (7) Handover and Reception: After arriving at the receiving hospital, the patient’s condition is handed over to ensure that the receiving hospital understands the patient’s condition and important information during the referral process.

[0196] (8) Follow-up treatment: The receiving hospital will continue to provide necessary treatment and care based on the patient’s condition.

[0197] In addition, the present invention provides a critically ill inter-hospital referral patient scoring and resource allocation system, comprising:

[0198] The referral demand determination module is used to determine the patient's referral needs by using preset condition score calculation rules and / or machine learning-based classification algorithms.

[0199] The hospital recommendation module is used to build a recommendation model for patients who are determined to need referral, combining the patient's condition, geographical location and the hospital's medical expertise information to generate a list of recommended hospitals that can accept the corresponding symptoms, and determine the transfer-to hospital and referral time window in the recommended hospital list.

[0200] The path and time calculation module is used to calculate the optimal referral path and the shortest time required for referral within the referral time window based on the referral hospital and referral time window, combined with geographic information and real-time traffic data.

[0201] The ambulance allocation module is used to dynamically calculate the patient's priority score based on a pre-built global optimization resource allocation model, combining the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance, and automatically allocate an ambulance to the patient according to the priority score within the available range of the ambulance.

[0202] In a specific embodiment, if Figure 7 As shown, the present invention provides a critically ill inter-hospital referral patient scoring and resource allocation system, which also includes:

[0203] Real-time positioning unit: Obtain the latitude and longitude of the current location based on GPS to determine the current geographic location of the ambulance, etc., and provide real-time location updates, which is crucial for subsequent route planning and ambulance dispatch.

[0204] Model operation unit: A large amount of patient referral data and condition data are integrated internally. This module is involved in the implementation and operation of customized annealing algorithms, greedy algorithms, etc. to optimize resource allocation and referral decisions.

[0205] Resource matching unit: internally integrates the hospital's real-time bed data, in-hospital ambulance data, longitude and latitude data, reception capacity for special diseases, etc., to determine the order of recommended hospitals.

[0206] Decision-making unit: Integrate the calculated data of the model operation unit, patient scoring unit, and resource matching unit to make an overall judgment on whether the corresponding hospital matches the patient.

[0207] Path planning unit: calculates the best path for the outgoing hospital, incoming hospital, and the current location of the ambulance, and estimates the time and road conditions.

[0208] Patient scoring unit: After the transferor fills in the patient's basic examination data, the severity of the patient's condition is calculated based on the patient's condition, examination report, vital signs data, etc.

[0209] Ambulance dispatch unit: calculates the priority of ambulance dispatch based on the longitude and latitude of the receiving hospital and the transferring hospital, road conditions, patient conditions, ambulance utilization rate, etc.

[0210] Message notification unit: After the referral message is pushed out, the intelligent voice module, SMS function module, and WeChat public account module integrated in the message notification unit will push the referral data and review data to the transferor, transferee, and patient.

[0211] Furthermore, an embodiment of the present invention provides a critically ill inter-hospital referral patient scoring and resource allocation device, comprising: at least one database; and a memory communicatively connected to the at least one database; wherein the memory stores instructions executable by the at least one database, and the instructions are executed by the at least one database so that the at least one database can execute the critically ill inter-hospital referral patient scoring and resource allocation method as described above.

[0212] Furthermore, an embodiment of the present invention provides a computer-readable medium having computer-executable instructions stored thereon, and when the executable instructions are executed by a processor, the above-mentioned critically ill inter-hospital referral patient scoring and resource allocation method is implemented.

[0213] Since the system / device described in the above embodiments of the present invention is a system / device used to implement the method of the above embodiments of the present invention, a person skilled in the art can understand the specific structure and deformation of the system / device based on the method described in the above embodiments of the present invention, and thus will not be described in detail here. All systems / devices used in the method of the above embodiments of the present invention belong to the scope of protection of the present invention.

[0214] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0215] The present invention is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present invention. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions.

[0216] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0217] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0218] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0219] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for scoring and allocating resources for critically ill inter-hospital referral patients, characterized in that: include: Determine the patient's referral needs using preset condition score calculation rules and / or machine learning-based classification algorithms; For patients who are determined to need referral, a recommendation model is built based on the patient's condition, geographic location, and the hospital's medical expertise to generate a list of recommended hospitals that can accept the corresponding condition, and the transferee hospital and referral time window are determined in the recommended hospital list; According to the receiving hospital and referral time window, combined with geographic information and real-time traffic data, the best referral route and the shortest time required for referral within the referral time window are calculated; Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. Within the available range of ambulances, ambulances are automatically assigned to patients according to the priority score.

2. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to claim 1, characterized in that: Using preset disease severity score calculation rules and / or machine learning-based classification algorithms, the patient's referral needs are determined to include: Obtain the patient's disease type, vital signs, laboratory test results and required equipment information; The preset disease score calculation rules are used to output the disease score in combination with the patient's vital signs and laboratory tests; The pre-built logistic regression model is trained using the patient's disease type and vital signs, and the optimal coefficients are solved by minimizing the loss function; Substitute the patient's characteristic values ​​into the trained logistic regression model formula to calculate the probability of emergency referral, and output the referral judgment result based on the set threshold; Determine the patient's referral needs based on the disease score and / or referral judgment results, combined with the required device information; in, The preset calculation rules of the disease score include: vital signs scoring items and laboratory examination scoring items. For each indicator in the vital signs scoring items and laboratory examination scoring items, the score increases by 1 point every time a preset increase or decrease value unit is reached until the preset maximum score limit is reached. The total score of the disease score is obtained by the sum of the vital signs score and the laboratory examination score; The formula for the logistic regression model is: P(y=1|X)=1 / (1+exp(-(β0+β1x1+β2x2+...+β n *x n ))); Where P(y=1|X) represents the probability that a patient needs emergency referral given the patient’s characteristic X. The optimal coefficients include β0…β n , β0 is the intercept term, β1...β n are the coefficients corresponding to the characteristics of each patient, x1…x n are the characteristic values ​​of the patient, and exp is the natural exponential function.

3. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to claim 1, characterized in that: For patients who are determined to need referral, a recommendation model is built based on the patient's condition, geographic location, and the hospital's medical expertise to generate a list of recommended hospitals that can accept the corresponding symptoms. The referral hospital and referral time window are determined in the recommended hospital list, including: Obtain information on the patient's condition identified for referral, the patient's geographic location, the hospital's geographic location, and the hospital's medical expertise; According to the preset weight coefficients, the recommendation model is constructed by combining the patient's condition, geographical location and the hospital's medical expertise; Determine the first feature item by calculating the similarity between the patient's condition and the hospital's medical expertise information; Determine the second feature item by measuring the geographical location similarity between the patient and the hospital or calculating the geographical coordinate distance; determining a third feature item based on at least one of a rating, a patient evaluation, and a professional certification in the medical expertise information of the hospital; Substituting the weight coefficient and the first feature item, the second feature item, and the third feature item into the pre-built recommendation model, the matching degree between the patient and each hospital is calculated; Sort each hospital according to the matching degree to generate a list of recommended hospitals that can receive the corresponding disease and determine the transfer-in hospital and referral time window in the recommended hospital list; Among them, the recommended model is: M = α*P(patient, hospital) + β*L(patient, hospital) + γ*H(patient, hospital); Wherein, M is the matching degree, α, β, and γ are the first weight coefficient, the second weight coefficient, and the third weight coefficient, respectively, P (patient, hospital) is the first feature item, L (patient, hospital) is the second feature item, and H (patient, hospital) is the third feature item.

4. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to claim 1, characterized in that: According to the receiving hospital and referral time window, combined with geographic information and real-time traffic data, the best referral route and the shortest time required for referral within the referral time window are calculated, including: Determine the referral time constraint according to the location information of the receiving hospital and the given referral time window; Based on the real-time traffic data including road congestion, traffic flow, and accident information, evaluate the factors affecting the referral route selection; According to the transferring-out hospital and the receiving hospital, a path planning algorithm is used to calculate multiple possible referral paths from the transferring-out hospital to the receiving hospital; For multiple possible referral paths, the estimated driving time of the ambulance on each possible referral path is obtained according to the factors affecting the selection of the referral path under the current road conditions; According to the referral time constraint and the estimated driving time of the ambulance on each possible referral path, multiple paths that meet the referral time window requirements are screened out; From multiple paths that meet the referral time window requirements, the best referral path is determined based on road condition factors including road stability information, total path length, and estimated driving time, and the shortest time required for referral within the referral time window along the best referral path is determined.

5. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to claim 1, characterized in that: According to the receiving hospital and referral time window, combined with geographic information and real-time traffic data, the optimal referral route and the shortest time required for referral within the referral time window are calculated, and the following are also included: Upon receipt of manual intervention information including the expected referral time, ambulance dispatch is immediately triggered to ensure that the ambulance can depart according to the predetermined referral time window and the optimal referral route.

6. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to any one of claims 1 to 5, characterized in that: Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated by combining the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance. Within the available range of ambulances, the patient is automatically assigned an ambulance according to the priority score, including: Obtain data on the patient's acuity score, minimum time required for referral within the referral time window, optimal referral route, and ambulance availability and status; Based on a pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance; Automatically assign ambulances to patients based on their real-time location, number, and priority score to ensure that resources are maximized and ambulance availability is not exceeded; Among them, the global optimization resource allocation model is: P = δ*S(patient) / D(patient, target hospital)+ε*A(ambulance); Wherein, δ and ε are the first allocation ratio and the second allocation ratio respectively, S(patient) represents the patient's condition score, D(patient, target hospital) represents the shortest time and the best referral path required for the patient to be transferred to the transfer hospital within the referral time window, A(ambulance) represents the availability and status of the ambulance, and the availability and status of the ambulance include the number, location and whether the ambulance is idle; and the global optimization resource allocation model is provided with a feedback mechanism for dynamically adjusting the weight coefficient in the global optimization resource allocation model according to the actual treatment effect, so as to optimize the resource allocation algorithm.

7. The method for scoring and allocating resources for critically ill inter-hospital referral patients according to any one of claims 1 to 5, characterized in that: Based on the pre-built global optimization resource allocation model, the patient's priority score is dynamically calculated based on the patient's condition, the shortest time required for referral within the referral time window, the best referral path, and the availability and status of the ambulance. After the ambulance is automatically assigned to the patient according to the priority score within the available range of the ambulance, it also includes: After the ambulance dispatch is completed, a message notification is automatically triggered, and the relevant information is sent to the receiving hospital in the form of voice, text message, or WeChat, and an application form containing the referral patient information is submitted for review and handover.

8. A critically ill inter-hospital referral patient scoring and resource allocation system, characterized in that: include: A referral demand determination module, used to determine the patient's referral demand by using a preset condition score calculation rule and / or a classification algorithm based on machine learning; The hospital recommendation module is used to build a recommendation model for patients who need to be referred, combining the patient's condition, geographical location and the hospital's medical expertise information to generate a list of recommended hospitals that can accept the corresponding symptoms, and determine the transfer-in hospital and referral time window in the recommended hospital list; The path and time calculation module is used to calculate the best referral path and the shortest time required for referral within the referral time window based on the referral hospital and referral time window, combined with geographic information and real-time traffic data; The ambulance allocation module is used to dynamically calculate the patient's priority score based on a pre-built global optimization resource allocation model, combining the patient's condition score, the shortest time required for referral within the referral time window, the optimal referral path, and the availability and status of the ambulance, and automatically allocate an ambulance to the patient according to the priority score within the available range of the ambulance.

9. A critically ill inter-hospital referral patient scoring and resource allocation device, characterized in that: include: at least one database; And a memory communicatively connected to the at least one database; wherein the memory stores instructions executable by the at least one database, and the instructions are executed by the at least one database so that the at least one database can execute the critically ill inter-hospital referral patient scoring and resource allocation method as described in any one of claims 1-7.

10. A computer-readable medium having computer-executable instructions stored thereon, characterized in that: When the executable instructions are executed by the processor, the critically ill inter-hospital referral patient scoring and resource allocation method as described in any one of claims 1-7 is implemented.

Images