Medical resource allocation method based on case type data analysis

Abstract

The invention relates to the technical field of medical resource allocation, and discloses a medical resource allocation method based on case type data analysis, and the method comprises the steps: collecting the case type index characteristics of a patient in a decision tree mode, and constructing the case type of the patient; performing priority scoring on the case types by adopting a dynamic priority scoring mode; constructing a medical resource allocation objective function and a medical resource allocation constraint item; and solving the medical resource allocation objective function to obtain the medical resource allocation quantity of the patient. According to the invention, quantitative calculation of disease time decay, disease severity and collateral circulation degree is carried out according to the disease duration of the patient, the routine examination score and the imaging examination result, and the calculation result is used as the medical priority to provide instant support for medical resource allocation. And an objective function is constructed and solved by maximizing the matching degree between the patient and the medical resource allocation quantity, so that an optimal allocation mode under limited resources is obtained, and the medical decision-making efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of medical resource allocation, and in particular to a medical resource allocation method based on case type data analysis. Background Art

[0002] At present, the global medical system is generally facing a structural shortage of acute critical care resources. Taking acute ischemic stroke as an example, the time window for intravenous thrombolytic therapy is only 4.5 hours, and the time window for intravascular thrombectomy is extended to 24 hours, but a professional interventional team is required. There are significant differences in the treatment priorities of different stroke cases: patients with large vessel occlusions with NIHSS scores ≥ 6 points need to be given priority for angiography; patients with atrial fibrillation have special needs for new oral anticoagulants; patients with stroke after awakening need MRI-DWI / FLAIR mismatch assessment, etc. The existing "first come, first served" allocation model causes 21.3% of high-risk patients to miss the golden treatment time. Summary of the invention

[0003] In view of this, the present invention provides a medical resource allocation method based on case type data analysis, which performs dynamic priority evaluation according to patient case types to achieve optimal medical resource allocation and improve golden time treatment rate.

[0004] To achieve the above object, the present invention provides a medical resource allocation method based on case type data analysis, comprising the following steps: S1: Construct a case type indicator feature system, collect the patient's case type indicator features using a decision tree, and construct the patient's case type.

[0005] S2: Use dynamic priority scoring to prioritize case types and use the priority scoring results as the patient’s medical priority.

[0006] S3: Construct the medical resource allocation objective function and medical resource allocation constraints based on the patient's medical priority and case type.

[0007] S4: Solve the medical resource allocation objective function under the constraints of the medical resource allocation constraint item, obtain the patient's medical resource allocation amount, and perform medical resource allocation.

[0008] As a further improvement method of the present invention: Optionally, the case type indicators in the case type indicator characteristic system include the patient's onset duration, routine examination indicator scores and imaging examination results; the routine examination indicator score is the patient's clinical assessment score during the routine examination, wherein the higher the routine examination indicator score, the more serious the patient's condition.

[0009] Optionally, the process of collecting the case type indicator characteristics by using a decision tree method is as follows: The duration of the patient's onset is collected, and routine examinations are performed on the N patients currently receiving treatment to obtain the patient's routine examination score. The patient's routine examination score is compared with the preset threshold score to obtain the patient's symptom severity. An imaging examination is performed based on the patient's symptom severity to obtain the patient's imaging examination result.

[0010] The patient's case type indicator feature is constructed as the patient's case type, and the case type of the nth patient is : .

[0011] in: represents the duration of onset of the nth patient, represents the score of the routine examination index of the nth patient, represents the severity of the symptoms of the nth patient, , 1 means the nth patient is in a critical condition. 0 means that the nth patient is in a mild condition. represents the imaging result of the nth patient, .

[0012] Optionally, the medical priority of the nth patient is calculated as follows: From the case type Extract the imaging examination result coefficient of the nth patient from and imaging findings , generating the degree of collateral circulation of the nth patient .

[0013] From the case type Extract the duration of onset of the nth patient and routine examination scores , combined with the degree of collateral circulation of the nth patient , use dynamic priority scoring method to score the priority and get the medical priority of the nth patient : ; ; ; ; .

[0014] in: represents the time decay term of the nth patient’s condition, represents the severity of the condition of the nth patient, represents the treatment gain of the nth patient, Indicates the preset control factor; Indicates the preset time period parameters. represents the attenuation control coefficient, Indicates duration of illness Dynamic time decay.

[0015] Optionally, medical resources required by the patient are generated according to the disease type, and the medical resources include interventional treatment, intensive drug treatment, intravenous thrombolytic therapy, and common drug treatment, wherein the process of generating the medical resources required by the nth patient is: like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resources required for the nth patient are interventional treatment. When the cycle reaches or exceeds the preset threshold, the medical resources required for the nth patient are intensive drug treatment.

[0016] like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resource required for the nth patient is intravenous thrombolysis. When the cycle degree reaches or exceeds the preset threshold, the medical resources required for the nth patient are ordinary drug treatment.

[0017] The medical resources required by the nth patient are .

[0018] Optionally, a medical resource allocation objective function is constructed in combination with the medical resources required by the patient and the medical priority. The expression of the constructed medical resource allocation objective function is: .

[0019] in: represents the objective function of medical resource allocation, , represents the sequence of medical resource allocation for N patients to be solved, represents the amount of medical resources allocated to the nth patient, Indicates the amount of medical resources allocated of medical costs; Indicates the amount of medical resources allocated to the nth patient The medical resources required for the nth patient The difference between represents the L2 norm, where The smaller it is, the higher the degree of match between the medical resources required by the nth patient and the amount of medical resources allocated.

[0020] Optionally, the expression of the medical resource allocation constraint item is: .

[0021] in: The hospital can conduct the first The maximum number of patients for each treatment modality, , where the 1st to 4th treatment methods are interventional therapy, intensive drug therapy, intravenous thrombolytic therapy, and conventional drug therapy; Represents the sequence of medical resource allocation Middle The number of times a treatment method occurs.

[0022] Optionally, the medical resource allocation objective function is solved under the constraint of the medical resource allocation constraint term, wherein the solution process is: The medical resource allocation constraint item is used as the penalty item of the medical resource allocation objective function, and the medical resource allocation objective function is subjected to penalty constraint: ; .

[0023] in: represents the penalty term based on the medical resource allocation constraint term, represents the penalty coefficient; Represents the objective function of medical resource allocation after penalty constraint.

[0024] Initialize and generate a set of medical resource allocation sequences that meet the constraints , set the momentum , learning rate and momentum coefficient , set the maximum number of iterations Max.

[0025] Combined with the gradient of the medical resource allocation amount sequence in the medical resource allocation objective function after the penalty constraint, the momentum, momentum coefficient and learning rate are iterated, and the medical resource allocation amount sequence is iterated based on the iterative result until the maximum number of iterations is reached. The medical resource allocation amount sequence obtained by the final iteration is used as the solution result, and the corresponding medical resources are allocated to the patient based on the treatment method represented by the medical resource allocation amount.

[0026] In order to solve the above problem, the present invention provides an electronic device, the electronic device comprising: A memory storing at least one instruction; Communication interface, enabling electronic equipment to communicate; and A processor executes instructions stored in the memory to implement the above-mentioned medical resource allocation method based on case type data analysis.

[0027] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned medical resource allocation method based on case type data analysis.

[0028] Compared with the prior art, the present invention proposes a medical resource allocation method based on case type data analysis, which has the following advantages: First, this scheme proposes a medical priority evaluation method combined with a decision tree. CT vascular imaging and magnetic resonance imaging decisions are made according to the patient's routine examination results to avoid wasting medical imaging examination resources, and then quantify the medical priority. The longer the onset time, the higher the time decay item of the disease. The higher the routine examination score, the higher the severity of the disease. The higher the degree of collateral circulation, the better the blood supply, the higher the probability of successful treatment and the higher the recovery status of treatment. The higher the three indicators, the higher the medical priority, and the more urgent the treatment of critically ill patients.

[0029] At the same time, this scheme proposes a medical resource allocation method, which generates the medical resources required by patients according to the type of disease, and constructs a medical resource allocation objective function based on the medical resources required by patients and medical priorities. In the process of solving the objective function, the nonlinear adjustment of the learning rate and momentum coefficient is used to improve the smoothness of the iterative optimization process, avoiding excessive updates caused by a large learning rate or momentum in the early stage of iteration. The learning rate and momentum are dynamically adjusted through gradient information, and the algorithm can respond differently in different optimization stages. For example, when the gradient is large, the algorithm will appropriately increase the learning rate to speed up the convergence speed; when the gradient becomes smaller and close to the optimal solution, the learning rate will decrease to avoid skipping the optimal solution, and the nonlinear adjustment of the momentum coefficient helps to avoid excessive oscillations near the local optimal solution, thereby accelerating the finding of the global optimal solution. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 A flowchart of a medical resource allocation method based on case type data analysis provided by one embodiment of the present invention.

[0031] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0032] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0033] The embodiment of the present application provides a medical resource allocation method based on case type data analysis. The execution subject of the medical resource allocation method based on case type data analysis includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided by the embodiment of the present application. In other words, the medical resource allocation method based on case type data analysis can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0034] Reference Figure 1 , Embodiment 1 of the present invention is: A method for allocating medical resources based on case type data analysis includes the following steps: S1: Construct a case type indicator feature system, collect the patient's case type indicator features using a decision tree, and construct the patient's case type.

[0035] The case type indicators in the case type indicator characteristic system include the patient's onset duration, routine examination indicator scores, and imaging examination results.

[0036] The routine examination index score is the clinical evaluation score of the patient during the routine examination process, wherein the higher the routine examination index score, the more serious the patient's condition is; as an embodiment of the present invention, the present invention adopts the NIHSS scoring method for clinical evaluation, and the NIHSS scoring dimensions include the patient's level of consciousness, vision, facial movement, limb movement, coordination and language, etc., and the range of the NIHSS scoring results is 0-42 points.

[0037] Specifically, the imaging examination results include CT vascular imaging examination results or magnetic resonance imaging examination results, wherein the magnetic resonance imaging examination results include tissue signal intensity and tissue boundary clarity, and the CT vascular imaging examination results include the number of vascular anastomoses and the degree of vascular occlusion.

[0038] As an embodiment of the present invention, the DeepASITN algorithm is used to analyze CT vascular imaging. The DeepASITN algorithm includes feature extraction and spatiotemporal interaction modeling to obtain the degree of vascular occlusion in CT vascular imaging, identify the vascular network, and detect the number of small blood vessels and anastomotic branches connected to the main blood vessels (such as the carotid artery or the cerebral artery) as the number of vascular anastomotic branches.

[0039] The process of collecting the case type indicator characteristics using the decision tree method is as follows: The duration of the patient's onset is collected, routine examinations are performed on N patients currently receiving treatment to obtain the patient's routine examination scores, the patient's routine examination scores are compared with the preset threshold scores to obtain the patient's symptom severity, and an imaging examination is performed based on the patient's symptom severity to obtain the patient's imaging examination results; specifically, if the patient's routine examination score is lower than the preset threshold score, a magnetic resonance imaging examination is performed on the patient to obtain the patient's magnetic resonance imaging results; if the patient's routine examination score reaches or exceeds the preset threshold score, a magnetic resonance imaging examination is performed on the patient to determine the patient's lesion location, and a CT vascular imaging examination is performed on the lesion location to obtain the patient's CT vascular imaging results.

[0040] The patient's case type indicator feature is constructed as the patient's case type, and the case type of the nth patient is : .

[0041] in: represents the duration of onset of the nth patient, represents the score of the routine examination index of the nth patient, represents the severity of the symptoms of the nth patient, , 1 means the nth patient is in a critical condition. 0 means that the nth patient is in a mild condition. represents the imaging result of the nth patient, .

[0042] As a preferred embodiment of the present invention, if ,but , represents the CT angiography examination result of the nth patient, , represents the number of vascular anastomoses of the nth patient, represents the degree of vascular occlusion of the nth patient, Indicates the preset threshold score.

[0043] like ,but , represents the MRI result of the nth patient, , represents the tissue signal intensity of the nth patient, Indicates the clarity of the tissue boundary of the nth patient.

[0044] As a preferred embodiment of the present invention, if N is greater than 8, the preset threshold score is reduced by 2.

[0045] S2: Use dynamic priority scoring to prioritize case types and use the priority scoring results as the patient’s medical priority.

[0046] The medical priority of the nth patient is calculated as follows: From the case type Extract the imaging examination result coefficient of the nth patient from and imaging findings , generating the degree of collateral circulation of the nth patient Specifically, the degree of collateral circulation The generation formula is: .

[0047] From the case type Extract the duration of onset of the nth patient and routine examination scores , combined with the degree of collateral circulation of the nth patient , use dynamic priority scoring method to score the priority and get the medical priority of the nth patient : ; ; ; ; .

[0048] in: represents the time decay term of the nth patient’s condition, represents the severity of the condition of the nth patient, represents the treatment gain of the nth patient, Indicates the preset control factor; Indicates the preset time period parameters. represents the attenuation control coefficient, Indicates duration of illness Specifically, the longer the onset time is, the higher the disease time decay item is, the higher the conventional examination score is, the higher the severity of the disease is, the higher the degree of collateral circulation is, the better the blood supply is, and the higher the probability of successful treatment and treatment recovery are.

[0049] S3: Construct the medical resource allocation objective function and medical resource allocation constraints based on the patient's medical priority and case type.

[0050] The medical resources required by the patient are generated according to the type of illness, and the medical resources include interventional treatment, intensive drug treatment, intravenous thrombolytic therapy and common drug treatment. The process of generating the medical resources required by the nth patient is as follows: like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resources required for the nth patient are interventional treatment. When the cycle reaches or exceeds the preset threshold, the medical resources required for the nth patient are intensive drug treatment.

[0051] like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resource required for the nth patient is intravenous thrombolysis. When the cycle degree reaches or exceeds the preset threshold, the medical resources required for the nth patient are ordinary drug treatment.

[0052] The medical resources required by the nth patient are .

[0053] The objective function of medical resource allocation is constructed by combining the medical resources required by patients and their medical priorities. The expression of the constructed objective function of medical resource allocation is: .

[0054] in: represents the objective function of medical resource allocation, , represents the sequence of medical resource allocation for N patients to be solved, represents the amount of medical resources allocated to the nth patient, Indicates the amount of medical resources allocated In a preferred embodiment of the present invention, the allocation of different medical resources is calculated by constructing a hospital specialty marginal cost matrix decision model The medical cost of the hospital specialty marginal cost matrix decision model is in matrix form, the behavior of the matrix is ​​the specialty of the hospital, including internal medicine, surgery, pediatrics, obstetrics and gynecology, etc., the columns of the matrix are the marginal costs of different types of medical resource allocation, the marginal costs are used as the medical costs of the medical resource allocation, and the medical resource allocation objective function after the penalty constraint is solved. The types of medical resource allocation include interventional treatment, intensive drug treatment, intravenous thrombolytic therapy, ordinary drug treatment and waiting treatment, where waiting treatment means not allocating medical resources to patients; Indicates the amount of medical resources allocated to the nth patient The medical resources required for the nth patient The difference between represents the L2 norm, where The smaller it is, the higher the degree of match between the medical resources required by the nth patient and the amount of medical resources allocated.

[0055] The amount of medical resource allocation is any one of interventional therapy, intensive drug therapy, intravenous thrombolytic therapy, common drug therapy, and waiting treatment. Specifically, in the process of difference calculation and objective function solution, the numerical representations of interventional therapy, intensive drug therapy, intravenous thrombolytic therapy, common drug therapy, and waiting treatment are 10, 7, 5, 3, and 1, respectively.

[0056] The expression of the medical resource allocation constraint term is: .

[0057] in: The hospital can conduct the first The maximum number of patients for each treatment modality, , where the 1st to 4th treatment methods are interventional therapy, intensive drug therapy, intravenous thrombolytic therapy, and conventional drug therapy; Represents the sequence of medical resource allocation Middle The number of times a treatment method occurs.

[0058] S4: Solve the medical resource allocation objective function under the constraints of the medical resource allocation constraint item, obtain the patient's medical resource allocation amount, and perform medical resource allocation.

[0059] Under the constraints of the medical resource allocation constraint term, the medical resource allocation objective function is solved, wherein the solution process is: The medical resource allocation constraint item is used as the penalty item of the medical resource allocation objective function, and the medical resource allocation objective function is subjected to penalty constraint: ; .

[0060] in: represents the penalty term based on the medical resource allocation constraint term, represents the penalty coefficient; Represents the objective function of medical resource allocation after penalty constraint.

[0061] Initialize and generate a set of medical resource allocation sequences that meet the constraints , set the momentum , learning rate and momentum coefficient , set the maximum number of iterations Max, then the sequence of medical resource allocation obtained in the dth iteration is , the learning rate is , the momentum coefficient is , the momentum is .

[0062] The momentum, momentum coefficient and learning rate are iterated in combination with the gradient of the medical resource allocation amount sequence in the medical resource allocation objective function after the penalty constraint, and the medical resource allocation amount sequence is iterated based on the iterative result until the maximum number of iterations is reached. The medical resource allocation amount sequence obtained by the final iteration is used as the solution result, and the corresponding medical resources are allocated to the patient based on the treatment method represented by the medical resource allocation amount.

[0063] The iterative formula of the momentum coefficient is: .

[0064] in: represents the preset maximum momentum coefficient, represents the momentum-adjusted index, represents the gradient influence coefficient, Represents the sequence of medical resource allocation The gradient of the objective function of medical resource allocation after penalty constraint, represents the L1 norm; as a preferred embodiment of the present invention, The momentum coefficient is controlled to increase from the initial value to the maximum value. In the early stage, the momentum coefficient is small, which helps to avoid over-reliance on historical gradients. As the number of iterations increases, the momentum gradually increases to accelerate convergence.

[0065] The momentum coefficient is dynamically adjusted according to the size of the current gradient. If the gradient is large, it means that the solution may be far away from the optimal point and a smaller momentum is needed to avoid excessive updates. When the gradient is small, the momentum coefficient can be appropriately increased to accelerate the approach to the optimal solution.

[0066] The iterative formula for the momentum is: .

[0067] The iterative formula of the learning rate is: .

[0068] The iterative formula of the medical resource allocation sequence is: .

[0069] As an embodiment of the present invention, when receiving 3 patients at the same time: ① 2 hours after onset, routine examination score 8 points, basilar artery occlusion; ② 1 hour after onset, routine examination score 14 points, internal carotid artery occlusion; ③ stroke after awakening, routine examination score 6 points, FLAIR negative. The system will give priority to assigning the interventional team to case ②, arranging case ① to undergo intravenous thrombolysis with tenecteplase, and case ③ to enter the MRI evaluation queue.

[0070] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0071] It should be noted that the serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. And the terms "including", "comprising" or any other variants thereof in this article are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "including a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0072] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0073] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A medical resource allocation method based on case type data analysis, characterized in that: The method comprises: S1: Construct a case type indicator feature system, collect the patient's case type indicator features using a decision tree method, and construct the patient's case type. The case type is the patient's multi-layer clinical decision information, including the patient's onset time, routine examination score, imaging examination result coefficient, and imaging examination result; The case type indicator characteristic system includes multiple case type indicators. The case type indicator characteristic is the indicator value of the patient under the case type indicator. The indicator values ​​of the patient under all case type indicators are constructed as the case type of the patient. S2: Use dynamic priority scoring to prioritize case types and use the priority scoring results as the patient's medical priority; S3: constructing a medical resource allocation objective function and medical resource allocation constraint items based on the patient's medical priority and case type, wherein the medical resource allocation objective function uses the medical resource allocation amount as a variable and aims to maximize the matching degree between the patient and the medical resource allocation amount; S4: Solve the medical resource allocation objective function under the constraints of the medical resource allocation constraint item, obtain the patient's medical resource allocation amount, and perform medical resource allocation.

2. A medical resource allocation method based on case type data analysis as claimed in claim 1, characterized in that: The case type indicators in the case type indicator characteristic system include the patient's onset duration, routine examination indicator scores and imaging examination results; the routine examination indicator score is the patient's clinical assessment score during the routine examination process, wherein the higher the routine examination indicator score, the more serious the patient's condition.

3. A medical resource allocation method based on case type data analysis as claimed in claim 2, characterized in that: The process of collecting the characteristic indicators of the case type using the decision tree method is as follows: Collect the duration of the patient's onset, perform routine examinations on the N patients currently receiving treatment, obtain the patient's routine examination score, make a decision comparison between the patient's routine examination score and the preset threshold score to obtain the patient's symptom severity, perform imaging examinations based on the patient's symptom severity, and obtain the patient's imaging examination results; The patient's case type indicator feature is constructed as the patient's case type, and the case type of the nth patient is : ； in: represents the duration of onset of the nth patient, represents the score of the routine examination index of the nth patient, represents the severity of the symptoms of the nth patient, , 1 means the nth patient is in a critical condition. 0 means that the nth patient is in a mild condition. represents the imaging result of the nth patient, .

4. A medical resource allocation method based on case type data analysis as claimed in claim 3, characterized in that: The medical priority of the nth patient is calculated as follows: From the case type Extract the imaging examination result coefficient of the nth patient from and imaging findings , generating the degree of collateral circulation of the nth patient ; From the case type Extract the duration of onset of the nth patient and routine examination scores , combined with the degree of collateral circulation of the nth patient , use dynamic priority scoring method to score the priority and get the medical priority of the nth patient : ； ； ； ； ； in: represents the time decay term of the nth patient’s condition, represents the severity of the condition of the nth patient, represents the treatment gain of the nth patient, Indicates the preset control factor; Indicates the preset time period parameters. represents the attenuation control coefficient, Indicates duration of illness Dynamic time decay.

5. A medical resource allocation method based on case type data analysis as claimed in claim 4, characterized in that: The medical resources required by the patient are generated according to the type of illness, and the medical resources include interventional treatment, intensive drug treatment, intravenous thrombolytic therapy and common drug treatment. The process of generating the medical resources required by the nth patient is as follows: like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resources required for the nth patient are interventional treatment. If the preset cycle degree threshold is reached or exceeded, the medical resources required for the nth patient are intensive drug treatment; like , the degree of collateral circulation of patients To judge, when the degree of collateral circulation If the degree of collateral circulation is lower than the preset threshold, the medical resource required for the nth patient is intravenous thrombolysis. If the preset cycle degree threshold is reached or exceeded, the medical resources required for the nth patient are ordinary drug treatment; The medical resources required by the nth patient are .

6. A medical resource allocation method based on case type data analysis as claimed in claim 1, characterized in that: The objective function of medical resource allocation is constructed by combining the medical resources required by patients and their medical priorities. The expression of the constructed objective function of medical resource allocation is: ； in: represents the objective function of medical resource allocation, , represents the sequence of medical resource allocation for N patients to be solved, represents the amount of medical resources allocated to the nth patient, Indicates the amount of medical resources allocated of medical costs; Indicates the amount of medical resources allocated to the nth patient The medical resources required for the nth patient The difference between represents the L2 norm, where The smaller it is, the higher the degree of match between the medical resources required by the nth patient and the amount of medical resources allocated.

7. A method for allocating medical resources based on case type data analysis as claimed in claim 6, characterized in that: The expression of the medical resource allocation constraint term is: ； in: The hospital can conduct the first The maximum number of patients for each treatment modality, , where the 1st to 4th treatment methods are interventional therapy, intensive drug therapy, intravenous thrombolytic therapy, and conventional drug therapy; Represents the sequence of medical resource allocation Middle The number of times a treatment method occurs.

8. A method for allocating medical resources based on case type data analysis as claimed in claim 7, characterized in that: Under the constraints of the medical resource allocation constraint term, the medical resource allocation objective function is solved, wherein the solution process is: The medical resource allocation constraint item is used as the penalty item of the medical resource allocation objective function, and the medical resource allocation objective function is subjected to penalty constraint: ； ； in: represents the penalty term based on the medical resource allocation constraint term, represents the penalty coefficient; represents the objective function of medical resource allocation after penalty constraint; Initialize and generate a set of medical resource allocation sequences that meet the constraints , set the momentum , learning rate and momentum coefficient , set the maximum number of iterations Max; Combined with the gradient of the medical resource allocation amount sequence in the medical resource allocation objective function after the penalty constraint, the momentum, momentum coefficient and learning rate are iterated, and the medical resource allocation amount sequence is iterated based on the iterative result until the maximum number of iterations is reached. The medical resource allocation amount sequence obtained by the final iteration is used as the solution result, and the corresponding medical resources are allocated to the patient based on the treatment method represented by the medical resource allocation amount.

Images