Intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources

The intelligent scheduling platform for neurosurgical rehabilitation resources dynamically matches patient status with resource status, solving the problems of resource idleness and patient waiting in the existing scheduling methods, and achieving efficient resource utilization and ensuring rehabilitation effects.

CN123120239APending Publication Date: 2026-11-17HUNAN MEDICAL TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611627552.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-21
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

The current methods of allocating neurosurgical rehabilitation resources fail to fully consider the dynamic changes in the rehabilitation process, resulting in a contradiction between idle resources and patients waiting for treatment, which reduces the efficiency of resource utilization and affects rehabilitation outcomes and experience.

Method used

A smart scheduling platform for neurosurgical rehabilitation resources is provided, including a historical data analysis unit, a real-time patient assessment unit, and a resource assessment unit. By analyzing historical and real-time data, a scheduling recommendation plan is generated, and the patient status and resource status are dynamically matched to cope with dynamic disturbances in the rehabilitation process.

Benefits of technology

It has improved the efficiency of management and scheduling of neurosurgical rehabilitation resources, enhanced the flexibility and reliability of the scheduling system, ensured the continuity of patient rehabilitation, and improved the efficiency of resource utilization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123120239A_ABST
    Figure CN123120239A_ABST
Patent Text Reader

Abstract

This invention relates to the field of resource management and scheduling technology, specifically to an intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources, solving the technical problem of low efficiency in existing management and scheduling technologies. The platform includes: a historical data analysis unit, used to divide the patient's rehabilitation timeline into multiple stages of rehabilitation symptom manifestation based on historical rehabilitation data, and to determine the potential delay in the rehabilitation training plan for each stage; a real-time patient assessment unit, used to determine the controllability of each patient's sudden symptoms; a resource assessment unit, used to determine the relative busyness of each available time slot for each rehabilitation training mode based on historical resource occupancy data and personnel conflict data; and a scheduling decision unit, used to generate a scheduling recommendation scheme based on the controllability of each patient's sudden symptoms and the relative busyness of each available time slot for each rehabilitation training mode to be performed by each patient.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of resource management and scheduling technology, specifically to an intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources. Background Technology

[0002] Neurosurgical rehabilitation primarily serves patients in the postoperative recovery period or those with impaired neurological function. Their rehabilitation process is significantly complex and varies greatly from person to person. The severity of their condition, surgical procedures, and underlying physical conditions differ among patients, leading to diverse rehabilitation pathways and a non-linear recovery process. Sudden changes in condition and complications may occur during rehabilitation, impacting the implementation of the rehabilitation plan. Furthermore, the supply of rehabilitation resources is constrained by factors such as the availability of medical equipment, the professional competence and scheduling of therapists, and the allocation of nursing staff. The dynamic changes in these factors further increase the difficulty of allocating rehabilitation resources.

[0003] Current methods of allocating neurosurgical rehabilitation resources primarily rely on fixed rehabilitation pathway assumptions and static resource availability predictions, representing a rigid planning model. This approach fails to adequately consider dynamic changes during the rehabilitation process and lacks flexible adjustment mechanisms to address unforeseen circumstances. When disturbances such as sudden changes in patient symptoms, equipment malfunctions, or temporary absences of therapists occur, the original scheduling plan easily falls into disarray, hindering rapid resource reallocation and plan adjustments. This not only leads to the coexistence of idle rehabilitation resources and patients waiting for treatment, reducing resource utilization efficiency, but also disrupts the continuity and stability of the rehabilitation process, impacting patient rehabilitation outcomes and experience. Summary of the Invention

[0004] To address the problem of low efficiency in existing management and scheduling technologies, the present invention aims to provide an intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources. The specific technical solution adopted is as follows: This application provides an intelligent scheduling platform for neurosurgical rehabilitation resources, including: The historical data analysis unit is used to divide the patient's rehabilitation timeline into multiple rehabilitation symptom manifestation stages based on historical rehabilitation data, and to determine the potential delay in the rehabilitation training plan for each rehabilitation symptom manifestation stage. The real-time patient assessment unit is used to perform symptom control analysis based on the current symptom data of each patient and the potential delay in the rehabilitation training plan of each patient's rehabilitation symptom stage, and to determine the control degree of each patient's sudden symptom. The control degree of sudden symptom is used to characterize the degree of consistency between the patient's symptom performance and the typical symptom characteristics of the rehabilitation symptom stage experienced. The resource assessment unit is used to determine the relative busy performance of each optional time period for each rehabilitation training mode based on historical resource usage data and personnel conflict data; the relative busy performance is used to characterize the resource strain of performing the corresponding rehabilitation training mode within the optional time period. The scheduling decision unit is used to generate a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busyness of each available time slot for each patient's rehabilitation training mode; the scheduling recommendation plan is used to characterize the available time slots for each patient's rehabilitation training mode.

[0005] In one possible implementation, the historical data analysis unit includes: The postponement performance calculation module is used to calculate the postponement performance of rehabilitation training for each unit period in the rehabilitation timeline based on the number of occurrences of various sudden symptoms within a unit period in historical rehabilitation data and the corresponding proportion of changes to the rehabilitation training plan. The phase division module is used to divide the rehabilitation timeline into multiple time clusters based on the performance of rehabilitation training suspension and a clustering algorithm. The potential delay determination module defines a continuous unit time period contained in each time cluster as a rehabilitation symptom performance stage, and determines the potential delay of the rehabilitation training plan for each rehabilitation symptom performance stage based on the rehabilitation training postponement performance of all unit time periods within each rehabilitation symptom performance stage.

[0006] In one possible implementation, the performance calculation module is shelved, specifically for: The frequency of occurrence of various sudden symptoms in each unit period of the historical patient's recovery timeline was counted, and the symptom weight of each sudden symptom was determined based on the frequency of occurrence. The number of times the rehabilitation training plan was modified due to each type of sudden symptom in each unit period of the historical patients' rehabilitation timeline was counted, and the proportion of rehabilitation training plan modifications for each type of sudden symptom was determined based on the number of modifications. The performance degree of rehabilitation training suspension is calculated based on the symptom weights and the corresponding proportions of changes to the rehabilitation training plan.

[0007] In one possible implementation, the phase division module is specifically used for: The characteristic distance between two unit periods is determined based on the degree of rehabilitation training suspension performance at any two unit periods on the rehabilitation timeline. Based on feature distance, clustering algorithms are used to perform cluster analysis on all unit time periods on the rehabilitation timeline to form multiple time clusters.

[0008] In one possible implementation, the real-time patient assessment unit includes: The stage positioning module is used to determine the stage of rehabilitation symptoms experienced by each patient from multiple stages of rehabilitation symptoms based on the current rehabilitation time point of each patient. The typical symptom set construction module is used to divide the sudden symptoms that appear in each stage of rehabilitation symptom manifestation into a core symptom set and a general symptom set based on the records of sudden symptoms that led to changes in the rehabilitation training plan in historical rehabilitation data. The symptom matching analysis module is used to obtain the symptom set of each patient during the rehabilitation symptom stage by using the symptom data of each patient, and compare the similarity of the symptom set of each patient during the rehabilitation symptom stage with the core symptom set and general symptom set corresponding to the rehabilitation symptom stage to obtain the symptom control coefficient and symptom uncontrollability coefficient. The controllability calculation module is used to calculate the controllability of each patient's sudden symptoms based on the symptom controllability coefficient, symptom uncontrollability coefficient, and the potential delay of the rehabilitation training plan corresponding to the stage of rehabilitation symptom performance.

[0009] In one possible implementation, the typical symptom set construction module is specifically used for: The statistics include all sudden symptoms that led to changes in the rehabilitation training plan during each stage of rehabilitation symptom presentation in historical rehabilitation data, as well as the number of times these sudden symptoms occurred. Based on the frequency of occurrence of each sudden symptom, the sudden symptoms are divided into a core symptom set and a general symptom set using the maximum inter-class classification algorithm; the average frequency of occurrence of sudden symptoms in the core symptom set is greater than the average frequency of occurrence of sudden symptoms in the general symptom set.

[0010] In one possible implementation, the degree of control calculation module is specifically used for: The potential delay of the rehabilitation training plan for each patient's rehabilitation symptom manifestation stage was normalized, and the normalized potential delay of the rehabilitation training plan was used as a weighting coefficient to calculate the controllability coefficient and loss-of-control coefficient of symptom manifestation for each rehabilitation symptom manifestation stage, so as to determine the controllability of sudden symptoms for each patient.

[0011] In one possible implementation, the resource assessment unit includes: The baseline busyness calculation module is used to obtain the number of optional time slots for each rehabilitation training mode in the historical period, and to determine the baseline busyness for each optional time slot for each rehabilitation training mode based on the number of optional time slots in the historical period. The conflict coefficient calculation module is used to obtain the number of personnel conflicts in the selectable time periods of each rehabilitation training mode in the historical period, and to determine the busy conflict coefficient corresponding to each selectable time period of each rehabilitation training mode based on the number of personnel conflicts in the selectable time periods in the historical period. The busy performance integration module is used to determine the relative busy performance of each optional time period for each rehabilitation training mode based on the baseline busy performance and busy conflict coefficient corresponding to each optional time period for each rehabilitation training mode.

[0012] In one possible implementation, the scheduling decision unit is specifically used for: Based on the degree of control of sudden symptoms among multiple patients under the same rehabilitation training mode and the relative busyness of each selectable time period under the same rehabilitation training mode, a patient priority sequence and a time period busy sequence are generated respectively. Based on the matching and association between patient priority sequences and busy time sequence sequences, a scheduling recommendation scheme is generated.

[0013] This application provides a collaborative management method, including: Based on historical rehabilitation data, the patient's rehabilitation timeline is divided into multiple stages of rehabilitation symptom manifestation, and the potential delay in rehabilitation training plan for each stage of rehabilitation symptom manifestation is determined. Based on the current symptom data of each patient and the potential delays in the rehabilitation training plan during each patient's rehabilitation symptom stage, a symptom control analysis was conducted to determine the controllability of each patient's sudden symptoms. The controllability of sudden symptoms is used to characterize the degree to which the patient's symptom presentation matches the typical symptom characteristics of the rehabilitation symptom stage experienced. Based on historical resource usage data and personnel conflict data, the relative busy performance of each optional time period for each rehabilitation training mode is determined; the relative busy performance is used to characterize the resource strain of performing the corresponding rehabilitation training mode within the optional time period. Based on the degree of control of each patient's sudden symptoms and the relative busyness of each available time slot for each patient's scheduled rehabilitation training mode, a scheduling recommendation scheme is generated; the scheduling recommendation scheme is used to characterize the available time slots for each patient's scheduled rehabilitation training mode.

[0014] The present invention has the following beneficial effects: In view of the technical problem of low efficiency in existing management and scheduling, this application provides an intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources. The historical data analysis unit can divide the patient's rehabilitation timeline into multiple stages of rehabilitation symptom manifestation based on historical rehabilitation data, and determine the potential delay of the rehabilitation training plan for each stage, thereby quantifying the risk of the rehabilitation process. The real-time patient assessment unit can perform symptom control analysis based on the current symptom data of each patient and the potential delay of the rehabilitation training plan for each stage of rehabilitation symptom manifestation experienced by each patient, determining the controllability of each patient's sudden symptoms, thus accurately grasping the individual patient status. The resource assessment unit can determine the relative busyness of each optional time period for each rehabilitation training mode based on historical resource occupancy data and personnel conflict data, to obtain the resource supply and demand situation. Finally, the scheduling decision unit generates a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busyness of each optional time period for each rehabilitation training mode to be performed by each patient, realizing dynamic matching between patient status and resource status. Compared with existing rigid scheduling schemes, this application can effectively cope with dynamic disturbances in the rehabilitation process, while ensuring the continuity of patient rehabilitation, improving the utilization efficiency of rehabilitation resources, enhancing the flexibility and reliability of the overall scheduling system, and improving the management and scheduling efficiency of neurosurgical rehabilitation resources. Attached Figure Description

[0015] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0016] Figure 1 This is a system architecture diagram of an intelligent scheduling platform for neurosurgical rehabilitation resources provided in one embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a historical data analysis unit provided in one embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a real-time patient assessment unit provided in one embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of a resource assessment unit provided in one embodiment of the present invention; Figure 5 This is a flowchart illustrating a collaborative management method provided in one embodiment of the present invention. Detailed Implementation

[0017] The following detailed description, in conjunction with the accompanying drawings and preferred embodiments, illustrates the specific implementation of the intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources proposed in this invention.

[0018] The specific scheme of the intelligent scheduling platform and collaborative management method for neurosurgical rehabilitation resources provided by the present invention will be described in detail below with reference to the accompanying drawings.

[0019] Please see Figure 1 The diagram illustrates the system architecture of an intelligent scheduling platform for neurosurgical rehabilitation resources provided in an embodiment of the present invention. The intelligent scheduling platform 10 for neurosurgical rehabilitation resources includes: a historical data analysis unit 20, a real-time patient assessment unit 30, a resource assessment unit 40, and a scheduling decision unit 50.

[0020] Historical data analysis unit 20 is used to divide the patient's rehabilitation timeline into multiple rehabilitation symptom manifestation stages based on historical rehabilitation data, and to determine the potential delay in the rehabilitation training plan for each rehabilitation symptom manifestation stage.

[0021] Historical rehabilitation data includes medical records of past neurosurgical rehabilitation patients, rehabilitation training execution records, records of sudden symptom occurrences, and records of rehabilitation plan adjustments. This data provides a foundation for the division of rehabilitation stages and the assessment of potential delays. It should be noted that all sudden symptoms in the above records are uniformly classified and mapped through a pre-set standardized symptom label library (e.g., mapped to independent symptom types such as headache, nausea, and limb spasms), thereby clearly defining the specific type of sudden symptom occurring at any given time.

[0022] The rehabilitation timeline refers to the complete time interval from when the patient enters the rehabilitation period until the rehabilitation goal is achieved or the patient is discharged. For example, the rehabilitation timeline can be set to 30 days according to the conventional cycle of neurosurgical rehabilitation, with the day as the basic time unit for data statistics and analysis.

[0023] A rehabilitation symptom manifestation stage refers to a set of consecutive time periods on the rehabilitation timeline that share similar characteristics of sudden symptom onset and a risk of delay in the rehabilitation plan. Each rehabilitation symptom manifestation stage corresponds to a specific level of rehabilitation risk. The potential delay rate of the rehabilitation training plan is used to quantify the degree of risk that the rehabilitation training plan cannot be executed as originally planned due to sudden symptoms within that rehabilitation symptom manifestation stage. The higher the value, the greater the likelihood that the rehabilitation plan will be delayed in that stage.

[0024] For example, in the rehabilitation phase, patients undergoing neurosurgery (such as after stroke or traumatic brain injury) require significantly different rehabilitation resources and treatment plans depending on the stage of their rehabilitation symptoms as their neurological function recovers. For instance, in the early stages of rehabilitation, patients may experience high levels of muscle tone, limb spasticity, or dysphagia. At this time, rehabilitation resources are primarily focused on bedside physical therapy and speech and swallowing therapy, and the risk of delays in planning is high due to fluctuations in the patient's vital signs. However, as patients progress into the middle and later stages of rehabilitation, their vital signs and typical symptoms tend to stabilize, and rehabilitation resources gradually shift towards more complex cognitive function training and enhanced activities of daily living (ADL) training.

[0025] The real-time patient assessment unit 30 is used to perform symptom control analysis based on the current symptom data of each patient and the potential delay in the rehabilitation training plan of each patient's rehabilitation symptom stage, and to determine the controllability of each patient's sudden symptoms.

[0026] Among them, the controllability of sudden symptoms is used to characterize the degree to which the patient's symptom manifestations conform to the typical symptom characteristics of the recovery symptom manifestation stages experienced.

[0027] For example, patient symptom data includes labeled information on various sudden symptoms exhibited by the patient during the recovery symptom stage (such as whether discrete symptom events such as headache, nausea, and limb spasms occurred), which can be collected through medical monitoring equipment or recorded by medical staff. This application can compare the overlap between the patient's symptom set and the typical symptom set of the recovery symptom stage. If the patient's symptoms highly overlap with the typical symptom set of the recovery symptom stage, it indicates that the patient's recovery process is in line with expectations, the symptoms are under control, and the controllability value is high; conversely, if the patient experiences a large number of atypical symptoms (i.e., symptom events not within the typical symptom set), it indicates that the symptoms may be out of control, and the controllability value is low.

[0028] Resource assessment unit 40 is used to determine the relative busyness of each optional time period for each rehabilitation training mode based on historical resource usage data and personnel conflict data.

[0029] Among them, the relative busy performance level is used to characterize the resource shortage of implementing the corresponding rehabilitation training mode within the selectable time period. The higher the value, the more intense the resource competition within the selectable time period and the more constraints there are in carrying out the corresponding rehabilitation training mode.

[0030] The aforementioned rehabilitation training models refer to specialized rehabilitation training programs designed according to the type of neurological functional impairment in patients. These include physical therapy models (such as limb function training), cognitive training models (such as memory recovery training), and speech therapy models. Different rehabilitation training models have different requirements for medical equipment and the professional qualifications of therapists. Optional time slots refer to the specific time intervals within a day during which each rehabilitation training model can be conducted. For example, each day can be divided into three optional time slots: morning (8:00-12:00), afternoon (14:00-18:00), and evening (19:00-21:00). More refined optional time slots can also be defined based on the actual scheduling of the hospital's rehabilitation department. This application does not impose any limitations on this.

[0031] Historical resource usage data includes the number of times each rehabilitation training mode was conducted in each available time slot, reflecting the regular usage of resources during that time slot. Personnel conflict data refers to records of situations where different rehabilitation training modes could not be conducted simultaneously due to insufficient human resources such as therapists and nurses within the same available time slot.

[0032] The scheduling decision unit 50 is used to generate a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busyness of each available time period of the rehabilitation training mode to be performed by each patient.

[0033] The scheduling recommendation scheme is used to characterize the available time slots for each patient's assigned rehabilitation training mode, clarifying the correspondence between patients, rehabilitation training modes, and available time slots, for review and confirmation by medical staff before implementation. Based on the principle of risk hedging, this application prioritizes patients with high control over sudden symptoms (stable symptoms, strong tolerance to plan changes) to relatively busy available time slots to improve resource utilization efficiency, while assigning patients with low control over sudden symptoms (unstable symptoms, potentially requiring plan adjustments) to relatively idle available time slots, reserving buffer space for possible plan changes and reducing the impact on the overall scheduling scheme.

[0034] In some embodiments, the scheduling decision unit 50 is specifically used for: Based on the controllability of sudden symptoms of multiple patients under the same rehabilitation training mode and the relative busyness of each selectable time period under the same rehabilitation training mode, a patient priority sequence and a time period busy sequence are generated. Based on the matching and association between the patient priority sequence and the time period busy sequence, a scheduling recommendation scheme is generated.

[0035] For example, this application can obtain a list of all patients requiring the same rehabilitation training mode, and sort them in descending order according to the degree of control over sudden symptoms for each patient in the list, generating a patient priority sequence. The higher the degree of control over sudden symptoms, the more stable the patient's symptoms are, the stronger their tolerance to changes in the plan, and the higher their position in the sequence. If multiple patients with the same degree of control over sudden symptoms are encountered, they are sorted in ascending order according to the time they entered the rehabilitation period to ensure the uniqueness of the sequence.

[0036] Simultaneously, a list of all available time slots for this rehabilitation training mode is obtained, and these slots are sorted in descending order based on their relative busyness level to generate a busy time slot sequence. A higher relative busyness level indicates more scarce resources and a greater risk of conflict for that available time slot, placing it earlier in the sequence. If multiple available time slots have the same relative busyness level, they are then sorted in ascending order based on their chronological order within the day to ensure the uniqueness and certainty of the sequence.

[0037] Subsequently, this application can obtain the maximum number of patients that can be admitted for each selectable time period, which is determined based on resources such as the number of available devices and therapists within that selectable time period.

[0038] In this way, this application can sequentially associate and match patient priority sequences with busy time slot sequences based on the maximum capacity of admitted patients. For example, according to the order of the busy time slot sequence, the maximum number of patients that can be admitted for each available time slot is allocated to patients who are ranked higher in the patient priority sequence and have not yet been allocated. After completing the above matching, this application can output a scheduling recommendation scheme for doctors to review and confirm.

[0039] Based on the above technical solutions, the historical data analysis unit 20 in this application can divide the patient's rehabilitation timeline into multiple rehabilitation symptom manifestation stages based on historical rehabilitation data, and determine the potential delay of the rehabilitation training plan for each rehabilitation symptom manifestation stage, thereby quantifying the risks of the rehabilitation process. The real-time patient assessment unit 30 can perform symptom control analysis based on the current symptom manifestation data of each patient and the potential delay of the rehabilitation training plan for each patient's rehabilitation symptom manifestation stage, determine the controllability of each patient's sudden symptoms, and thus accurately grasp the individual patient status. The resource assessment unit 40 can determine the relative busy performance of each optional time period of each rehabilitation training mode based on historical resource occupancy data and personnel conflict data, so as to obtain the resource supply and demand situation. Finally, the scheduling decision unit 50 generates a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busy performance of each optional time period of the rehabilitation training mode to be performed by each patient, thereby realizing the dynamic matching of patient status and resource status. Compared with existing rigid scheduling schemes, this application can effectively cope with dynamic disturbances in the rehabilitation process, while ensuring the continuity of patient rehabilitation, improving the utilization efficiency of rehabilitation resources, enhancing the flexibility and reliability of the overall scheduling system, and improving the management and scheduling efficiency of neurosurgical rehabilitation resources.

[0040] As one possible embodiment of this application, combined with Figure 1 ,like Figure 2 As shown, the historical data analysis unit 20 includes: a shelving performance calculation module 21, a stage division module 22, and a potential delay determination module 23.

[0041] The postponement performance calculation module 21 is used to calculate the postponement performance of rehabilitation training for each unit period in the rehabilitation timeline based on the number of occurrences of various sudden symptoms within a unit period in the historical rehabilitation data and the corresponding proportion of changes to the rehabilitation training plan.

[0042] In some embodiments, the performance calculation module 21 is suspended, specifically for: The frequency of occurrence of various sudden symptoms in each unit period of the historical patient's rehabilitation timeline was counted. Based on the frequency of occurrence, the symptom weight of each sudden symptom was determined. Then, the number of times the rehabilitation training plan was modified due to each sudden symptom in each unit period of the historical patient's rehabilitation timeline was counted. Based on the number of modifications, the proportion of rehabilitation training plan modification for each sudden symptom was determined. Based on the symptom weight and the corresponding proportion of rehabilitation training plan modification, the rehabilitation training postponement performance degree was calculated.

[0043] For example, this application can, for each unit period of the rehabilitation timeline (e.g., one unit period can correspond to one day), first count the occurrence frequency of each type of sudden symptom (such as headache, high blood pressure) within that unit period, and then use linear normalization, that is, divide the occurrence frequency of each sudden symptom by the total occurrence frequency of all sudden symptoms within that unit period, thereby converting the occurrence frequency into symptom weights, ensuring that the sum of the symptom weights of each sudden symptom is 1. The symptom weights can quantify the prevalence of each sudden symptom within that unit period. It should be noted that if the total occurrence frequency of all sudden symptoms within that unit period is 0 (i.e., no sudden symptoms occurred within that period), then there is no need to perform the above symptom weight normalization calculation, and the rehabilitation training suspension performance of that unit period is directly assigned a value of 0, to characterize that the rehabilitation plan during that period was not interfered with by any sudden symptoms.

[0044] The aforementioned number of modifications refers to the number of times the planned rehabilitation training mode was altered (e.g., suspended, postponed, or adjusted) due to the occurrence of this sudden symptom. The rehabilitation training plan modification ratio is the ratio of this number of modifications to the total number of times all rehabilitation training modes that should have been executed in the historical original rehabilitation training plans within this unit of time. If the total number of times all rehabilitation training modes that should have been executed in the historical original rehabilitation training plans within this unit of time is 0, then the above ratio calculation is skipped, and the rehabilitation training plan modification ratio is directly assigned a value of 0. The modification ratio can quantify the degree of interference of each sudden symptom on the progress of rehabilitation plan execution.

[0045] Thus, this application can comprehensively assess the performance of rehabilitation training suspension based on the above two factors. For example, the performance of rehabilitation training suspension can satisfy the following formula: in, The first in the recovery timeline Performance of rehabilitation training suspension per unit time period The first in the recovery timeline The number of types of sudden symptoms that occur within a unit of time period. The first in the recovery timeline The first occurrence within a unit time period Symptom weights for sudden onset symptoms The first in the recovery timeline The first occurrence within a unit time period The proportion of changes to rehabilitation training plans for sudden symptoms.

[0046] in, After linear normalization, the sum of the symptom weights of each emerging symptom is equal to 1. This application can then use these symptom weights to... right Weighted calculations are performed to quantify the degree of rehabilitation training suspension performance in each time period, with symptom weights. The higher the percentage of changes to the rehabilitation training plan The larger the value, the more significantly the symptom interferes with the rehabilitation plan for that unit of time, and the greater the corresponding degree of postponement of rehabilitation training.

[0047] The phase division module 22 is used to divide the rehabilitation timeline into multiple time clusters based on the rehabilitation training postponement performance and a clustering algorithm.

[0048] In some embodiments, the stage division module 22 is specifically used for: Based on the performance of rehabilitation training suspension between any two unit periods on the rehabilitation timeline, the characteristic distance between the two unit periods is determined. Then, based on the characteristic distance, a clustering algorithm is used to perform cluster analysis on all unit periods on the rehabilitation timeline to form multiple time clusters.

[0049] The feature distance is used to measure the similarity of the risk of rehabilitation procrastination between two time periods. For example, this application can use the absolute value of the difference in rehabilitation training procrastination performance between two time periods, and simultaneously introduce the time interval between the two time periods on the rehabilitation timeline (e.g., the absolute value of the difference in day sequence) as a time penalty term. The weighted sum of the performance difference and the time interval is used as the feature distance between the two time periods. The smaller the feature distance, the more similar the procrastination risk of the two time periods, and the more suitable they are to be classified into the same stage.

[0050] This application can employ the density-based spatial clustering of applications with noise (DBSCAN) algorithm. This algorithm does not require pre-specifying the number of clusters and can automatically identify dense sets of unit time periods with similar characteristics as time clusters, while excluding outliers.

[0051] The potential delay determination module 23 is used to define the continuous unit time period contained in each time cluster as a rehabilitation symptom performance stage, and to determine the potential delay of the rehabilitation training plan for each rehabilitation symptom performance stage based on the rehabilitation training postponement performance of all unit time periods within each rehabilitation symptom performance stage.

[0052] Since a time interval penalty term has been introduced into the clustering feature distance, the time clusters formed by the clusters already possess strong continuity on the timeline. To ensure strict compliance with the stage definition, this application further traverses each time cluster, defining a continuous unit time period within the time cluster as a stage of rehabilitation symptom manifestation, thereby achieving the stage division of the rehabilitation timeline.

[0053] For example, this application can arithmetically average the rehabilitation training postponement rates across all time periods within each rehabilitation symptom stage to obtain the potential delay rate of the rehabilitation training plan for the corresponding rehabilitation symptom stage. This potential delay rate reflects the average postponement risk level of that rehabilitation symptom stage, providing a stage weighting basis for subsequent assessments of patient symptom control. For instance, if a rehabilitation symptom stage comprises three time periods with postponement rates of 0.6, 0.7, and 0.8, the potential delay rate of the rehabilitation training plan for that stage is (0.6 + 0.7 + 0.8) / 3 = 0.7.

[0054] Based on the above technical solution, this application further refines the historical data analysis unit 20 to achieve precise segmentation of rehabilitation symptom manifestation stages and quantitative assessment of the potential delay in rehabilitation training plans. Specifically, the shelving performance calculation module 21 quantifies the rehabilitation risk of each unit time period by integrating the symptom weights and modification ratios of sudden symptoms. The stage segmentation module 22 utilizes a clustering algorithm to achieve dynamic clustering of the rehabilitation timeline, avoiding the limitations of traditional fixed stage segmentation. The potential delay determination module 23 further realizes stage segmentation and risk quantification for each rehabilitation symptom manifestation stage, making the segmentation of rehabilitation symptom manifestation stages more closely reflect the risk characteristics of the actual rehabilitation process. This provides more accurate basic data for subsequent patient assessment and resource allocation, improving the scientific decision-making of the entire scheduling platform.

[0055] As one possible embodiment of this application, combined with Figure 1 ,like Figure 3 As shown, the real-time patient assessment unit 30 includes: a stage positioning module 31, a typical symptom set construction module 32, a symptom matching analysis module 33, and a controllability calculation module 34.

[0056] The stage positioning module 31 is used to determine the stage of rehabilitation symptoms experienced by each patient from multiple stages of rehabilitation symptoms based on the current rehabilitation time nodes of each patient.

[0057] The patient's recovery timeline can be represented by the cumulative recovery time since the patient entered the recovery period. The stage positioning module 31 can take the time when the patient entered the recovery period as the starting point and the current recovery timeline as the ending point, and extract all the recovery symptom manifestation stages covered within the continuous time interval as the recovery symptom manifestation stages experienced by each patient. In this way, by establishing the correspondence between the patient's real-time recovery progress and the subdivided recovery stages, it can be ensured that subsequent symptom assessments can be carried out based on the typical characteristics of the corresponding recovery symptom manifestation stages, avoiding deviations caused by cross-stage assessments.

[0058] The typical symptom set construction module 32 is used to divide the sudden symptoms that appear in each stage of rehabilitation symptom manifestation into a core symptom set and a general symptom set based on the records of sudden symptoms that lead to changes in the rehabilitation training plan in historical rehabilitation data.

[0059] In some embodiments, the typical symptom set construction module 32 is specifically used for: The system statistically analyzes all sudden symptoms that led to changes in the rehabilitation training plan during each stage of rehabilitation symptom presentation in historical rehabilitation data, along with the frequency of occurrence of each sudden symptom. Then, based on the frequency of occurrence of each sudden symptom, the system uses a maximum inter-class classification algorithm to divide each sudden symptom into a core symptom set and a general symptom set.

[0060] Among these, the average frequency of sudden symptoms in the core symptom set is greater than the average frequency of sudden symptoms in the general symptom set. The maximum inter-class classification algorithm is used to divide sudden symptoms into two categories based on their frequency of occurrence by determining a target threshold, maximizing the inter-class difference and minimizing the intra-class difference. For example, the maximum inter-class classification algorithm could be the maximum inter-class variance method. It should be noted that when the total number of sudden symptom types recorded during the recovery symptom manifestation stage is less than or equal to two, the maximum inter-class classification algorithm is skipped, and the symptom with the highest relative frequency is directly assigned to the core symptom set, while the rest are assigned to the general symptom set.

[0061] Since the core symptom set represents the most common symptoms that most frequently interfere with the rehabilitation plan during the rehabilitation symptom presentation stage, it is the typical symptom feature of the rehabilitation symptom presentation stage. The general symptom set, on the other hand, represents the occasional symptoms that interfere with the rehabilitation symptom presentation stage and are atypical symptom features.

[0062] Therefore, this application can further identify the core symptom set and the general symptom set based on the average occurrence frequency of the two types of sudden symptoms, that is, the average occurrence frequency of sudden symptoms in the core symptom set is greater than the average occurrence frequency of sudden symptoms in the general symptom set.

[0063] The symptom matching analysis module 33 is used to obtain the symptom set of each patient during the rehabilitation symptom stage by using the symptom data of each patient, and to compare the similarity of the symptom set of each patient during the rehabilitation symptom stage with the core symptom set and general symptom set corresponding to the rehabilitation symptom stage, respectively, to obtain the symptom control coefficient and symptom uncontrollability coefficient.

[0064] The symptom set exhibited by the patient during the recovery symptom manifestation stage refers to the set of sudden symptoms displayed by the patient during that stage. The similarity comparison described above can be achieved by calculating the proportion of the patient's actual symptoms in different typical levels, thus accurately measuring the degree of matching between the two.

[0065] If the patient's symptom set is an empty set during the recovery symptom manifestation stage, the symptom control coefficient for that stage is set to a preset maximum value (e.g., 1.0), and the symptom uncontrollability coefficient is set to a preset minimum value (e.g., 0.0), to indicate that the patient has no sudden symptoms during that stage and fully meets or even exceeds typical recovery expectations.

[0066] If not empty, the symptom control coefficient is the concentration of symptoms exhibited by the patient in the corresponding recovery symptom stage, the number of symptoms belonging to the core symptom set of that stage, and the proportion of the total number of symptoms in the patient's symptom set (i.e., the core symptom percentage). The higher the value, the more the patient's symptoms conform to the conventional characteristics of that recovery symptom stage, and the more controlled the recovery process is.

[0067] The symptom control failure coefficient is the proportion of the number of symptoms that a patient exhibits in the corresponding recovery symptom stage, which belongs to the general symptom set of that stage, to the total number of symptoms in the patient's symptom set (i.e., the proportion of atypical symptoms). The higher the value, the more common the patient is to have symptoms that are not common in that recovery symptom stage. Even if only a few rare symptoms appear, the proportion will increase significantly, thus accurately triggering an early warning and indicating that the recovery process may be out of control.

[0068] The controllability calculation module 34 is used to calculate the controllability of each patient's sudden symptoms based on the symptom controllability coefficient, the symptom uncontrollability coefficient, and the potential delay of the rehabilitation training plan corresponding to the rehabilitation symptom stage.

[0069] In some embodiments, the degree of control calculation module 34 is specifically used for: The potential delay of the rehabilitation training plan for each patient's rehabilitation symptom manifestation stage was normalized, and the normalized potential delay of the rehabilitation training plan was used as a weighting coefficient to calculate the controllability coefficient and loss-of-control coefficient of symptom manifestation for each rehabilitation symptom manifestation stage, so as to determine the controllability of sudden symptoms for each patient.

[0070] For example, the degree of control over sudden symptoms satisfies the following formula: in, To ensure the patient's sudden onset symptoms were under control, This refers to the number of stages of recovery symptoms experienced by the patient. is the time decay factor, where It is a constant between 0 and 1 (for example, taking the empirical value of 0.8). The current time corresponds to the stage number of the rehabilitation symptom manifestation stage. (If the unit time period corresponding to the current time was excluded as an outlier in the above cluster analysis and does not belong to any rehabilitation symptom manifestation stage, then the stage number of the nearest classified rehabilitation symptom manifestation stage to the current time is extracted as the stage number corresponding to the current time.) (participating in calculations) For the first The sequence number of each stage of recovery symptom presentation, then Indicates the first The time interval between each stage of recovery symptom manifestation and the current stage is a factor; the larger the interval, the more significant the attenuation, making the recent stage have a much greater impact on controllability than the earlier stages. The first time this patient experienced Potential delays in rehabilitation training programs at each stage of rehabilitation symptom presentation. The first time this patient experienced The control coefficient of symptom manifestations at each stage of recovery symptom manifestation. The first time this patient experienced The symptom control coefficient of each recovery symptom stage. This is a normalization function used to normalize the potential delays in the rehabilitation training plan during the patient's symptom presentation stages into weighted coefficients, ensuring that the sum of the weights is 1.

[0071] Symptom control coefficient The higher the level, the greater the rate of symptom control failure. The lower the weight, the more controlled the patient's symptoms are at this stage of recovery. Characterizing the importance of risks at different stages, weights The larger the value, the more significant the impact on the overall controllability. Through the weighted calculations described above, the overall controllability of the patient's sudden symptoms is obtained. Sudden symptoms under control The higher the value, the more the patient's overall recovery progress is in line with expectations, the symptoms are under control, the greater the tolerance to changes in the recovery plan, and the better the control over sudden symptoms. The lower the value, the more likely the patient's symptoms may be out of control, and more room for adjustment needs to be reserved in resource allocation.

[0072] Based on the above technical solution, this application achieves accurate quantitative assessment of patient symptoms by further refining the real-time patient assessment unit 30. The stage positioning module 31 can determine the rehabilitation symptom manifestation stage experienced by each patient from multiple rehabilitation symptom manifestation stages according to the current rehabilitation time node of each patient, ensuring the pertinence of the assessment. The typical symptom set construction module 32, based on the records of sudden symptoms that led to changes in the rehabilitation training plan in historical rehabilitation data, divides the sudden symptoms manifested in the rehabilitation symptom manifestation stage into a core symptom set and a general symptom set, realizing a two-dimensional division of core symptoms and other symptoms. The symptom matching analysis module 33 further realizes the quantitative analysis of symptom control and uncontrollability by comparing the patient's symptoms with the above two dimensions of symptoms. The controllability calculation module 34 integrates the risk weights of each stage, and the final sudden symptom controllability can comprehensively and objectively reflect the patient's current rehabilitation status.

[0073] As one possible embodiment of this application, combined with Figure 1 ,like Figure 4 As shown, the resource assessment unit 40 includes: a baseline busyness calculation module 41, a conflict coefficient calculation module 42, and a busyness performance integration module 43.

[0074] The baseline busyness calculation module 41 is used to obtain the number of optional time slots for each rehabilitation training mode in the historical period, and to determine the baseline busyness corresponding to each optional time slot for each rehabilitation training mode based on the number of optional time slots in the historical period.

[0075] The baseline busyness is used to characterize the regular busyness of the rehabilitation training mode within the selectable time period. The higher the value, the higher the regular resource occupancy rate and the less idle resources there are during the selectable time period.

[0076] For example, the baseline busyness can be represented by the ratio of the average number of events held during the available time period to the maximum number of events held during the available time period in the historical period. If the maximum number of events held during the available time period in the historical period is 0, then there is no need to perform the aforementioned ratio calculation, and the baseline busyness of the available time period can be directly assigned to 0.

[0077] The selection of the historical period should consider both the timeliness and representativeness of the data. For example, the past 30 days can be selected as the historical period to ensure that the data reflects recent resource usage patterns. The number of times a rehabilitation training model is implemented refers to the actual number of times that rehabilitation training model is performed within a certain selectable time period.

[0078] The conflict coefficient calculation module 42 is used to obtain the number of personnel conflicts in the selectable time periods of each rehabilitation training mode in the historical period, and to determine the busy conflict coefficient corresponding to each selectable time period of each rehabilitation training mode based on the number of personnel conflicts in the selectable time periods in the historical period.

[0079] Among them, the number of personnel conflicts refers to the total number of medical staff involved when the scheduling of therapists, nurses and other medical staff conflicts within the same available time period, resulting in competition for the same group of medical staff between this rehabilitation training mode and other rehabilitation training modes.

[0080] For example, this application can accumulate and average the number of personnel conflicts in the selectable time periods within a historical period to obtain the average number of personnel conflicts. Then, the average number of personnel conflicts in each selectable time period is normalized (e.g., maximum and minimum value normalization, where the maximum value is the maximum of the average number of personnel conflicts in the selectable time periods within the historical period, and the minimum value is the minimum of the average number of personnel conflicts in the selectable time periods within the historical period), mapping the average number of personnel conflicts to between 0 and 1, which serves as the busy conflict coefficient. It should be noted that if the average number of personnel conflicts in each selectable time period is completely equal (i.e., the maximum and minimum values ​​are the same, and the range is 0), the above normalization process is skipped: if the equal average is 0, the busy conflict coefficient for all selectable time periods is directly assigned to 0 to indicate that there are no personnel conflicts in any time period; if the equal average is greater than 0, the busy conflict coefficient for all selectable time periods is directly assigned to 0.5 to indicate that the level of personnel conflict in each time period is consistent and there is no distinguishing feature.

[0081] The busy conflict coefficient is used to characterize the degree of resource tension caused by human resource conflicts during the available time period. The higher the value, the more intense the competition for human resources during that time period, and the more constraints there are on the implementation of the rehabilitation training model.

[0082] Busyness performance integration module 43 is used to determine the relative busyness performance of each optional time period of each rehabilitation training mode based on the baseline busyness and busyness conflict coefficient corresponding to each optional time period of each rehabilitation training mode.

[0083] For example, the relative busy performance level satisfies the following formula: in, For the first The first rehabilitation training model The relative busyness of each selectable time period For the first The first rehabilitation training model The baseline busyness for each selectable time period For the first The first rehabilitation training model Busyness / conflict coefficient for each selectable time period For the normalization function (e.g., maximum-minimum normalization, where the maximum value is the value of each selectable time period for each rehabilitation training mode). The maximum and minimum values ​​are for each selectable time period of each rehabilitation training mode. The minimum value in the range is used to map the calculation result to a value between 0 and 1. It should be noted that if the selectable time periods for each rehabilitation training mode... If the calculation results are completely consistent (i.e., the maximum value equals the minimum value and the range is 0), then the above normalization process is skipped, and the relative busyness of all selectable time periods is directly assigned a preset constant (such as 0.5) to characterize the consistent resource tension and lack of differentiation in each time period.

[0084] Baseline Busyness The busy conflict coefficient reflects the risk of routine occupancy of hardware resources such as equipment and venues. This reflects the competitive risk of human resources such as medical staff. This application can integrate the two risks in the above way to comprehensively quantify the resource shortage of the rehabilitation training mode during the available time period. The more saturated the hardware resources are and the more intense the human resource conflict, the higher the relative busyness performance.

[0085] Based on the above technical solution, this application further refines the resource assessment unit 40 to achieve a multi-dimensional assessment of the workload of rehabilitation resources. The baseline workload calculation module 41 quantifies the risk of routine occupancy of hardware resources, the conflict coefficient calculation module 42 quantifies the competition risk of human resources such as medical staff, and the workload performance integration module 43 realizes the organic integration of the two types of risks. The above solution can more comprehensively reflect the characteristics of resource supply and demand, and helps to improve resource utilization efficiency and the feasibility of scheduling schemes.

[0086] Please see Figure 5 The diagram illustrates a flowchart of a collaborative management method according to an embodiment of the present invention, which includes the following steps: Step 501: Based on historical rehabilitation data, divide the patient's rehabilitation timeline into multiple rehabilitation symptom manifestation stages, and determine the potential delay in the rehabilitation training plan for each rehabilitation symptom manifestation stage.

[0087] Step 502: Based on the current symptom data of each patient and the potential delay in the rehabilitation training plan during the rehabilitation symptom stages experienced by each patient, conduct symptom control analysis to determine the controllability of each patient's sudden symptoms.

[0088] Among them, the controllability of sudden symptoms is used to characterize the degree to which the patient's symptom manifestations conform to the typical symptom characteristics of the recovery symptom manifestation stages experienced.

[0089] Step 503: Based on historical resource usage data and personnel conflict data, determine the relative busyness of each optional time period for each rehabilitation training mode.

[0090] Among them, the relative busy performance level is used to characterize the resource shortage level for performing the corresponding rehabilitation training mode within the selectable time period.

[0091] Step 504: Generate a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busyness of each available time period for each patient's rehabilitation training mode.

[0092] The scheduling recommendation scheme is used to characterize the available time slots for the rehabilitation training modes to be performed for each patient.

[0093] For related information, please refer to the above embodiments; further details will not be provided here.

[0094] It should be noted that the order of description of the above embodiments does not indicate superiority or inferiority in execution, and the flowcharts in the accompanying drawings are not limited to a specific sequential order; some steps can be processed in parallel. Each embodiment focuses on describing the differences from other embodiments; similar or identical parts can be referred to interchangeably.

Claims

1. A smart scheduling platform for neurosurgical rehabilitation resources, characterized in that, include: The historical data analysis unit is used to divide the patient's rehabilitation timeline into multiple rehabilitation symptom manifestation stages based on historical rehabilitation data, and to determine the potential delay in the rehabilitation training plan for each rehabilitation symptom manifestation stage. The real-time patient assessment unit is used to perform symptom control analysis based on the current symptom data of each patient and the potential delay in the rehabilitation training plan of each patient's rehabilitation symptom stage, and to determine the control degree of sudden symptom onset for each patient; the control degree of sudden symptom onset is used to characterize the degree of consistency between the patient's symptom presentation and the typical symptom characteristics of the rehabilitation symptom stage experienced. The resource assessment unit is used to determine the relative busy performance of each optional time period for each rehabilitation training mode based on historical resource occupancy data and personnel conflict data; the relative busy performance is used to characterize the resource strain of performing the corresponding rehabilitation training mode during the optional time period. The scheduling decision unit is used to generate a scheduling recommendation plan based on the controllability of each patient's sudden symptoms and the relative busyness of each available time period for each patient's rehabilitation training mode. The scheduling recommendation scheme is used to characterize the selectable time slots for the rehabilitation training modes to be performed for each patient.

2. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 1, characterized in that, The historical data analysis unit includes: The postponement performance calculation module is used to calculate the postponement performance of rehabilitation training for each unit period in the rehabilitation timeline based on the number of occurrences of various sudden symptoms within a unit period in historical rehabilitation data and the corresponding proportion of changes to the rehabilitation training plan. The phase division module is used to divide the rehabilitation timeline into multiple time clusters based on the rehabilitation training postponement performance and a clustering algorithm. The potential delay determination module defines a continuous unit time period contained in each time cluster as a rehabilitation symptom performance stage, and determines the potential delay of the rehabilitation training plan for each rehabilitation symptom performance stage based on the rehabilitation training postponement performance of all unit time periods within each rehabilitation symptom performance stage.

3. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 2, characterized in that, The shelving performance calculation module is specifically used for: The frequency of occurrence of various sudden symptoms in each unit period of the historical patient's recovery timeline is counted, and the symptom weight of each sudden symptom is determined based on the frequency of occurrence. The number of times the rehabilitation training plan was modified due to each type of sudden symptom in each unit period of the historical patient's rehabilitation timeline was counted, and the proportion of rehabilitation training plan modification for each type of sudden symptom was determined based on the number of modifications. The rehabilitation training postponement performance degree is calculated based on the symptom weights and the corresponding proportions of changes to the rehabilitation training plan.

4. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 2, characterized in that, The stage division module is specifically used for: The characteristic distance between the two unit time periods is determined based on the degree of rehabilitation training suspension performance between any two unit time periods on the rehabilitation timeline. Based on the aforementioned feature distance, the clustering algorithm is used to perform cluster analysis on all unit time periods on the rehabilitation timeline to form the multiple time clusters.

5. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 1, characterized in that, The real-time patient assessment unit includes: The stage positioning module is used to determine the stage of rehabilitation symptoms experienced by each patient from the multiple stages of rehabilitation symptom manifestation based on the current rehabilitation time node of each patient. The typical symptom set construction module is used to divide the sudden symptoms that appear in each stage of rehabilitation symptom manifestation into a core symptom set and a general symptom set based on the records of sudden symptoms that led to changes in the rehabilitation training plan in historical rehabilitation data. The symptom matching analysis module is used to obtain the symptom set exhibited by each patient during the rehabilitation symptom stage through the symptom performance data of each patient, and to compare the similarity of the symptom set exhibited by each patient during the rehabilitation symptom stage with the core symptom set and the general symptom set corresponding to the rehabilitation symptom stage, respectively, to obtain the symptom performance control coefficient and the symptom performance uncontrollable coefficient. The controllability calculation module is used to calculate the controllability of each patient's sudden symptoms based on the controllability coefficient of the symptom manifestation, the uncontrollability coefficient of the symptom manifestation, and the potential delay of the rehabilitation training plan corresponding to the rehabilitation symptom manifestation stage.

6. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 5, characterized in that, The typical symptom set construction module is specifically used for: The statistics include all sudden symptoms that led to changes in the rehabilitation training plan during each stage of rehabilitation symptom presentation in historical rehabilitation data, as well as the number of times these sudden symptoms occurred. Based on the frequency of occurrence of each sudden symptom, the sudden symptoms are divided into a core symptom set and a general symptom set using a maximum inter-class classification algorithm; the average frequency of occurrence of sudden symptoms in the core symptom set is greater than the average frequency of occurrence of sudden symptoms in the general symptom set.

7. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 5, characterized in that, The controllability calculation module is specifically used for: The potential delay of the rehabilitation training plan for each patient's rehabilitation symptom manifestation stage was normalized, and the normalized potential delay of the rehabilitation training plan was used as a weighting coefficient to calculate the controllability coefficient and loss-of-control coefficient of symptom manifestation for each rehabilitation symptom manifestation stage, so as to determine the controllability of sudden symptoms for each patient.

8. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 1, characterized in that, The resource assessment unit includes: The baseline busyness calculation module is used to obtain the number of optional time slots for each rehabilitation training mode in a historical period, and to determine the baseline busyness for each optional time slot for each rehabilitation training mode based on the number of optional time slots in the historical period. The conflict coefficient calculation module is used to obtain the number of personnel conflicts in the selectable time periods of each rehabilitation training mode in the historical period, and to determine the busy conflict coefficient corresponding to each selectable time period of each rehabilitation training mode based on the number of personnel conflicts in the selectable time periods in the historical period. The busy performance integration module is used to determine the relative busy performance of each optional time period for each rehabilitation training mode based on the baseline busy performance and busy conflict coefficient corresponding to each optional time period for each rehabilitation training mode.

9. The intelligent scheduling platform for neurosurgical rehabilitation resources according to claim 1, characterized in that, The scheduling decision unit is specifically used for: Based on the degree of control of sudden symptoms among multiple patients under the same rehabilitation training mode and the relative busyness of each selectable time period under the same rehabilitation training mode, a patient priority sequence and a time period busy sequence are generated respectively. Based on the matching and association between the patient priority sequence and the busy time sequence, a scheduling recommendation scheme is generated.

10. A collaborative management method, characterized in that, include: Based on historical rehabilitation data, the patient's rehabilitation timeline is divided into multiple stages of rehabilitation symptom manifestation, and the potential delay in rehabilitation training plan for each stage of rehabilitation symptom manifestation is determined. Based on the current symptom data of each patient and the potential delay in the rehabilitation training plan during each patient's rehabilitation symptom stage, a symptom control analysis is conducted to determine the controllability of the sudden symptoms of each patient; the controllability of the sudden symptoms is used to characterize the degree to which the patient's symptom manifestations conform to the typical symptom characteristics of the rehabilitation symptom stage experienced. Based on historical resource usage data and personnel conflict data, the relative busyness of each optional time period for each rehabilitation training mode is determined; the relative busyness is used to characterize the resource strain of performing the corresponding rehabilitation training mode within the optional time period. Based on the degree of control over each patient's sudden symptoms and the relative busyness of each available time slot for each patient's rehabilitation training mode, a scheduling recommendation plan is generated. The scheduling recommendation scheme is used to characterize the selectable time slots for the rehabilitation training modes to be performed for each patient.