Multi-device collaborative control and nursing management method and system for organ rehabilitation

Abstract

The present application relates to the technical field of organ rehabilitation, and more particularly to a method and system for multi-device collaborative control and nursing management of organ rehabilitation, which constructs a dynamic topology graph by cross-modal fusion of multi-channel organ function monitoring signals, identifies nodes and paths of organ function collaborative recovery, performs function target point matching of devices and nursing, obtains an interaction spectrum combined with physical field superposition mapping, generates a collaborative strategy including device power distribution, timing arrangement, nursing positioning and skill matching based on game theory, and realizes precise collaboration of multiple devices and nursing, thereby improving rehabilitation efficiency and safety.

Description

Technical Field

[0001] This invention relates to the field of organ rehabilitation technology, and in particular to a method and system for multi-device collaborative control and nursing management of organ rehabilitation. Background Technology

[0002] In organ rehabilitation, multiple rehabilitation devices (such as respiratory trainers, external counterpulsation devices, and neuromuscular electrical stimulators) are typically used simultaneously in conjunction with nursing interventions to promote the coordinated recovery of multiple organ functions. Current practices generally involve separately collecting monitoring signals for each organ (such as electrocardiogram, electroencephalogram, electromyography, and blood oxygenation), assessing the condition of each organ using single-channel or simple multi-parameter threshold judgments, and independently setting parameters and controlling the start / stop of each rehabilitation device based on the equipment manual or clinical experience. Nurses then perform nursing procedures such as turning over, back percussion, and postural drainage at fixed times or in specific positions according to standardized operating procedures. There is a lack of real-time communication and dynamic coordination between equipment operation and nursing actions.

[0003] Current methods for monitoring data fusion are at a low level, typically performing only simple synchronization or overlay analysis between signals from different organs. They fail to deeply extract the correlation characteristics, such as changes in coupling strength and information transmission delays, between organ subsystems during functional recovery. This makes it difficult for clinical assessments to accurately capture the dynamic stages of multi-organ synergistic recovery, leading to lags in rehabilitation goal setting and equipment parameter adjustments. The action fields of rehabilitation equipment (e.g., current distribution of electrical stimulation, pressure area of ​​mechanical massage) and the influence fields of nursing operations (e.g., vibration transmission range of back percussion, impact of positional changes on organ location) often overlap or interfere in physical space. Current practices lack a systematic identification of the synergistic or inhibitory effects of these two interventions on the organ functional network, resulting in offsetting or overstimulating effects between different devices and between devices and nursing care, affecting rehabilitation efficiency and increasing the risk of patient discomfort. Summary of the Invention

[0004] The present invention provides a method and system for collaborative control and nursing management of multiple devices for organ rehabilitation, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for collaborative control and nursing management of multiple devices for organ rehabilitation, comprising:

[0006] Cross-modal fusion processing was performed on multi-channel organ function monitoring signal data of target patients to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and a dynamic topology graph structure representing the functional state of organs was constructed.

[0007] Identify multi-connection node sets and single-path-dependent connection paths for the coordinated recovery of organ function on the dynamic topology graph structure;

[0008] Based on the location information of the multi-connection node set and the single-path-dependent connection path, the real-time operating status data of various types of rehabilitation equipment and the spatiotemporal trajectory data of the operations performed by nursing staff are matched for functional targets.

[0009] By constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operations, we can identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network, and obtain the interaction spectrum of the equipment-nursing-organ network.

[0010] Based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, and combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

[0011] Cross-modal fusion processing is performed on multi-channel organ function monitoring signal data of target patients to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and to construct a dynamic topology graph structure representing the organ functional state, including:

[0012] The different modal signals in the multi-channel organ function monitoring signal data are time-stamped and aligned. The initial signal data from the ECG monitoring channel, blood oxygen monitoring channel, respiratory monitoring channel and blood pressure monitoring channel are mapped to a unified time axis and defined as a modal signal respectively, so as to obtain a time-synchronized multimodal signal sequence.

[0013] Frequency domain decomposition is performed on each mode signal in the time-synchronized multimodal signal sequence to extract the spectral feature vector of each mode signal within a preset frequency band. The cross-correlation coefficient between the spectral feature vectors of different mode signals is calculated and constructed into a cross-modal spectral correlation matrix.

[0014] Based on the cross-correlation coefficients of the cross-modal spectrum correlation matrix for modal pairs that exceed a preset correlation threshold, the information transmission delay characteristics are determined by calculating the time difference between the arrival times of the signal peaks between the modal pairs, and the functional coupling strength between the corresponding organ subsystems of the modal pairs is defined according to the magnitude of the cross-correlation coefficients.

[0015] The different organ subsystems are defined as nodes in a dynamic topology graph structure. The functional coupling strength between the organ subsystems is defined as the edge weight connecting the nodes. The information transmission delay feature is defined as the delay attribute of the edge, thus constructing the dynamic topology graph structure.

[0016] Identifying multi-connected node sets and single-path-dependent connection paths for coordinated organ function recovery on the dynamic topology graph structure includes:

[0017] Traverse all nodes in the dynamic topology graph structure, count the number of edges directly connected to the current node as the connectivity value of the current node, and obtain the connectivity distribution of all nodes in the dynamic topology graph structure.

[0018] Calculate the statistical distribution characteristic parameters of the connectivity values ​​of all nodes based on the connectivity distribution, and filter the nodes whose connectivity values ​​exceed the statistical distribution characteristic parameters as candidate multi-connection nodes;

[0019] For each candidate multi-connection node, the collaborative recovery contribution of the candidate multi-connection node in the organ function information transmission network is calculated based on the information transmission delay characteristics and edge weights corresponding to the edges connected to the candidate multi-connection node. Candidate multi-connection nodes whose collaborative recovery contribution exceeds a preset contribution threshold are determined as nodes in the set of multi-connection nodes.

[0020] For any two nodes in the dynamic topology graph structure, by searching all reachable paths connecting the two nodes, identify the node pairs with a total number of reachable paths equal to one, and mark the unique reachable path as the single-path-dependent connection path.

[0021] Based on the location information of the multi-connection node set and the single-path-dependent connection path, the real-time operating status data of various types of rehabilitation equipment and the spatiotemporal trajectory data of the operations performed by nursing staff are matched for functional targets, including:

[0022] Extract the organ subsystem identifier corresponding to each node in the multi-connection node set and the organ subsystem identifier corresponding to the node connected by the single path dependency connection path, and construct a spatial distribution mapping table of organ functional target points.

[0023] The real-time operating status data includes the organ part identifiers of the device and the device output parameters. The organ part identifiers of the device are matched with the organ subsystem identifiers in the organ function target spatial distribution mapping table. Rehabilitation devices whose organ part identifiers of the device match the set of multiple connection nodes are identified as key devices, and rehabilitation devices whose organ part identifiers of the device match the single path-dependent connection path are identified as protective devices.

[0024] The spatiotemporal trajectory data includes the organ site identifier of the operation and the operation intensity parameter. The organ site identifier of the operation is matched with the organ subsystem identifier in the organ function target spatial distribution mapping table. Nursing operations in which the organ site identifier of the operation hits the set of multiple connected nodes are identified as critical nursing operations, and nursing operations in which the organ site identifier of the operation hits the single path-dependent connection path are identified as protective nursing operations.

[0025] By constructing a spatial superposition mapping mechanism between the physical field of equipment and the influence domain of nursing operations, the synergistic enhancement and mutual inhibition regions of equipment action and nursing intervention in the organ function network are identified, resulting in the interaction spectrum of the equipment-nursing-organ network, including:

[0026] Based on the output parameters of the key equipment and the organ location identification of the equipment, the spatial diffusion range of the physical field of the key equipment in the organ functional network is calculated.

[0027] Based on the operation intensity parameter and the organ site where the operation was applied, the spatial propagation range of the operation influence domain of the key nursing operation in the organ function network is calculated.

[0028] The spatial diffusion range of the key equipment and the spatial propagation range of the key nursing operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as key location collaboration areas.

[0029] The spatial diffusion range of the protective device and the spatial propagation range of the protective care operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as vulnerable path collaboration areas.

[0030] For each node in the critical location collaboration region and the vulnerable path collaboration region, the parameter compatibility between the device output parameters and the operation intensity parameters is calculated. When the parameter compatibility is higher than a preset compatibility threshold, the current node is marked as a node in the collaboration enhancement region. When the parameter compatibility is lower than the preset compatibility threshold, the node is marked as a node in the mutual inhibition region, thus obtaining the interaction spectrum of the device-care-organ network.

[0031] Based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, and combined with patient physiological stress response data and physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes device power allocation sequences, device action timing arrangement schemes, nursing operation spatial positioning instructions, and nursing staff skill requirement matching tables.

[0032] Extract the node identifiers of the cooperative enhancement region and the node identifiers of the mutually inhibiting region in the interaction spectrum, and statistically analyze the spatial distribution density of the node identifiers of the cooperative enhancement region and the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure.

[0033] The patient physiological stress response data includes the patient's physiological tolerance threshold to the device output parameters and the patient's physiological tolerance threshold to the operation intensity parameters; the physical field distribution data of the rehabilitation environment includes the physical field intensity distribution at different spatial locations in the rehabilitation environment.

[0034] A mechanism for solving the equipment-nursing resource equilibrium based on a game theory framework is constructed. The equilibrium solution of the game theory framework is solved to obtain the optimal configuration values ​​of equipment output parameters and operation intensity parameters.

[0035] Based on the optimized configuration values ​​of the device output parameters and the optimized configuration values ​​of the operation intensity parameters, a device power allocation sequence, a device action timing arrangement scheme, a nursing operation space positioning instruction, and a nursing staff skill requirement matching table are generated and combined into a collaborative control and nursing management strategy.

[0036] A mechanism for resolving the equipment-nursing resource equilibrium based on a game theory framework is constructed. The equilibrium solution within the game theory framework is obtained, yielding optimized configuration values ​​for equipment output parameters and operational intensity parameters, including:

[0037] The key equipment, the protective equipment, the key nursing operation, and the protective nursing operation are taken as game participants, and the equipment output parameters and the operation intensity parameters are taken as game strategy variables.

[0038] The functional gain weight is calculated based on the spatial distribution density of the node identifiers of the collaborative enhancement region in the dynamic topology graph structure, and the functional loss weight is calculated based on the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure. A game payoff function is constructed based on the functional gain weight and the functional loss weight.

[0039] The physiological tolerance threshold of the patient to the output parameters of the device and the physiological tolerance threshold of the patient to the operation intensity parameters are used as upper limits of the game strategy variables, and the distribution of physical field intensity at different spatial locations in the rehabilitation environment is used as the spatial range constraint of the device to construct the game constraint conditions.

[0040] Based on the game participants, the game strategy variables, the game payoff function, and the game constraints, the Nash equilibrium solution of the game theory framework is obtained, and the optimal configuration values ​​of the equipment output parameters and the operation intensity parameters are obtained.

[0041] A second aspect of the present invention provides a multi-device collaborative control and nursing management system for organ rehabilitation, comprising:

[0042] The topology building unit is used to perform cross-modal fusion processing on the multi-channel organ function monitoring signal data of the target patient, extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, construct a dynamic topology graph structure that characterizes the organ function state, and identify the multi-connection node set and single-path-dependent connection path of organ function collaborative recovery on the dynamic topology graph structure.

[0043] The target matching unit is used to perform functional target matching between the real-time operating status data of multiple types of rehabilitation equipment and the spatiotemporal trajectory data of the operation performed by the nursing staff, based on the location information of the set of multiple connection nodes and the single path-dependent connection path.

[0044] The action mapping unit is used to identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network by constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operation, and to obtain the interaction spectrum of equipment-nursing-organ network.

[0045] The collaborative strategy unit is used to generate a collaborative control and nursing management strategy based on the distribution characteristics of the collaborative enhancement region and the mutual inhibition region in the interaction spectrum, combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

[0046] A third aspect of the present invention provides an electronic device, comprising:

[0047] processor;

[0048] Memory used to store processor-executable instructions;

[0049] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0050] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0051] The beneficial effects of this application are as follows:

[0052] This method constructs a dynamic topology graph reflecting functional coupling and time delay characteristics by performing cross-modal fusion of multi-channel organ function monitoring signals. This allows for the accurate capture of the collaborative recovery status and information transmission dynamics between organ subsystems, significantly improving the accuracy and real-time performance of identifying multi-organ interactions during patient rehabilitation. Based on the multi-connected node set and single-path-dependent connection paths identified on the dynamic topology graph, it enables the accurate location of key hubs and vulnerable links in the organ function network, providing a highly reliable basis for the targeted allocation of subsequent equipment and nursing resources.

[0053] By matching the operational status data of various types of rehabilitation equipment with the spatiotemporal trajectories of nursing staff operations to functional targets, and constructing a spatial superposition mapping mechanism between the physical action field of the equipment and the influence domain of nursing operations, it is possible to clearly distinguish between synergistic enhancement regions and mutually inhibiting regions, forming an interaction spectrum of the equipment-nursing-organ network. This interaction spectrum intuitively presents the distribution of positive and negative effects of different interventions in the organ functional network, effectively avoiding conflicts and cancellations between equipment effects and nursing interventions, and greatly improving the synergistic efficiency and safety of multiple interventions.

[0054] Based on the constructed interaction spectrum, and combined with patient physiological stress response data and the physical field distribution of the rehabilitation environment, a collaborative control and nursing management strategy is automatically generated using a game theory-based device-nursing resource equilibrium solution mechanism. This strategy includes device power allocation sequences, action timing arrangement schemes, nursing operation spatial positioning instructions, and skill demand matching tables. This mechanism can dynamically balance power competition among multiple devices and the spatiotemporal constraints of nursing resources. While ensuring patient physiological tolerance, it maximizes the coverage of collaborative enhancement areas and minimizes the impact of mutually inhibiting areas, thereby significantly shortening the rehabilitation cycle, reducing resource waste, and improving the individualized adaptation and execution stability of the overall rehabilitation plan. Attached Figure Description

[0055] Figure 1 A flowchart illustrating the multi-device collaborative control and nursing management method for organ rehabilitation;

[0056] Figure 2 A schematic diagram illustrating the process of constructing the network interaction spectrum of equipment-nursing-organs. Detailed Implementation

[0057] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0058] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0059] Figure 1 This is a flowchart illustrating the multi-device collaborative control and nursing management method for organ rehabilitation according to an embodiment of the present invention.

[0060] The multi-device collaborative control and nursing management methods for organ rehabilitation include:

[0061] Cross-modal fusion processing was performed on multi-channel organ function monitoring signal data of target patients to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and a dynamic topology graph structure representing the functional state of organs was constructed.

[0062] Identify multi-connection node sets and single-path-dependent connection paths for the coordinated recovery of organ function on the dynamic topology graph structure;

[0063] Based on the location information of the multi-connection node set and the single-path-dependent connection path, the real-time operating status data of various types of rehabilitation equipment and the spatiotemporal trajectory data of the operations performed by nursing staff are matched for functional targets.

[0064] By constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operations, we can identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network, and obtain the interaction spectrum of the equipment-nursing-organ network.

[0065] Based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, and combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

[0066] In one optional implementation, cross-modal fusion processing is performed on the multi-channel organ function monitoring signal data of the target patient to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and to construct a dynamic topology graph structure representing the organ functional state, including:

[0067] The different modal signals in the multi-channel organ function monitoring signal data are time-stamped and aligned. The initial signal data from the ECG monitoring channel, blood oxygen monitoring channel, respiratory monitoring channel and blood pressure monitoring channel are mapped to a unified time axis and defined as a modal signal respectively, so as to obtain a time-synchronized multimodal signal sequence.

[0068] Frequency domain decomposition is performed on each mode signal in the time-synchronized multimodal signal sequence to extract the spectral feature vector of each mode signal within a preset frequency band. The cross-correlation coefficient between the spectral feature vectors of different mode signals is calculated and constructed into a cross-modal spectral correlation matrix.

[0069] Based on the cross-correlation coefficients of the cross-modal spectrum correlation matrix for modal pairs that exceed a preset correlation threshold, the information transmission delay characteristics are determined by calculating the time difference between the arrival times of the signal peaks between the modal pairs, and the functional coupling strength between the corresponding organ subsystems of the modal pairs is defined according to the magnitude of the cross-correlation coefficients.

[0070] The different organ subsystems are defined as nodes in a dynamic topology graph structure. The functional coupling strength between the organ subsystems is defined as the edge weight connecting the nodes. The information transmission delay feature is defined as the delay attribute of the edge, thus constructing the dynamic topology graph structure.

[0071] Multi-channel organ function monitoring signal data originates from ECG, blood oxygen, respiration, and blood pressure monitoring channels. These four channels differ significantly in their physical nature, sampling frequency, and signal dimensions. ECG signals are typically measured in millivolts with sampling rates between 250Hz and 1000Hz; blood oxygen signals are expressed as percentage saturation with relatively low sampling rates; respiration signals reflect the frequency of chest wall fluctuations; and blood pressure signals record the periodic changes in diastolic and systolic blood pressure in millimeters of mercury. Because the clock references of the acquisition devices for each channel are not entirely consistent, directly comparing signals from different channels introduces systematic time bias. Therefore, it is necessary to first align the timestamps of the signals from each channel.

[0072] Using the most accurate clock source in the acquisition system as the reference time axis, the local timestamps of each channel signal are mapped to this unified time axis through linear interpolation or resampling. For channels with inconsistent sampling rates, anti-aliasing low-pass filtering that preserves the signal spectral characteristics is used before resampling, ensuring that all channel signals have the same time resolution on the unified time axis. After alignment, the ECG monitoring channel, pulse oximetry monitoring channel, respiration monitoring channel, and blood pressure monitoring channel each correspond to a modal signal, and the four modal signals together constitute a time-synchronized multimodal signal sequence. This step is a prerequisite for subsequent cross-modal spectral analysis; any residual time offset will cause systematic errors in the cross-correlation calculation results, thus affecting the accuracy of functional coupling strength.

[0073] Short-time Fourier transform or wavelet transform is used to convert the time-domain signal into a time-frequency representation. To meet the practical needs of organ function analysis, the preset frequency band typically covers the extremely low frequency range (0.003Hz to 0.04Hz, corresponding to thermoregulation and endocrine rhythms), the low frequency range (0.04Hz to 0.15Hz, corresponding to sympathetic regulation), and the high frequency range (0.15Hz to 0.4Hz, corresponding to parasympathetic and respiratory regulation). Within each preset frequency band, statistical quantities such as power spectral density, band energy proportion, and spectral centroid are extracted from the signal. These statistical quantities are then concatenated to form the spectral feature vector of that modal signal within that frequency band. After generating corresponding spectral feature vectors for each of the four modal signals, the cross-correlation coefficient between the spectral feature vectors of any two modal signals is calculated.

[0074] The cross-correlation coefficient is calculated based on the Pearson correlation or normalized cross-correlation method, assuming the mode... The spectral eigenvector is Modality The spectral eigenvector is The cross-relationship between the two This reflects the degree of linear correlation between the two organ subsystems in the frequency domain. The cross-correlation coefficients of all mode pairs... Arranged by modal number, construct a cross-modal spectral correlation matrix. The first in the matrix Line number The elements of a column are modalities. With mode cross-relationships between Because the cross-correlation coefficients are symmetric, It is a symmetric matrix with diagonal elements of 1.

[0075] Cross-modal spectral correlation matrix After the structure is built, filter out the number of cross-relationships. Exceeding the preset association threshold The modal pairs. Preset association threshold. The value of needs to be determined based on the specific clinical scenario, and is usually set between 0.5 and 0.7 to exclude weak correlations or spurious associations caused by noise. For those meeting the criteria... The modal pairs are then further analyzed to calculate the information transmission delay characteristics between them. The method for determining the information transmission delay is as follows: in the time-synchronized multimodal signal sequence, the modes are detected separately... Signals and modes Calculate the time difference between the arrival times of the signal peaks and the arrival times of the two peaks. To improve the robustness of time delay estimation, the median of the time differences over multiple consecutive cardiac or respiratory cycles can be taken to obtain a stable information transmission time delay characteristic. The positive or negative direction of the time delay reflects the directionality of information transmission: if... This indicates the mode Functional changes in corresponding organ subsystems precede modal changes. Occurs in the corresponding organ subsystem; if If the direction is reversed, then the direction is reversed.

[0076] The definition of functional coupling strength is directly based on the cross-correlation coefficient. The numerical value. The closer it is to 1, the closer the functional linkage between the two organ subsystems and the higher the coupling strength. The closer to the preset association threshold The greater the coefficient of coupling, the weaker the coupling strength. In practical applications, the functional coupling strength can be directly taken as [value missing]. It itself, or after normalization mapping, is mapped to The interval is used to ensure consistent dimensionality in subsequent graph structures.

[0077] The dynamic topology graph structure is constructed using different organ subsystems as nodes. The ECG monitoring channel corresponds to the cardiac subsystem node, the blood oxygen monitoring channel corresponds to the pulmonary circulation and blood oxygen regulation subsystem node, the respiration monitoring channel corresponds to the respiratory system node, and the blood pressure monitoring channel corresponds to the vascular and circulatory regulation subsystem node. For the number of cross-relationships... Exceeding the preset association threshold For modal pairs, establish an edge between the corresponding two nodes to increase the functional coupling strength. Assigning the edge weight to the edge will reduce the information transmission delay. Assign the time delay attribute to this edge. For cross-correlation coefficients not exceeding the threshold... The modal pairs, where no edge connections are established between corresponding nodes, indicate that there is no significant functional relationship between the two organ subsystems within the current time window.

[0078] As the patient's recovery progresses, the statistical characteristics of the monitoring signals from each channel will change, affecting the cross-modal spectral correlation matrix. The values ​​of each element will also be updated accordingly. By setting a sliding time window, the calculation is recalculated at fixed time intervals. It also updates the weights and time delay attributes of each edge in the graph structure, enabling the dynamic topology graph structure to reflect the current state of the patient's organ function network in real time. When the weight of a certain edge... From below the threshold Become above the threshold When an edge is added to the graph, it indicates a new functional coupling relationship between the two organ subsystems; conversely, when the edge is removed, it indicates that the original coupling relationship has weakened or disappeared. This dynamic update mechanism enables subsequent analysis of multi-connected node sets and single-path-dependent connection paths in the graph structure based on the latest organ functional state, ensuring the timeliness and accuracy of the collaborative control strategy.

[0079] In one optional implementation, identifying the set of multiple-connection nodes and single-path-dependent connection paths for collaborative organ function recovery on the dynamic topology graph structure includes:

[0080] Traverse all nodes in the dynamic topology graph structure, count the number of edges directly connected to the current node as the connectivity value of the current node, and obtain the connectivity distribution of all nodes in the dynamic topology graph structure.

[0081] Calculate the statistical distribution characteristic parameters of the connectivity values ​​of all nodes based on the connectivity distribution, and filter the nodes whose connectivity values ​​exceed the statistical distribution characteristic parameters as candidate multi-connection nodes;

[0082] For each candidate multi-connection node, the collaborative recovery contribution of the candidate multi-connection node in the organ function information transmission network is calculated based on the information transmission delay characteristics and edge weights corresponding to the edges connected to the candidate multi-connection node. Candidate multi-connection nodes whose collaborative recovery contribution exceeds a preset contribution threshold are determined as nodes in the set of multi-connection nodes.

[0083] For any two nodes in the dynamic topology graph structure, by searching all reachable paths connecting the two nodes, identify the node pairs with a total number of reachable paths equal to one, and mark the unique reachable path as the single-path-dependent connection path.

[0084] After obtaining the dynamic topology graph structure, it is necessary to identify two types of structural features that are crucial for the coordinated recovery of organ function: sets of nodes with multiple connections and single-path-dependent connection paths. These two types of structures correspond to hub nodes and vulnerable transmission channels in the organ function network, respectively, and serve as the topological basis for subsequent targeted configuration of equipment and nursing resources.

[0085] In a dynamic topology graph structure, each node corresponds to an organ subsystem or functional monitoring channel. Edges between nodes represent the functional coupling relationships between organs, and the weight of each edge is determined by the strength of this functional coupling. The process involves traversing all nodes in the graph and counting the number of edges directly connected to each node. This count is defined as the connectivity degree of the current node, denoted as . ,in Number the nodes. After completing the traversal, the connectivity distribution sequence of all nodes in the graph is obtained. ,in The number represents the total number of nodes in the graph. The connectivity distribution reflects the topological position of each organ subsystem in the functional network. Nodes with high connectivity usually undertake the task of converging and forwarding functional information from multiple organs, and have a stronger coordinating role in the rehabilitation process.

[0086] After obtaining the connectivity distribution, the statistical distribution characteristics of the connectivity values ​​of all nodes are calculated. Specifically, the mean of the connectivity distribution is calculated. with standard deviation ,by As the filtering threshold, This is a preset standard deviation factor, which can be adjusted according to the sensitivity requirements of hub nodes in clinical scenarios. A typical value range is between 1 and 2. Connectivity value. Nodes exceeding this threshold are selected as candidate multi-connection nodes, forming a candidate set. This selection strategy adaptively determines the threshold based on statistical distribution, avoiding the problem of insufficient adaptability of fixed thresholds for different patients or different stages of rehabilitation. For patients with relatively sparse organ functional network structures, the mean and standard deviation are both small, and the threshold is lowered accordingly, ensuring that a sufficient number of candidate nodes are selected; for patients with relatively dense functional network connections, the threshold is increased accordingly to ensure that the selected candidate nodes have true topological hub significance.

[0087] For each candidate multi-connection node, its contribution to collaborative recovery in the organ function information transmission network is further calculated. The contribution to collaborative recovery considers two dimensions: first, the information transmission delay characteristics corresponding to each edge connected to the candidate node; the smaller the delay, the more timely the functional information interaction between the node and adjacent organs, and the stronger the promoting effect on collaborative recovery; second, the edge weights; the larger the edge weights, the higher the corresponding functional coupling strength, and the stronger the node's integration ability in the functional network. For candidate multi-connection nodes... Let its set of neighboring nodes be . , with adjacent nodes The edge weights between them are The corresponding information transmission delay is Then the contribution of collaborative recovery It can be represented as: The physical meaning of this formula is that connections with higher edge weights and lower latency accumulate greater contributions. The latency term is normalized by adding one to avoid singular values ​​when the latency is zero. The calculated collaborative recovery contribution is then used. Compared with the preset contribution threshold Compare and satisfy The candidate multi-connection nodes are ultimately determined as nodes in the set of multi-connection nodes. A preset contribution threshold is used. It can be set based on statistical data of historical patient groups or the experience of clinical experts, and can also be adaptively updated according to the dynamic changes in the patient's functional status during the rehabilitation process. Through this two-stage screening mechanism, the resulting set of multi-connected nodes not only meets the high connectivity requirements at the topological level, but also has significant advantages in the timeliness and intensity of functional information transmission, and can accurately locate the core hubs for the coordinated recovery of organ function.

[0088] Single-path dependency connection path identification focuses on the connection relationship between any two nodes in a dynamic topology graph. For any pair of nodes in the graph... Search all reachable paths connecting these two nodes and count the total number of reachable paths. A path is considered reachable if and only if the nodes are... When the total number of reachable paths between nodes is equal to 1, the node pair is identified as a single-path dependent node pair, and its unique reachable path is marked as a single-path dependent connection path. In the actual search process, a depth-first search algorithm can be used to traverse the graph structure and record the nodes... Starting from the node If the path count is 1 after the search is completed, then the path is confirmed to be a single-path dependent connection path.

[0089] Single-path-dependent connections hold significant clinical importance in organ function networks. These connections lack redundant channels for functional information transmission between the two organ subsystems. If any node or edge on this path experiences functional degradation, it directly disrupts functional coordination between the two subsystems, creating a vulnerable link in the rehabilitation process. Therefore, identifying single-path-dependent connections can help predict rehabilitation risks in advance, providing a priority direction for nursing interventions and equipment deployment. In the subsequent functional target matching phase, nodes and edges involved in single-path-dependent connections will be given higher resource allocation priority, ensuring these vulnerable channels receive adequate equipment support and nursing coverage during rehabilitation.

[0090] The set of multiple connected nodes and single-path-dependent connections together constitute the key structural features in a dynamic topology graph. The former represents the hub region of the functional network, while the latter represents the vulnerable channel of the functional network. They are often complementary in spatial distribution: hub nodes are usually located at the intersection of multiple paths, while single-path-dependent connections often appear at the edge of the network or between relatively isolated organ subsystems. Combining these two types of structural features allows for a holistic understanding of the topological pattern of the organ functional network, providing a complete structured basis for the subsequent collaborative deployment of various types of rehabilitation equipment and nursing procedures.

[0091] In one optional implementation, based on the location information of the multi-connection node set and the single-path-dependent connection path, the functional target matching of the real-time operating status data of various types of rehabilitation equipment with the spatiotemporal trajectory data of the operations performed by nursing staff includes:

[0092] Extract the organ subsystem identifier corresponding to each node in the multi-connection node set and the organ subsystem identifier corresponding to the node connected by the single path dependency connection path, and construct a spatial distribution mapping table of organ functional target points.

[0093] The real-time operating status data includes the organ part identifiers of the device and the device output parameters. The organ part identifiers of the device are matched with the organ subsystem identifiers in the organ function target spatial distribution mapping table. Rehabilitation devices whose organ part identifiers of the device match the set of multiple connection nodes are identified as key devices, and rehabilitation devices whose organ part identifiers of the device match the single path-dependent connection path are identified as protective devices.

[0094] The spatiotemporal trajectory data includes the organ site identifier of the operation and the operation intensity parameter. The organ site identifier of the operation is matched with the organ subsystem identifier in the organ function target spatial distribution mapping table. Nursing operations in which the organ site identifier of the operation hits the set of multiple connected nodes are identified as critical nursing operations, and nursing operations in which the organ site identifier of the operation hits the single path-dependent connection path are identified as protective nursing operations.

[0095] Each node in the dynamic topology graph carries a unique organ subsystem identifier, derived from the source annotation of multi-channel organ function monitoring signals, such as the encoding of specific organ locations like the heart, liver, kidneys, and lungs. For single-path-dependent connection paths, both ends of the path and the nodes traversed along it also carry organ subsystem identifiers, reflecting the various organ subsystems involved in the functional transmission chain of the path. The organ subsystem identifiers of these two types of nodes are organized according to their positional relationship in the topology graph to form a spatial distribution mapping table of organ function target points. This mapping table uses the organ subsystem identifier as the index key, recording the category of each organ subsystem in the functional network: whether it belongs to a set of multiple-connection nodes, a single-path-dependent connection path, or a regular node. It also retains the relative spatial coordinate information of the organ subsystem in the topology graph for subsequent spatial overlay mapping.

[0096] The real-time operational status data of various types of rehabilitation equipment contains two key types of information: the organ site identification on which the equipment is acting and the equipment output parameters. The organ site identification, provided by the equipment configuration file or sensor positioning data, indicates which organ(s) the equipment is currently applying physical action to. For example, if a low-frequency electrical stimulation device is currently acting on the kidney area, its organ site identification would be the code corresponding to the kidney. The equipment output parameters include quantifiable operational indicators such as power, frequency, stimulation intensity, and duration of action. These parameters are used as quantitative inputs for subsequent construction of the equipment's physical action field.

[0097] The identification of the organ site targeted by the device is compared one by one with the spatial distribution mapping table of organ function targets to determine whether the identification matches an organ subsystem covered by a set of multiple connected nodes. If it matches, the rehabilitation device is marked as a critical device, meaning that the device's area of ​​action falls on a core node in the organ function network with high connectivity and a significant contribution to the overall functional recovery. Changes in its operating state will have a cascading effect on multiple adjacent organ subsystems. Therefore, in subsequent collaborative control strategies, it is necessary to prioritize ensuring its power stability and timing continuity. If the identification of the organ site targeted by the device matches an organ subsystem involved in a single-path-dependent connection path, the rehabilitation device is marked as a protective device. A single-path-dependent connection path means that this path is the only channel for functional information transmission between two organ subsystems. Once this channel is disturbed, the collaborative recovery of related organ functions will face the risk of interruption. Therefore, devices acting on such areas need to operate in a protective mode to avoid excessive physical stimulation causing additional damage to this path, while ensuring that the functional transmission on this path is not interfered with by the device's side effects.

[0098] The spatiotemporal trajectory data of nursing staff performing operations is collected in real time through wearable positioning devices, operation recording sensors, or nursing information systems. The spatiotemporal trajectory data includes the identification of the organ site to which the operation is applied and operation intensity parameters. The identification of the organ site to which the operation is applied is determined by the position of the nursing staff's hands, the contact point of the operation tool, or the operation site field in the medical record. The operation intensity parameters include quantitative indicators such as the pressure applied, the turning angle, the rate of drainage operation, and the duration of physical therapy.

[0099] The organ site identifier of the operation is matched with the spatial distribution mapping table of organ functional targets to determine the category of the organ subsystem involved in the operation. If the organ site identifier of the operation matches an organ subsystem covered by a set of multiple connected nodes, the nursing operation is marked as a critical nursing operation. The organ region corresponding to the critical nursing operation is a hub node in the functional network. The nurse's operation behavior in this region directly affects the functional state of multiple organ subsystems. Therefore, it is necessary to finely control its operation intensity parameters and allocate high-priority resource support in the collaborative control strategy. If the organ site identifier of the operation matches an organ subsystem involved in a single path-dependent connection path, the nursing operation is marked as a protective nursing operation. The goal of protective nursing operations is to maintain the functional transmission integrity of single path-dependent connection paths. The operation intensity parameters need to be strictly constrained within a safe threshold range to prevent nursing behavior from causing additional physiological stress to vulnerable single functional pathways.

[0100] After completing the above matching, the classification results of key equipment, protective equipment, key nursing operations, and protective nursing operations, along with their corresponding equipment output parameters and operation intensity parameters, are stored together to form a functional target matching result set. This result set provides a complete input data foundation for subsequently constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operations. This allows the effects of equipment and nursing interventions to be accurately located at the topological level of the organ functional network, thereby supporting the identification and analysis of synergistic enhancement regions and mutually inhibiting regions.

[0101] In practical applications, the same organ subsystem can be acted upon by multiple devices simultaneously, and may also involve multiple types of nursing operations. In such cases, it is necessary to aggregate all devices and operations that hit the same organ subsystem identifier, and determine whether the combined effect of the device output parameters and operation intensity parameters exceeds the current functional capacity of the organ subsystem. If multiple critical devices simultaneously act on the organ subsystem corresponding to the same multi-connection node, it is necessary to mark the node as having a multi-source superposition risk in the functional target matching result set. This allows the game theory framework to constrain and optimize resource allocation for this node when solving for device-nursing resource equilibrium, avoiding excessive intervention that could exacerbate organ functional stress responses.

[0102] In one optional implementation, by constructing a spatial superposition mapping mechanism between the physical field of the equipment and the influence domain of nursing operations, the synergistic enhancement regions and mutually inhibiting regions of equipment action and nursing intervention in the organ function network are identified, resulting in the interaction spectrum of the equipment-nursing-organ network, including:

[0103] Based on the output parameters of the key equipment and the organ location identification of the equipment, the spatial diffusion range of the physical field of the key equipment in the organ functional network is calculated.

[0104] Based on the operation intensity parameter and the organ site where the operation was applied, the spatial propagation range of the operation influence domain of the key nursing operation in the organ function network is calculated.

[0105] The spatial diffusion range of the key equipment and the spatial propagation range of the key nursing operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as key location collaboration areas.

[0106] The spatial diffusion range of the protective device and the spatial propagation range of the protective care operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as vulnerable path collaboration areas.

[0107] For each node in the critical location collaboration region and the vulnerable path collaboration region, the parameter compatibility between the device output parameters and the operation intensity parameters is calculated. When the parameter compatibility is higher than a preset compatibility threshold, the current node is marked as a node in the collaboration enhancement region. When the parameter compatibility is lower than the preset compatibility threshold, the node is marked as a node in the mutual inhibition region, thus obtaining the interaction spectrum of the device-care-organ network.

[0108] like Figure 2 As shown, the method includes:

[0109] After completing the functional target matching, it is necessary to further clarify the actual range of action of equipment intervention and nursing operations in the organ functional network, and identify the spatial superposition effect of the two. To this end, for each key device, based on its output parameters (including physical quantities such as output power, electromagnetic field strength, vibration frequency, and thermal radiation intensity) and the organ site it acts on, the spatial diffusion range of the device's physical field in the organ functional network is calculated. Specifically, the diffusion originates from the node corresponding to the organ site acted on by the device, and propagates to surrounding nodes according to the attenuation law of the physical field. The degree of attenuation is determined by the topological distance between nodes and the physiological tissue conduction coefficient. When the field strength at a node exceeds a preset effective field strength threshold, that node is included in the spatial diffusion range of the device. Different types of devices have different physical field characteristics. For example, the field diffusion of low-frequency electrical stimulation devices mainly propagates along neural pathways, while the field diffusion of ultrasound therapy devices is more characterized by the spherical attenuation of volume waves. Therefore, when calculating the diffusion range, the corresponding propagation model must be selected based on the device type.

[0110] For critical nursing procedures, the spatial propagation range of the operation's influence domain within the organ function network is calculated based on operation intensity parameters (including quantitative indicators such as force applied, frequency of operation, duration, and temperature of hot compress) and the organ site where the operation is applied. The calculation method for the operation influence domain is similar to that of a physical action field, starting from the node corresponding to the operation site and expanding along the edges of the organ function network towards adjacent nodes. The propagation intensity decreases as the number of topological hops increases. When the operation influence intensity at a node exceeds a preset effective threshold, that node is included in the spatial propagation range of the nursing operation. The propagation of the nursing operation's influence domain also needs to consider the differences in operation types. For example, the expansion paths of the influence domain of a patient repositioning operation on abdominal organs and on peripheral circulation differ significantly, requiring the selection of appropriate propagation weight coefficients based on the nursing operation type identifier.

[0111] After calculating the spatial diffusion range of critical equipment and the spatial propagation range of critical nursing procedures separately, they are spatially superimposed on a dynamic topology graph. The superposition operation uses nodes as the basic unit. For each node in the dynamic topology graph, it is determined whether it simultaneously belongs to both the spatial diffusion range of critical equipment and the spatial propagation range of critical nursing procedures. If a node satisfies both conditions, it is marked as a spatially overlapping node. The set of all spatially overlapping nodes constitutes the critical location synergy region. The physical meaning of the critical location synergy region is that the organ function nodes within this region are simultaneously covered by both the physical field of critical equipment and the influence domain of critical nursing procedures, representing the core spatial region where equipment intervention and nursing procedures can potentially produce synergistic effects.

[0112] Similarly, for protective equipment and protective nursing procedures, the spatial diffusion range of the protective equipment and the spatial propagation range of the protective nursing procedures are calculated separately, and then spatially superimposed on the dynamic topology structure. The set of nodes that simultaneously belong to both coverage areas is identified and marked as vulnerable path collaborative regions. Vulnerable path collaborative regions correspond to organ function nodes on single-path-dependent connection paths. These nodes play an irreplaceable role in function transmission within the network. The fact that both protective equipment and protective nursing procedures jointly cover this region signifies that the dual protective intervention for vulnerable paths is spatially superimposed. Further evaluation is needed to determine whether this superposition produces a synergistic protective effect or carries the risk of over-intervention.

[0113] After identifying the critical location synergy region and the vulnerable path synergy region, the parameter compatibility between the device output parameters and the operational intensity parameters is calculated for each node in both regions. The parameter compatibility calculation comprehensively considers the superposition of the two types of parameters in terms of physical dimensions, the consistency of their action directions, and the degree of temporal synchronization. A higher parameter compatibility value is assigned when the device output parameters and operational intensity parameters are consistent in their action directions (e.g., both promote organ blood perfusion or both inhibit inflammatory responses) and have effective temporal overlap; a lower parameter compatibility value is assigned when their action directions are opposite (e.g., the device applies heat stimulation while the nursing procedure applies cold compresses) or when there is a significant temporal misalignment.

[0114] The device output parameters are denoted as The operational intensity parameter is denoted as The normalization direction consistency coefficient between the two is denoted as... The temporal overlap rate is denoted as Then the parameter compatibility at the node It can be represented as ,in For the normalized parameter amplitude matching term, the value ranges from 0 to 1. and When the amplitudes are similar, this term approaches 1; the greater the difference in amplitudes, the smaller this term becomes. When the directions of action are the same, a positive value is taken; when the directions are opposite, a negative value or a zero value is taken, so that parameter compatibility can distinguish between synergistic and inhibitory situations.

[0115] Compatibility of the calculated parameters Compatibility threshold with preset Comparison. When Higher than When this occurs, it indicates that the equipment output parameters and operational intensity parameters have good physical compatibility and temporal coordination at this node. Equipment intervention and nursing operations at this node can produce a synergistic enhancement effect, and this node is marked as a node in the synergistic enhancement region. Below If the two types of parameters conflict in direction or are out of chronological order at this node, the effect of the equipment and the nursing intervention weaken each other or even produce antagonistic effects at this node, and this node is marked as a node in the mutual inhibition region.

[0116] After traversing all nodes in the key location collaboration region and the vulnerable path collaboration region and completing the above marking, the sets of nodes in the collaboration enhancement region and the mutual inhibition region, along with the parameter compatibility values, region types, associated equipment identifiers, and associated nursing operation identifiers corresponding to each node, are integrated into an interaction spectrum of the device-nursing-organ network. The interaction spectrum uses a dynamic topology graph as its carrier, overlaying spatial coverage information of devices and nursing operations, parameter compatibility assessment results, and regional functional attribute annotations onto the graph structure. It can intuitively present the specific locations where the current device configuration and nursing plan produce synergistic enhancement effects in the organ function network, as well as the distribution of nodes with mutual inhibition risks, providing accurate spatial constraint inputs for subsequent resource equilibrium solutions based on a game theory framework. The dynamic update frequency of the interaction spectrum is synchronized with the update frequency of the dynamic topology graph, ensuring that the distribution of collaboration enhancement regions and mutual inhibition regions can reflect the latest organ function network state in real time when the patient's physiological state changes, thereby supporting the continuous optimization and adjustment of collaborative control and nursing management strategies.

[0117] In one optional implementation, based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, combined with patient physiological stress response data and physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

[0118] Extract the node identifiers of the cooperative enhancement region and the node identifiers of the mutually inhibiting region in the interaction spectrum, and statistically analyze the spatial distribution density of the node identifiers of the cooperative enhancement region and the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure.

[0119] The patient physiological stress response data includes the patient's physiological tolerance threshold to the device output parameters and the patient's physiological tolerance threshold to the operation intensity parameters; the physical field distribution data of the rehabilitation environment includes the physical field intensity distribution at different spatial locations in the rehabilitation environment.

[0120] A mechanism for solving the equipment-nursing resource equilibrium based on a game theory framework is constructed. The equilibrium solution of the game theory framework is solved to obtain the optimal configuration values ​​of equipment output parameters and operation intensity parameters.

[0121] Based on the optimized configuration values ​​of the device output parameters and the optimized configuration values ​​of the operation intensity parameters, a device power allocation sequence, a device action timing arrangement scheme, a nursing operation space positioning instruction, and a nursing staff skill requirement matching table are generated and combined into a collaborative control and nursing management strategy.

[0122] The node identifier sets corresponding to the cooperative enhancement regions and the mutual inhibition regions are extracted from the interaction spectrum. For each type of region, the spatial distribution density of each node identifier is statistically analyzed in the coordinate space of the dynamic topology graph structure. Specifically, using the functional coordinate system of the dynamic topology graph structure as a reference, the graph space is divided into several local sub-regions. The number of cooperative enhancement nodes and the number of mutual inhibition nodes falling into each sub-region are normalized and statistically analyzed to obtain the spatial density distribution map of the cooperative enhancement region. Spatial density distribution map of mutually inhibiting regions Density difference This reflects the net demand intensity for equipment and nursing intervention resources at different locations within the organ function network: Regions with positive values ​​should be prioritized for allocation of more intensive collaborative intervention resources. For areas with negative values, the intensity of intervention needs to be reduced or the intervention method adjusted to avoid further inhibiting the recovery of organ function.

[0123] The patient physiological stress response data includes two key constraint parameters: the patient's physiological tolerance threshold to the device output parameters. Physiological tolerance threshold of the patient to the intensity parameters of the procedure These two thresholds were obtained through statistical analysis of patients' historical physiological monitoring data, specifically including response curves of stress response indicators such as heart rate variability, cortisol secretion levels, and blood pressure fluctuations under different intervention intensities. The critical value of the intervention intensity that elicits a significant stress response was extracted as the upper bound of the tolerance threshold. The physical field distribution data of the rehabilitation environment describes the physical field quantities such as electromagnetic field intensity, sound field intensity, and thermal field intensity at different locations in the rehabilitation space. The values ​​of each spatial location were recorded. The overall physical field strength at that location is The distribution of physical field intensity affects the spatial attenuation characteristics of equipment performance. Equipment efficiency is higher in high-field-intensity regions than in low-field-intensity regions. Therefore, resource allocation should consider this factor. It is incorporated into the optimization calculation as a spatial weighting factor.

[0124] Based on the above information, a game theory-based mechanism for resolving the equipment-nursing resource equilibrium is constructed. The set of various types of rehabilitation equipment and the nursing staff are modeled as two types of players in a game. The strategy space of the equipment players is the output parameter configuration vector of each device, and the strategy space of the nursing players is the operational intensity configuration vector of each nursing staff member. The payoff function for the equipment players is... Defined as: meeting the patient's physiological tolerance constraints Under the premise of maximizing the density-weighted cumulative value of collaborative enhancement at each device's action node, a penalty term is applied to the action contribution of devices falling into the mutual inhibition region, and the penalty weight is proportional to... The local density is proportional to the benefit function of the nursing stakeholders. Defined as: meeting the patient's physiological tolerance constraints Under the premise of maximizing the spatial range of the synergistic enhancement area covered by nursing operations, while imposing penalties on nursing operations that fall into mutually inhibiting areas. The strategy choices of the two types of participants influence each other: adjustments to the equipment output parameters will change the spatial distribution of the physical field, thereby affecting the effective range of nursing operations; changes in the intensity of nursing operations will cause changes in the patient's local physiological state, thereby affecting the actual effect of the equipment.

[0125] The game equilibrium solution employs the Nash equilibrium framework, seeking strategy combinations that satisfy the following condition: given that the nursing participant's strategy remains unchanged, the equipment participant cannot further improve its performance by unilaterally adjusting its strategy. Given that the strategies of the participating devices remain unchanged, nursing participants cannot further improve their services by unilaterally adjusting their strategies. Since the payoff function includes spatial density weighting and physical field intensity correction terms, directly obtaining an analytical solution is difficult. Therefore, an iterative optimal response algorithm is used for numerical solution. In each iteration, the current strategy of the nursing participant is fixed, and the payoff function of the equipment participant is updated using gradient ascent to obtain the updated values ​​of the equipment output parameters. Subsequently, the update strategy of the equipment participant is fixed, and the payoff function of the nursing participant is updated using gradient ascent to obtain the updated values ​​of the operational intensity parameters. The iteration process continues until the policy changes of both types of participants are lower than the preset convergence threshold. At this point, the Nash equilibrium solution is obtained, which is the optimal configuration value of the device output parameters. Optimized configuration values ​​of operational intensity parameters .

[0126] During the solution process, the physical field strength The contribution values ​​of each node in the equipment benefit function are adjusted using spatial weighting coefficients, which appropriately amplifies the equipment intervention benefits at locations with higher physical field intensity. This guides the optimization results to concentrate the deployment of high-intensity equipment interventions in spatial locations with better physical field conditions. Simultaneously, the patient's physiological tolerance threshold serves as a hard constraint. After each iteration update, the strategy vector is projected to ensure that the optimized configuration value never exceeds the upper tolerance bound, thus guaranteeing patient safety.

[0127] Based on the obtained and This further generates four specific collaborative control and nursing management strategy outputs. The equipment power allocation sequence, based on the optimized output parameter configuration values ​​of each device, combined with the device's rated power range and energy consumption constraints, will... The output parameters of each device are converted into a specific power value sequence, arranged according to the time axis of the rehabilitation cycle, forming a power change time series table for each device throughout the rehabilitation process. Based on the power allocation sequence, the device action timing arrangement scheme, according to the spatial density distribution of the synergistic enhancement region and the topological constraints of single-path-dependent connection paths in the organ function network, arranges the start and end times of multiple devices to avoid multiple devices simultaneously exerting high intensity on mutually inhibiting nodes, ensuring that the timing coordination between devices matches the recovery rhythm of the organ function network. Nursing operation spatial positioning commands are based on... The optimized intensity configuration values ​​for each nursing procedure, combined with the positional mapping relationship of the collaborative enhancement area nodes in physical space, generate specific location coordinates and directional guidance for nurses to perform procedures on the patient's body surface or bedside space. This ensures that nursing interventions are precisely positioned at the body surface projection location corresponding to the collaborative enhancement area. The nurse skill requirement matching table, based on the optimized intensity configuration values ​​and spatial positioning requirements of each nursing procedure, matches each procedure with nurses possessing the corresponding skill level according to the nurses' skill profile database. It outputs the corresponding allocation relationship between nurses and procedures, ensuring that high-intensity or high-precision procedures are undertaken by nurses with matching skill levels. The above four types of outputs combine to form a complete collaborative control and nursing management strategy for the rehabilitation management system to schedule and execute.

[0128] In one optional implementation, a device-nursing resource equilibrium solution mechanism based on a game theory framework is constructed. Solving for the equilibrium solution within the game theory framework yields optimized configuration values ​​for device output parameters and operational intensity parameters, including:

[0129] The key equipment, the protective equipment, the key nursing operation, and the protective nursing operation are taken as game participants, and the equipment output parameters and the operation intensity parameters are taken as game strategy variables.

[0130] The functional gain weight is calculated based on the spatial distribution density of the node identifiers of the collaborative enhancement region in the dynamic topology graph structure, and the functional loss weight is calculated based on the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure. A game payoff function is constructed based on the functional gain weight and the functional loss weight.

[0131] The physiological tolerance threshold of the patient to the output parameters of the device and the physiological tolerance threshold of the patient to the operation intensity parameters are used as upper limits of the game strategy variables, and the distribution of physical field intensity at different spatial locations in the rehabilitation environment is used as the spatial range constraint of the device to construct the game constraint conditions.

[0132] Based on the game participants, the game strategy variables, the game payoff function, and the game constraints, the Nash equilibrium solution of the game theory framework is obtained, and the optimal configuration values ​​of the equipment output parameters and the operation intensity parameters are obtained.

[0133] In constructing the equipment-nursing resource equilibrium solution mechanism within a game theory framework, key equipment, protective equipment, key nursing operations, and protective nursing operations are incorporated into the game system as independent players. Specifically, key equipment refers to rehabilitation equipment with a direct functional mapping relationship to the synergistic enhancement region in the interaction spectrum; protective equipment refers to rehabilitation equipment whose scope of action covers mutually inhibiting regions and whose output intensity needs to be constrained; key nursing operations refer to operation types performed by nursing staff whose spatiotemporal trajectories fall within the synergistic enhancement region; and protective nursing operations refer to nursing intervention behaviors that need to avoid causing additional incentives to the mutually inhibiting regions. The equipment output parameters are then considered. With operational strength parameters As the strategy variables of each game participant, each participant independently chooses a strategy within its own strategy space in order to maximize its own payoff function value.

[0134] The construction of the game payoff function relies on the spatial distribution density information of co-enhancing and mutually inhibiting regions within the dynamic topology graph structure. For co-enhancing regions, the spatial clustering degree of their node identifiers within the dynamic topology graph structure is statistically analyzed, and the spatial density distribution map of the co-enhancing regions is generated. As the basis for calculating functional gain weights, the denser the node distribution, the more significant the contribution of that region to the coordinated recovery of organ function, and the higher the corresponding functional gain weight. The higher the density distribution, the better. Similarly, for mutually inhibiting regions, the spatial density distribution map of the mutually inhibiting regions is used. Calculate the functional loss weights The more concentrated the distribution of nodes in the inhibition region, the stronger the mutual inhibition effect between the equipment and nursing intervention, and the greater the corresponding functional loss weight.

[0135] Revenue function of equipment participants It consists of a function gain term and a function loss penalty term. The function gain term reflects the device's output parameters within the cooperative enhancement region. The organ function recovery benefits that can be activated, and the functional gain weight. Positive correlation; the function loss penalty term reflects the cost of increased inhibition effect caused by excessively high output parameters within the mutual inhibition region, and is related to the function loss weight. and equipment output parameters The projection intensity is positively correlated with the inhibition region. The payoff function of the nursing stakeholders. Employing a symmetrical structure, the functional gain term and the operational intensity parameter of the nursing operation within the synergistic enhancement area are used. and functional gain weight Relatedly, the functional loss penalty item is related to the intensity of the nursing operation and the functional loss weight within the mutually inhibiting area. Related. By measuring the difference between synergistic enhancement density and mutual inhibition density. By introducing a weight adjustment term in the payoff function, when the density of synergistic enhancement is significantly higher than that of mutual inhibition, the payoff function gives a higher weight to the gain term, thereby guiding game participants to concentrate resources in the synergistic enhancement region; when the difference between the two is small or even reversed, the payoff function automatically increases the loss penalty weight to inhibit excessive investment by participants in the inhibition region.

[0136] The construction of game-theoretic constraints includes two types of constraints. The first type is the physiological tolerance threshold constraint: setting an upper bound on the patient's physiological tolerance threshold to the device's output parameters. As a device policy variable The upper limit, that is ; set the upper limit of the patient's physiological tolerance threshold to the procedural intensity parameter. As a nursing strategy variable The upper limit, that is This constraint ensures that during the game-theoretic process, no equilibrium solution will exceed the patient's physiological tolerance, fundamentally guaranteeing patient safety. The second type is the constraint on the spatial range of equipment: this involves different spatial locations within the rehabilitation environment... Comprehensive physical field strength at the location This serves as a boundary condition for equipment spatial deployment and power allocation. Specifically, it refers to the spatial location of the equipment. The actual intensity of the action at a location must not exceed the comprehensive physical field intensity at that location. Exceeding the preset physical field safety threshold prevents local physical field overload from harming the patient or rehabilitation environment. These two types of constraints together constitute the feasible region of the game's strategy space, and the search process for the Nash equilibrium solution is strictly confined to this feasible region.

[0137] After fully defining the game participants, strategy variables, payoff functions, and constraints, an iterative optimal response algorithm is used to solve for the Nash equilibrium. In each iteration, the equipment participant, under the condition of fixing the current strategy of the nursing participant, solves the optimal response within the feasible region to achieve the desired Nash equilibrium. Maximize the optimal device output parameters; under the condition of fixed device participant's current strategy, solve within the feasible region to maximize the optimal device output parameters. The optimal operational strength parameter is maximized. The two participants alternately update their policies until the change in policy variables in two consecutive iterations is less than the iteration convergence threshold. The game is determined to have reached a Nash equilibrium. At this point, the optimized configuration values ​​of the device output parameters are output. Optimized configuration values ​​for operational intensity parameters This is the Nash equilibrium solution, which means that under the current game framework and constraints, no single participant can further increase its own gains by unilaterally changing its strategy, thereby achieving a global equilibrium allocation of equipment and nursing resources in the organ function network.

[0138] and The physical meanings correspond to the power distribution sequence of various rehabilitation devices within different time windows, and the intensity level of caregivers performing operations in different spatial locations. Mapping to the device action timing scheme allows for the determination of the start-up, operation, and pause times for each device, enabling the devices to create a temporal superposition effect within the synergistic enhancement region; Mapping these commands to the spatial positioning of nursing operations provides nurses with precise references for operational locations and intensity, enabling complementary coverage of nursing interventions and equipment effects in space. Through an equilibrium-solving mechanism within a game theory framework, the allocation of equipment and nursing resources no longer relies on empirical judgments but is based on the topological structure and physical field distribution of the organ function network. This maximizes synergistic benefits and minimizes mutual inhibition losses while meeting the patient's physiological safety constraints, providing a quantitative parameter basis for the subsequent generation of complete collaborative control and nursing management strategies.

[0139] A second aspect of the present invention provides a multi-device collaborative control and nursing management system for organ rehabilitation, comprising:

[0140] The topology building unit is used to perform cross-modal fusion processing on the multi-channel organ function monitoring signal data of the target patient, extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, construct a dynamic topology graph structure that characterizes the organ function state, and identify the multi-connection node set and single-path-dependent connection path of organ function collaborative recovery on the dynamic topology graph structure.

[0141] The target matching unit is used to perform functional target matching between the real-time operating status data of multiple types of rehabilitation equipment and the spatiotemporal trajectory data of the operation performed by the nursing staff, based on the location information of the set of multiple connection nodes and the single path-dependent connection path.

[0142] The action mapping unit is used to identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network by constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operation, and to obtain the interaction spectrum of equipment-nursing-organ network.

[0143] The collaborative strategy unit is used to generate a collaborative control and nursing management strategy based on the distribution characteristics of the collaborative enhancement region and the mutual inhibition region in the interaction spectrum, combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

[0144] A third aspect of the present invention provides an electronic device, comprising:

[0145] processor;

[0146] Memory used to store processor-executable instructions;

[0147] The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0148] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0149] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0150] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for multi-device collaborative control and nursing management in organ rehabilitation, characterized in that, include: Cross-modal fusion processing was performed on multi-channel organ function monitoring signal data of target patients to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and a dynamic topology graph structure representing the functional state of organs was constructed. Identify multi-connection node sets and single-path-dependent connection paths for the coordinated recovery of organ function on the dynamic topology graph structure; Based on the location information of the multi-connection node set and the single-path-dependent connection path, the real-time operating status data of various types of rehabilitation equipment and the spatiotemporal trajectory data of the operations performed by nursing staff are matched for functional targets. By constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operations, we can identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network, and obtain the interaction spectrum of the equipment-nursing-organ network. Based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, and combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

2. The method according to claim 1, characterized in that, Cross-modal fusion processing is performed on multi-channel organ function monitoring signal data of target patients to extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, and to construct a dynamic topology graph structure representing the organ functional state, including: The different modal signals in the multi-channel organ function monitoring signal data are time-stamped and aligned. The initial signal data from the ECG monitoring channel, blood oxygen monitoring channel, respiratory monitoring channel and blood pressure monitoring channel are mapped to a unified time axis and defined as a modal signal respectively, so as to obtain a time-synchronized multimodal signal sequence. Frequency domain decomposition is performed on each mode signal in the time-synchronized multimodal signal sequence to extract the spectral feature vector of each mode signal within a preset frequency band. The cross-correlation coefficient between the spectral feature vectors of different mode signals is calculated and constructed into a cross-modal spectral correlation matrix. Based on the cross-correlation coefficients of the cross-modal spectrum correlation matrix for modal pairs that exceed a preset correlation threshold, the information transmission delay characteristics are determined by calculating the time difference between the arrival times of the signal peaks between the modal pairs, and the functional coupling strength between the corresponding organ subsystems of the modal pairs is defined according to the magnitude of the cross-correlation coefficients. The different organ subsystems are defined as nodes in a dynamic topology graph structure. The functional coupling strength between the organ subsystems is defined as the edge weight connecting the nodes. The information transmission delay feature is defined as the delay attribute of the edge, thus constructing the dynamic topology graph structure.

3. The method according to claim 1, characterized in that, Identifying multi-connected node sets and single-path-dependent connection paths for coordinated organ function recovery on the dynamic topology graph structure includes: Traverse all nodes in the dynamic topology graph structure, count the number of edges directly connected to the current node as the connectivity value of the current node, and obtain the connectivity distribution of all nodes in the dynamic topology graph structure. Calculate the statistical distribution characteristic parameters of the connectivity values ​​of all nodes based on the connectivity distribution, and filter the nodes whose connectivity values ​​exceed the statistical distribution characteristic parameters as candidate multi-connection nodes; For each candidate multi-connection node, the collaborative recovery contribution of the candidate multi-connection node in the organ function information transmission network is calculated based on the information transmission delay characteristics and edge weights corresponding to the edges connected to the candidate multi-connection node. Candidate multi-connection nodes whose collaborative recovery contribution exceeds a preset contribution threshold are determined as nodes in the set of multi-connection nodes. For any two nodes in the dynamic topology graph structure, by searching all reachable paths connecting the two nodes, identify the node pairs with a total number of reachable paths equal to one, and mark the unique reachable path as the single-path-dependent connection path.

4. The method according to claim 1, characterized in that, Based on the location information of the multi-connection node set and the single-path-dependent connection path, the real-time operating status data of various types of rehabilitation equipment and the spatiotemporal trajectory data of the operations performed by nursing staff are matched for functional targets, including: Extract the organ subsystem identifier corresponding to each node in the multi-connection node set and the organ subsystem identifier corresponding to the node connected by the single path dependency connection path, and construct a spatial distribution mapping table of organ functional target points. The real-time operating status data includes the organ part identifiers of the device and the device output parameters. The organ part identifiers of the device are matched with the organ subsystem identifiers in the organ function target spatial distribution mapping table. Rehabilitation devices whose organ part identifiers of the device match the set of multiple connection nodes are identified as key devices, and rehabilitation devices whose organ part identifiers of the device match the single path-dependent connection path are identified as protective devices. The spatiotemporal trajectory data includes the organ site identifier of the operation and the operation intensity parameter. The organ site identifier of the operation is matched with the organ subsystem identifier in the organ function target spatial distribution mapping table. Nursing operations in which the organ site identifier of the operation hits the set of multiple connected nodes are identified as critical nursing operations, and nursing operations in which the organ site identifier of the operation hits the single path-dependent connection path are identified as protective nursing operations.

5. The method according to claim 4, characterized in that, By constructing a spatial superposition mapping mechanism between the physical field of equipment and the influence domain of nursing operations, the synergistic enhancement and mutual inhibition regions of equipment action and nursing intervention in the organ function network are identified, resulting in the interaction spectrum of the equipment-nursing-organ network, including: Based on the output parameters of the key equipment and the organ location identification of the equipment, the spatial diffusion range of the physical field of the key equipment in the organ functional network is calculated. Based on the operation intensity parameter and the organ site where the operation was applied, the spatial propagation range of the operation influence domain of the key nursing operation in the organ function network is calculated. The spatial diffusion range of the key equipment and the spatial propagation range of the key nursing operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as key location collaboration areas. The spatial diffusion range of the protective device and the spatial propagation range of the protective care operation are spatially superimposed on the dynamic topology structure, and the set of nodes that overlap in space are identified as vulnerable path collaboration areas. For each node in the critical location collaboration region and the vulnerable path collaboration region, the parameter compatibility between the device output parameters and the operation intensity parameters is calculated. When the parameter compatibility is higher than a preset compatibility threshold, the current node is marked as a node in the collaboration enhancement region. When the parameter compatibility is lower than the preset compatibility threshold, the node is marked as a node in the mutual inhibition region, thus obtaining the interaction spectrum of the device-care-organ network.

6. The method according to claim 4, characterized in that, Based on the distribution characteristics of the synergistic enhancement region and the mutual inhibition region in the interaction spectrum, and combined with patient physiological stress response data and physical field distribution data of the rehabilitation environment, a collaborative control and nursing management strategy is generated by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes device power allocation sequences, device action timing arrangement schemes, nursing operation spatial positioning instructions, and nursing staff skill requirement matching tables. Extract the node identifiers of the cooperative enhancement region and the node identifiers of the mutually inhibiting region in the interaction spectrum, and statistically analyze the spatial distribution density of the node identifiers of the cooperative enhancement region and the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure. The patient physiological stress response data includes the patient's physiological tolerance threshold to the device output parameters and the patient's physiological tolerance threshold to the operation intensity parameters; the physical field distribution data of the rehabilitation environment includes the physical field intensity distribution at different spatial locations in the rehabilitation environment. A mechanism for solving the equipment-nursing resource equilibrium based on a game theory framework is constructed. The equilibrium solution of the game theory framework is solved to obtain the optimal configuration values ​​of equipment output parameters and operation intensity parameters. Based on the optimized configuration values ​​of the device output parameters and the optimized configuration values ​​of the operation intensity parameters, a device power allocation sequence, a device action timing arrangement scheme, a nursing operation space positioning instruction, and a nursing staff skill requirement matching table are generated and combined into a collaborative control and nursing management strategy.

7. The method according to claim 6, characterized in that, A mechanism for resolving the equipment-nursing resource equilibrium based on a game theory framework is constructed. The equilibrium solution within the game theory framework is obtained, yielding optimized configuration values ​​for equipment output parameters and operational intensity parameters, including: The key equipment, the protective equipment, the key nursing operation, and the protective nursing operation are taken as game participants, and the equipment output parameters and the operation intensity parameters are taken as game strategy variables. The functional gain weight is calculated based on the spatial distribution density of the node identifiers of the collaborative enhancement region in the dynamic topology graph structure, and the functional loss weight is calculated based on the spatial distribution density of the node identifiers of the mutually inhibiting region in the dynamic topology graph structure. A game payoff function is constructed based on the functional gain weight and the functional loss weight. The physiological tolerance threshold of the patient to the output parameters of the device and the physiological tolerance threshold of the patient to the operation intensity parameters are used as upper limits of the game strategy variables, and the distribution of physical field intensity at different spatial locations in the rehabilitation environment is used as the spatial range constraint of the device to construct the game constraint conditions. Based on the game participants, the game strategy variables, the game payoff function, and the game constraints, the Nash equilibrium solution of the game theory framework is obtained, and the optimal configuration values ​​of the equipment output parameters and the operation intensity parameters are obtained.

8. A multi-device collaborative control and nursing management system for organ rehabilitation, used to implement the method as described in any one of claims 1-7, characterized in that, include: The topology building unit is used to perform cross-modal fusion processing on the multi-channel organ function monitoring signal data of the target patient, extract the functional coupling strength and information transmission delay characteristics between different organ subsystems, construct a dynamic topology graph structure that characterizes the organ function state, and identify the multi-connection node set and single-path-dependent connection path of organ function collaborative recovery on the dynamic topology graph structure. The target matching unit is used to perform functional target matching between the real-time operating status data of multiple types of rehabilitation equipment and the spatiotemporal trajectory data of the operation performed by the nursing staff, based on the location information of the set of multiple connection nodes and the single path-dependent connection path. The action mapping unit is used to identify the synergistic enhancement region and the mutual inhibition region of equipment action and nursing intervention in the organ function network by constructing a spatial superposition mapping mechanism between the physical action field of equipment and the influence domain of nursing operation, and to obtain the interaction spectrum of equipment-nursing-organ network. The collaborative strategy unit is used to generate a collaborative control and nursing management strategy based on the distribution characteristics of the collaborative enhancement region and the mutual inhibition region in the interaction spectrum, combined with the patient's physiological stress response data and the physical field distribution data of the rehabilitation environment, by constructing a device-nursing resource equilibrium solution mechanism based on a game theory framework. This strategy includes a device power allocation sequence, a device action timing arrangement scheme, nursing operation spatial positioning instructions, and a nursing staff skill requirement matching table.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images