An AI energy platform closed-loop control system for urinary endoluminal surgery

Abstract

This invention relates to the field of medical device technology, specifically disclosing a closed-loop control system for an AI energy platform used in urological surgery. By real-time acquisition of visual and bioimpedance data of the surgical area, a topological relationship graph of the surgical area is constructed, with tissue partitions as nodes and spatial adjacency and functional connectivity as edges. Key node groups representing critical blood vessels are identified, and by analyzing all associated paths from non-critical nodes to key node groups, the propagation risk of energy operation is calculated, generating quantified risk assessment parameters. These parameters are then fused with the risk assessment parameters of the target node, tissue type probability, and real-time bioimpedance change rate. Matching energy control commands are generated to drive a laser or plasma generator to execute precise energy output. Changes in tissue state after the action are captured and fed back, forming a closed-loop control. This invention achieves proactive quantitative protection of key anatomical structures and adaptive precise energy control, improving surgical safety and operational efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of medical devices, in particular to an AI energy platform closed-loop control system for urinary endoluminal surgery. BACKGROUND

[0002] Currently, in the endoluminal surgery of urology, the use of energy platforms (such as laser, plasma) has become very common. In order to improve the precision and safety of the operation, the existing technology begins to introduce artificial intelligence assistance, which identifies the tissue type by analyzing the endoscopic visual image or single modal sensing data (such as spectrum), and attempts to automatically adjust the energy output power to a certain extent according to the identification result. However, such systems usually simply weight and fuse different sensing data (such as vision, spectrum, bioimpedance) or analyze them independently, and their decision logic focuses on the classification of the tissue properties at the "current contact point", but lacks overall modeling and understanding of the spatial topological relationship and functional association between different tissue segments in the surgical area.

[0003] The existing technology has the following shortcomings:

[0004] In the endoluminal surgery of urology, the existing AI energy control system faces the dilemma of multi-modal sensing data conflict and clinical safety priority decision. Specifically, when the system judges that the current area is tumor tissue that needs to be removed based on visual and spectral data, it may also detect abnormal impedance signals due to the proximity of large blood vessels. The existing technology lacks a mechanism that can deeply integrate the global safety constraint of "protecting key blood vessel structures" beyond the local resection goal into the real-time control logic, resulting in the system making dangerous energy decisions that conform to the local identification logic but violate the overall surgical safety principle when data conflicts occur, and failing to autonomously avoid the potential damage risk to large blood vessels within the control cycle. SUMMARY

[0005] The purpose of the present application is to provide an AI energy platform closed-loop control system for urinary endoluminal surgery to solve the problems in the above background.

[0006] The purpose of the present application can be achieved by the following technical solutions:

[0007] An AI energy platform closed-loop control system for urinary endoluminal surgery, comprising:

[0008] a topological relationship graph construction module for constructing a topological relationship graph of the surgical area based on multi-modal sensing data containing vision and bioimpedance obtained from the surgical area in real time;

[0009] a key structure identification module for identifying a key node group corresponding to a preset key anatomical structure from all nodes based on the feature vectors of the nodes in the topological relationship graph;

[0010] The risk propagation analysis module is configured to receive a topological relationship graph containing a key node group identifier, and based on a topological structure of the corresponding topological relationship graph, calculate a potential influence of an energy operation applied to any non-key node on the key node group by analyzing an associated path connected via edges between the non-key node and the key node group, and generate a risk assessment parameter corresponding to each non-key node;

[0011] The safety control instruction generation module is configured to fuse the feature vector of the target node and the risk assessment parameter in real time for a target node on which the current surgical instrument acts, and generate an energy control instruction matched with the target node;

[0012] The closed-loop control execution module is configured to adjust laser or plasma parameters output in real time according to the energy control instruction, and act on a surgical region; a state change of the surgical region after the action is captured by multi-modal sensing data and fed back to the topological relationship graph construction module to form a closed-loop control.

[0013] As a further scheme of the present application, the surgical region topological relationship graph specifically comprises:

[0014] The surgical topological relationship graph is composed of nodes and edges connecting the nodes, the nodes correspond to physical partitions of the surgical region, each node has a feature vector fused based on multi-modal sensing data of the corresponding partition, and the edges are used to represent spatial adjacency association or functional connectivity association between the physical partitions corresponding to the nodes.

[0015] As a further scheme of the present application, the key node group corresponding to the preset key anatomical structure is identified from all nodes, specifically comprising:

[0016] Based on a preset key anatomical structure spectral feature threshold, the feature vectors of all nodes in the topological relationship graph are initially screened, and the nodes whose spectral feature vectors satisfy the threshold are marked as candidate key nodes;

[0017] The biological impedance dynamic change curve corresponding to the candidate key node is obtained, the periodic pulsation amplitude of the biological impedance dynamic change curve in a unit time is calculated, and the pulsation amplitude is compared with a preset blood vessel feature amplitude range to screen out nodes meeting the requirements to form a set of to-be-verified key nodes;

[0018] For each node in the set of to-be-verified key nodes, the feature vectors of all first-order adjacent nodes of the node in the topological relationship graph are extracted, a cooperative consistency parameter of the corresponding node and its adjacent nodes in the spectral and impedance features is calculated, and if the cooperative consistency parameter is higher than a preset cooperative threshold, the corresponding node is determined as a final key node, and the corresponding node is added to the key node group.

[0019] As a further scheme of the present application, the generation process of the risk assessment parameter specifically comprises:

[0020] For each non-critical node in the topological relationship graph, enumerate all associated paths from the non-critical node to all critical nodes in the critical node group, and the associated path is composed of a series of edges connected at both ends;

[0021] For each enumerated associated path, calculate a path attenuation factor representing the degree of attenuation of energy influence on the corresponding associated path according to the preset weight of each edge on the associated path and the total number of edges contained in the path;

[0022] Superimpose the path attenuation factors of different associated paths of the same target non-critical node to all critical nodes to obtain a comprehensive attenuation value, and map the risk assessment parameter according to the size of the comprehensive attenuation value.

[0023] As a further scheme of the present application, the calculation process of the path attenuation factor is:

[0024] According to the type of each edge constituting the associated path, an initial weight is set for each edge; wherein the edge connecting the spatial adjacent physical partitions is assigned a first weight value, and the edge connecting the physical partitions with functional connectivity is assigned a second weight value greater than the first weight value;

[0025] According to the type of each edge constituting the associated path, an initial weight is set for each edge; wherein the edge connecting the spatial adjacent physical partitions is assigned a first weight value, and the edge connecting the physical partitions with functional connectivity is assigned a second weight value greater than the first weight value;

[0026] Multiply the equivalent weights of all edges on the associated path, and then divide the multiplication result by the total number of edges contained in the corresponding associated path, and the quotient obtained is the path attenuation factor.

[0027] As a further scheme of the present application, the generation of the energy control instruction matched with the target node specifically includes:

[0028] Compare the received risk assessment parameter of the target node with the preset plurality of continuous safety interval thresholds to determine the safety level to which the risk assessment parameter belongs;

[0029] From the feature vector of the target node, the probability distribution representing the organization type and the real-time biological impedance change rate are parsed out;

[0030] Combine the safety level, the probability distribution of the organization type and the biological impedance change rate to form a multi-dimensional query index;

[0031] Based on the multi-dimensional query index, search and match in a pre-defined energy parameter mapping table to directly map an energy control instruction composed of a specific power value, an action time and an energy waveform mode.

[0032] As a further scheme of the present application: the analysis process of the biological impedance change rate is:

[0033] The multiple numerical components in the feature vector representing the spectral response of different tissues are compared with the preset spectral discrimination thresholds corresponding to the multiple preset tissue types respectively.

[0034] According to the comparison results, the initial probability values for each preset tissue type are calculated.

[0035] The dispersion degree between the initial probability values corresponding to all tissue types is calculated to obtain the overall confidence index.

[0036] When the overall confidence index is lower than the preset confidence threshold, the normalization smoothing processing based on the corresponding confidence index is performed on all initial probability values to obtain the final tissue type probability distribution. The biological impedance values at the current and previous time points are extracted from the feature vector, the change amount of the biological impedance values within a unit time is calculated to obtain the biological impedance change rate.

[0037] As a further scheme of the present application: the laser or plasma parameters are adjusted and the surgical area is affected, specifically including:

[0038] The energy control instruction is received, and the energy control instruction contains the final energy form identifier, power value, energy action time and specific waveform mode code determined after the safety decision;

[0039] If the energy form identifier is laser, a set of wavelength tuning signals and pulse frequency modulation signals for controlling the laser generator are generated according to the power value and the waveform mode code to adjust the laser parameters;

[0040] If the energy form identifier is plasma, a set of radio frequency power driving signals and working gas flow adjustment signals for controlling the plasma generator are generated according to the power value and the waveform mode code to adjust the plasma parameters.

[0041] The beneficial effects of the present application are:

[0042] (1) Traditional surgery relies on the doctor's experience to make risk judgments, while this invention constructs a topological relationship map of the surgical area to digitally model the spatial and functional relationships between tissues. The system not only accurately identifies key node groups representing blood vessels based on spectral and bioimpedance characteristics, but also introduces risk propagation analysis. This analysis quantifies potential risks (risk assessment parameters) by enumerating all possible transmission paths of energy from the operation point to the key nodes and calculating the comprehensive attenuation value. This transforms the clinical principle of "protecting blood vessels" from an abstract concept that relies on subjective experience into an objective parameter that can be quantified in real time based on topological network calculations. When generating energy commands, this risk assessment parameter serves as the primary constraint, which can directly force the system to suppress high-energy cutting modes when the risk is too high and prioritize the use of protective strategies, thereby proactively avoiding the risk of accidental damage to large blood vessels at the source of decision-making, and shifting safety control from "post-event remediation" to "pre-event prevention".

[0043] (2) This invention integrates multimodal sensing data with real-time risk analysis, enabling energy control commands to dynamically match the microscopic state of the tissue. The system parses the probability distribution of tissue type and the real-time rate of change of bioimpedance from the feature vector of the target node, the latter reflecting the tissue's water content and thermal effect. Combined with the safety level, the optimal combination of power, action time, and waveform mode is directly mapped through a multidimensional query index. This closed-loop control based on real-time physiological feedback allows energy output to adapt to different tissue types (such as tumors, normal mucosa) and their instantaneous state changes (such as drying, carbonization). It reduces the absolute dependence on operator experience, lowers the problem of excessive tissue damage or low cutting efficiency caused by improper manual adjustment, and maintains the best energy effect through automatic optimization, thereby improving the overall surgical cutting efficiency, coagulation effect, and operational consistency while ensuring safety boundaries. Attached Figure Description

[0044] The invention will now be further described with reference to the accompanying drawings.

[0045] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0046] 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.

[0047] Please see Figure 1 As shown, this invention is a closed-loop control system for an AI energy platform used in urinary tract surgery, comprising:

[0048] Topology graph construction module: used to construct a topology graph of the surgical area based on multimodal sensor data including visual and bioimpedance data acquired in real time from the surgical area;

[0049] Key structure identification module: Based on the feature vectors of nodes in the topology graph, identify key node groups corresponding to preset key anatomical structures from all nodes;

[0050] Risk propagation analysis module: It is used to receive the topology graph containing the key node group identifier, and based on the topology of the corresponding topology graph, it analyzes the associated paths between non-key nodes and key node groups through edge connections, calculates the potential impact of energy operations applied to any non-key node on the key node group, and generates risk assessment parameters corresponding to each non-key node.

[0051] Safety control command generation module: used to generate energy control commands that match the target node by real-time fusion of the target node's own feature vector and risk assessment parameters for the target node being acted upon by the current surgical instrument;

[0052] Closed-loop control execution module: Based on energy control commands, it adjusts the output laser or plasma parameters in real time and acts on the surgical area; the changes in the state of the surgical area after the action are captured by multimodal sensor data and fed back to the topology graph construction module to form closed-loop control.

[0053] In the topology mapping module, the construction of the surgical area topology map begins with the real-time acquisition of multimodal sensor data. Visual data is acquired via a miniature camera integrated into the endoscope tip. Bioimpedance data is acquired via a pair of miniature electrodes mounted on the working end of the surgical instrument. When the instrument contacts the tissue, this electrode pair applies a safe alternating current with an amplitude of 1 volt and a frequency range between 5 kHz and 500 kHz to the local tissue at a frequency of 1000 times per second, and simultaneously measures its voltage response. The complex impedance value at the contact point is calculated, and its magnitude is used as the real-time bioimpedance data.

[0054] Based on synchronized multimodal data, physical partitioning and node generation of the surgical area are performed. For each frame of real-time image, a simple linear iterative clustering algorithm is used for superpixel segmentation, dividing the image into approximately 500 regions of similar size and continuous color and texture, each defined as a physical partition. Each physical partition corresponds to a node in the topological graph. A feature vector with a dimension of 8 is generated for each node, constructed as follows: the average values ​​of the red, green, and blue channels within the superpixel region corresponding to the node are calculated as the first three components; the relative reflection intensities near the three characteristic wavelengths of 540 nm, 577 nm, and 600 nm are extracted from the spectral sensor data corresponding to the same spatial coordinates, as the fourth to sixth components; the bioimpedance modulus values ​​of all measurement points within the physical partition are averaged to obtain the seventh component; the standard deviation of the bioimpedance within the partition over the most recent 100 millisecond time window is calculated as the eighth component.

[0055] Edges in the topological graph are established based on two types of associations. The first is spatial adjacency: if the physical partitions represented by two nodes share a boundary on the image plane, an edge is established between these two nodes, and this edge is marked as a "spatial adjacency edge". The second is functional connectivity: the Euclidean distance of the eigenvectors between any two nodes is calculated, especially comparing the dynamic trends of their bioimpedance components (the 7th and 8th components). If the bioimpedance values ​​of two nodes show a highly synchronized upward or downward trend over five consecutive sampling periods (the correlation coefficient of their direction of change is greater than 0.8), then it is considered that these two partitions may have blood flow or electrophysiological connectivity, and an edge is established between the corresponding two nodes; this edge is marked as a "functional connectivity edge".

[0056] During the procedure, the aforementioned topological graph is dynamically updated. Each time a new video frame is acquired and processed, superpixel segmentation is re-performed, and the spatial location and visual feature vectors of the nodes are updated. Bioimpedance data is refreshed at a higher frequency, and the bioimpedance-related feature components (7th and 8th components) of the nodes are updated in real time. The set of edges is also adjusted accordingly: spatially adjacent edges are reconstructed based on the new superpixel adjacency relationships; functionally connected edges are updated or removed based on the latest continuous bioimpedance data after recalculating their correlation coefficients.

[0057] In the key structure identification module, key node groups are identified from all nodes in the topology graph, starting with an initial screening based on spectral features. The preset key anatomical structure is the blood vessel, and its spectral feature threshold is set according to the absorption characteristics of oxyhemoglobin at specific wavelengths. Specifically, for a node's feature vector, its 4th to 6th components are extracted, corresponding to the relative reflection intensities near wavelengths of 540 nm, 577 nm, and 600 nm. The preset threshold condition is that the value of the 5th component (577 nm) must be lower than the values ​​of both the 4th component (540 nm) and the 6th component (600 nm), and the difference between the three must be greater than 0.15. Each node in the topology graph is traversed, and its corresponding component of the feature vector is compared with this threshold condition. All nodes that meet the above conditions are marked as candidate key nodes and proceed to the next verification step.

[0058] Next, a secondary screening is performed based on the pulsation characteristics of bioimpedance. The bioimpedance magnitude sequence for each candidate key node within the most recent 1000 millisecond time window is obtained, forming a dynamic change curve of bioimpedance. The periodic pulsation amplitude of this curve is calculated as follows: first, all local maxima and local minima within the time window are identified; then, the absolute value of the difference between each adjacent local maximum and local minimum is calculated; finally, these absolute values ​​are averaged, and the resulting average value is the periodic pulsation amplitude. The preset vascular characteristic amplitude range is 0.5 ohms to 3.0 ohms. The calculated pulsation amplitude of each candidate key node is compared with this range, and only nodes whose amplitude falls within this range are retained. These nodes that pass the comparison constitute a set of key nodes to be verified.

[0059] Finally, the final determination is made by analyzing the consistency between nodes and their neighbors. For each node in the set of key nodes to be verified, all first-order adjacent nodes directly connected to it by an edge are identified in the topological graph. The feature vectors of the node and each of its adjacent nodes are extracted. The coherence consistency parameter is calculated through the following steps: First, the Pearson correlation coefficients between the node and each of its adjacent nodes on the spectral feature components (4th to 6th components) are calculated to obtain a set of spectral correlation coefficients; simultaneously, the Pearson correlation coefficients between the node and each of its adjacent nodes on the bioimpedance component (7th component) are calculated to obtain a set of impedance correlation coefficients. Then, the average of all spectral correlation coefficients is added to the average of all impedance correlation coefficients, and the sum is the coherence consistency parameter of the node. The preset coherence threshold is 0.7. If the calculated coherence consistency parameter of a node is higher than 0.7, the node is determined to be the final key node and added to the key node group. This step ensures that the identified key nodes not only conform to vascular characteristics in their own characteristics, but also that the tissue characteristics of their local area are consistent, thus improving the accuracy of identification.

[0060] In the risk propagation analysis module, the generation of risk assessment parameters begins with the analysis of a topological graph containing key node group identifiers. First, for each non-critical node in the topological graph, all associated paths leading to each critical node within the key node group are enumerated. The enumeration process employs a depth-first search strategy, starting from the current non-critical node and traversing along the edges connecting nodes to search for all paths leading to the target critical node. During this process, the maximum path search depth is set to 5, meaning the total number of edges in an associated path does not exceed 5, to avoid infinite computational complexity and to ignore excessively long propagation paths based on the physical law that energy influence decays with distance. All found connected sequences consisting of a series of edges connected end-to-end are recorded as associated paths from that non-critical node to the specific critical node.

[0061] For each enumerated associated path, a path attenuation factor needs to be calculated to quantify the degree of attenuation of energy influence as it propagates along the path. The calculation process begins by assigning an initial weight to each edge on the path. This weight is differentiated based on the edge type: if an edge is marked as a "spatial adjacency edge," it is assigned a first weight value, which is preset to 0.7; if an edge is marked as a "functional connectivity edge," it is assigned a second weight value, which is preset to 0.9. The second weight value is greater than the first weight value, reflecting that the energy conduction efficiency through functional connectivity is higher than that through simple spatial proximity.

[0062] Subsequently, based on the energy type currently used in the plan, an energy attenuation correction coefficient is introduced into the initial weights to obtain the equivalent weight of each edge under a specific energy level. The current energy types are divided into laser and plasma. If laser energy is used, the correction coefficient is set as follows: 1.0 for spatially adjacent edges and 1.1 for functionally connected edges. If plasma energy is used, the correction coefficient is set as follows: 0.9 for spatially adjacent edges and 1.2 for functionally connected edges. The equivalent weight is calculated by multiplying the initial weight value of an edge by the correction coefficient corresponding to the edge type and the current energy type; the product is the equivalent weight of that edge.

[0063] After obtaining the equivalent weights of all edges on an associated path, the path attenuation factor of that path is calculated. The specific calculation process is as follows: the equivalent weights of all edges from the start to the end of the path are multiplied sequentially to obtain a product. Then, this product is divided by the total number of edges contained in the associated path. The final quotient is defined as the path attenuation factor of the associated path. The physical significance of this calculation process is that the product simulates the cumulative attenuation effect of energy propagation along the path as it passes through each edge, while dividing by the total number of edges normalizes the process for paths of different lengths, making the attenuation factors of paths of different lengths comparable.

[0064] After calculating all path attenuation factors, for the same non-critical node, the influence of different paths from it to all critical nodes needs to be considered. Specifically, the path attenuation factors of all associated paths from the non-critical node to each critical node within the critical node group are summed to obtain a total value, called the comprehensive attenuation value. This value represents the potential intensity of the impact of an energy operation applied from the non-critical node propagating through all possible paths to the entire critical node group.

[0065] Finally, based on the calculated overall attenuation value, a pre-defined linear mapping relationship is used to generate the final risk assessment parameters for the target non-critical node. This mapping relationship is defined as follows: the minimum possible value of the overall attenuation is 0, and the maximum possible empirical value is 10. When the overall attenuation value is less than or equal to 2, the risk assessment parameter is mapped to 0, representing negligible risk; when the overall attenuation value is greater than 2 and less than or equal to 5, the risk assessment parameter is mapped to the difference between the overall attenuation value and 2; when the overall attenuation value is greater than 5, the risk assessment parameter is uniformly mapped to 3, representing the highest level of risk. Through these steps, a quantified risk assessment parameter is generated for each non-critical node, and this parameter is directly used for subsequent energy control decisions.

[0066] In the safety control command generation module, the generation of energy control commands begins with the fusion processing of risk assessment parameters and feature vectors of the target node. First, the received risk assessment parameters of the target node are... (This is based on the aforementioned risk propagation analysis) and is compared with three preset consecutive safety interval thresholds. These three thresholds are set as follows: , , This led to the definition of four security levels. The judgment rule is: if ,but (Safety); If ,but (Low risk); if ,but (Medium risk); if ,but (High risk). This security level constitutes the primary constraint for the generation of subsequent instructions.

[0067] Subsequently, the probability distribution of organization type is parsed from the feature vector of the target node. The feature vector... It contains multiple components, of which the 4th to 6th components , , Corresponding spectral response values ​​at wavelengths of 540 nm, 577 nm, and 600 nm. Preset tissue types include normal mucosa. Tumor tissue and bleeding area Each type of organization Corresponding to a set of preset spectral discrimination thresholds, including the lower limit and upper limit For each organization type, an initial matching score is calculated. The expression for the initial matching score is: ;

[0068] in, Indicates the initial matching score. This is an indicator function; its value is 1 when the condition inside the parentheses is true, and 0 otherwise. Indicates the first The lower limit for each organization type Indicates the first The upper limit of the number of organization types, eigenvector Each component represents a fraction. This score indicates the number of spectral components of the feature vector that fall within a preset threshold range for this tissue type. Initial probability value. It can be calculated using the following formula: ;in, Indicates the first Matching scores for different organizational types;

[0069] This calculation uses the Softmax function to transform the matching scores into a probability distribution, such that the sum of the initial probabilities of all organization types is 1.

[0070] Next, the reliability of this initial probability distribution is evaluated. Initial probability values ​​are calculated for all tissue types. The degree of dispersion between them yields the overall confidence index. Here, entropy is used to measure the degree of dispersion (i.e., uncertainty): ;

[0071] Overall confidence index Defined as: ;in, This represents the maximum entropy value for the three types of distributions. The value ranges from 0 to 1, with a larger value indicating a higher degree of certainty in the probability distribution. The preset confidence threshold... Set it to 0.6. When At that time, it was considered that the initial classification confidence was insufficient, and all initial probability values ​​needed to be recalculated. Normalization and smoothing are performed based on this confidence index to obtain the final organization type probability distribution. The calculation expression for the organization type probability distribution is as follows: ;in, This represents the probability distribution of the final organization type. The total number of organization types, It is a confidence weight used to smoothly adjust the initial probability distribution towards a uniform distribution when the initial classification confidence is insufficient. Its value is equal to the overall confidence index. This process performs linear interpolation between the initial probability and a uniform distribution. The lower the confidence level, the closer the result tends to a uniform distribution, reflecting a conservative principle under high uncertainty. If... Then directly order .

[0072] Simultaneously, the rate of change of bioimpedance is extracted from the feature vector. Specifically, the bioimpedance value sequence for the most recent five time points is obtained from the seventh component of the feature vector and the stored historical values ​​from the previous four time points. Rate of change of bioimpedance The slope of the linear regression of this sequence over a unit time period (500 milliseconds) is calculated. That is, a univariate linear regression is performed with the time point as the independent variable and the impedance value as the dependent variable; the resulting slope is the linear regression slope. The unit is ohms per second.

[0073] Then, the security level obtained above The final organization type probability distribution (Usually, the tissue type with the highest probability is selected) (as a representative) and the rate of change in bioimpedance The three elements are combined to form a multidimensional query index: .in, It was quantified into three categories: (like ), (like ), (like ).

[0074] Finally, based on this multidimensional query index, a lookup and matching are performed in a predefined energy parameter mapping table. This mapping table is... The triplet serves as the key, and its corresponding value is a specific energy control command triplet (power, duration, energy waveform pattern). For example, the index... Possibly mapped to instructions This indicates a power of 80 watts, an action time of 100 milliseconds, and a pulse waveform mode. By directly looking up a table, energy control commands that strictly match the real-time state of the target node can be obtained, thus achieving comprehensive decision-making based on safety constraints, organizational characteristics, and dynamic physiological responses.

[0075] In the closed-loop control execution module, the execution of closed-loop control begins with the reception and parsing of energy control commands. The commands are received in the form of a data packet, which explicitly contains four key parameters: the final energy form identifier (integer 1 for laser, integer 2 for plasma), the power value (in watts), the energy application time (in milliseconds), and a specific waveform pattern code (integer, e.g., 1 for continuous wave, 2 for pulsed wave). This command is a direct output of the previous safety decision-making process.

[0076] If the resolved energy form is identified as laser (i.e., value 1), then a specific laser control signal is generated based on the power value and waveform pattern encoding in the instruction. First, based on the power value, it is converted into the injection current value required to drive the laser diode through a preset power-current lookup table. For example, for a thulium laser generator with a wavelength of 1940 nm, 80 watts of power may correspond to a driving current of 2.5 amps. An analog voltage signal proportional to this current value is generated as a wavelength tuning signal, which maintains the wavelength stability of the laser output through a proportional-integral controller. Simultaneously, a pulse frequency modulation signal is generated based on the waveform pattern encoding: if the encoding is 2 (pulse wave), then based on a preset optimized frequency matched to the target tissue type (e.g., 30 Hz to 50 Hz is commonly used for prostate tissue), a square wave digital signal with a corresponding frequency and duty cycle (usually set to 50%) is generated to modulate the aforementioned continuous laser output, making it pulsed. These two sets of signals are synchronously sent to the laser generator's hardware control interface to precisely adjust its output laser parameters.

[0077] If the resolved energy form is identified as plasma (i.e., value 2), then the control signal for the plasma generator is generated based on the power value and waveform pattern code in the instruction. First, the power value is converted into a radio frequency (RF) power drive signal. Specifically, the power value is divided by a conversion factor (e.g., a factor of 10, representing 10 watts per volt) to obtain a base voltage value. This base value is then modulated according to the waveform pattern code: if it's a continuous wave mode, a constant voltage is output; if it's a pulse wave mode, it's switched on and off using a low-frequency pulse signal (e.g., 10 Hz). This modulated voltage signal is used to control the output amplitude of the RF power supply. Simultaneously, a working gas flow rate regulation signal is generated. The gas flow rate (in liters per minute) is linearly determined based on the power value, calculated as follows: the base flow rate equals the power value multiplied by a gas flow rate coefficient (e.g., 0.05 liters per minute per watt), ensuring it remains within a safe range (e.g., 1 to 5 liters per minute). This signal controls the opening of a proportional solenoid valve to precisely regulate the flow rate of argon or a mixed gas delivered to the tip of the surgical instrument. These two sets of signals work together to regulate the intensity, stability, and cutting-solidification characteristics of the plasma plume.

[0078] The generated control signal drives the energy generator, causing laser or plasma energy to be applied through the tip of the surgical instrument to the physical partition corresponding to the current target node in the surgical area. The energy causes the local tissue to undergo the desired changes, such as cutting, vaporization, or coagulation.

[0079] Following the application of energy, the new state of the surgical area is captured in real time by multimodal sensors. An endoscopic camera captures visual data of changes in color and morphology within the treated area; contact electrodes measure new bioimpedance values ​​caused by tissue dehydration, carbonization, or coagulation. This newly acquired sensor data is immediately transmitted to the initial data processing step of this method.

[0080] The newly acquired sensor data was used to update the initially constructed surgical area topology map. Specifically, the visual and bioimpedance components in the feature vectors of the target node and its surrounding affected nodes were recalculated and replaced based on the new data. Simultaneously, based on the new spatial adjacency relationships and bioimpedance synchronicity after tissue changes, the "edges" between nodes may also be added, deleted, or relabeled. This updated topology map, containing the latest state of the tissue after energy intervention, will serve as a new input for the next control cycle, thus forming a complete, real-time closed-loop control system from "perception-decision-execution" to "re-perception."

[0081] The working principle of this invention is as follows: First, visual and bioimpedance data of the surgical area are acquired in real time using an endoscopic camera and microelectrode pairs on surgical instruments, and a dynamically updated topological graph of the surgical area is constructed based on this data. This graph uses superpixel partitions as nodes, integrates color, spectral, and bioimpedance features, and establishes edges between nodes through spatial adjacency and functional connectivity. Next, based on the unique spectral absorption characteristics and bioimpedance pulsation patterns of blood vessels, the system identifies key node groups representing critical blood vessels from all nodes. Then, by analyzing all associated paths from any non-critical node to the key node group to be operated on, and combining the type and weight of the edges on the paths with the currently used energy type, the system calculates the attenuation degree of energy influence propagating along these paths, and comprehensively generates risk assessment parameters for the operation location. Subsequently, the system integrates the risk assessment parameters of the target node, the probability distribution of tissue type, and the real-time bioimpedance change rate, and generates specific energy control instructions containing power, action time, and waveform mode by querying a preset energy parameter mapping table. Finally, the instruction is converted into a control signal to drive the laser generator or plasma generator, applying precise energy to the surgical area. The changes in the tissue state after the action are captured by the sensor again and used to update the topology map in real time, thereby starting the next control cycle and forming a closed loop of continuous perception, analysis, decision-making and execution.

[0082] The foregoing has provided a detailed description of one embodiment of the present invention, but this description is merely a preferred embodiment and should not be construed as limiting the scope of the invention. All equivalent variations and modifications made within the scope of the claims of this invention should still fall within the patent coverage of this invention.

Claims

1. A closed-loop control system for an AI energy platform used in urological surgery, characterized in that, include: Topology graph construction module: used to construct a topology graph of the surgical area based on multimodal sensor data including visual and bioimpedance data acquired in real time from the surgical area; Key structure identification module: Based on the feature vectors of nodes in the topology graph, identify key node groups corresponding to preset key anatomical structures from all nodes; Risk propagation analysis module: This module receives a topology graph containing key node group identifiers and, based on the topology of the corresponding graph, analyzes the associated paths between non-key nodes and key node groups via edges. It calculates the potential impact of energy operations applied to any non-key node on the key node group and generates risk assessment parameters corresponding to each non-key node. The generation process of these risk assessment parameters specifically includes: For each non-critical node in the topological graph, enumerate all associated paths from the non-critical node to all critical nodes in the critical node group. Each associated path consists of a series of edges connected end to end. For each enumerated associated path, a path attenuation factor is calculated based on the preset weights of each edge on the associated path and the total number of edges contained in the path, which characterizes the degree of attenuation of energy influence on the corresponding associated path. The path attenuation factors of different associated paths from the same non-critical node to all critical nodes are superimposed to obtain a comprehensive attenuation value. Risk assessment parameters are generated based on the magnitude of the comprehensive attenuation value. The calculation process for the path attenuation factor is as follows: Based on the type of each edge that constitutes the associated path, an initial weight is assigned to each edge; wherein, the edge connecting adjacent physical partitions in space is assigned a first weight value, and the edge connecting physical partitions with functional connectivity is assigned a second weight value greater than the first weight value. Based on the energy type used in the current plan, energy attenuation correction coefficients are introduced for the first and second weight values ​​respectively to obtain the equivalent weight of each edge under the action of a specific energy. Multiply the equivalent weights of all edges on the associated path together, and then divide the result by the total number of edges contained in the corresponding associated path. The quotient is the path decay factor. Safety control command generation module: used to generate energy control commands that match the target node by real-time fusion of the target node's own feature vector and risk assessment parameters for the target node being acted upon by the current surgical instrument; Closed-loop control execution module: Based on energy control commands, it adjusts the output laser or plasma parameters in real time and acts on the surgical area; the changes in the state of the surgical area after the action are captured by multimodal sensor data and fed back to the topology graph construction module to form closed-loop control.

2. The closed-loop control system for an AI energy platform used in urinary tract surgery according to claim 1, characterized in that, The surgical area topology diagram specifically includes: The surgical topology graph consists of nodes and edges connecting the nodes. Nodes correspond to physical partitions of the surgical area. Each node has a feature vector formed by the fusion of multimodal sensor data based on the corresponding partition. Edges are used to represent the spatial adjacency or functional connectivity between the physical partitions corresponding to the nodes.

3. The closed-loop control system for an AI energy platform used in urinary tract surgery according to claim 1, characterized in that, The identification of key node groups corresponding to preset key anatomical structures from all nodes specifically includes: Based on the preset key anatomical structure spectral feature threshold, the feature vectors of all nodes in the topological relationship graph are initially screened, and nodes whose spectral feature vectors meet the threshold are marked as candidate key nodes. Obtain the bioimpedance dynamic change curves corresponding to candidate key nodes, calculate the periodic pulsation amplitude of the bioimpedance dynamic change curves per unit time; compare the pulsation amplitude with the preset vascular characteristic amplitude range, screen out nodes that meet the requirements, and form a set of key nodes to be verified. For each node in the set of key nodes to be verified, extract the feature vectors of all first-order adjacent nodes in the topology graph, and calculate the coordination consistency parameter of the corresponding node and its adjacent nodes in terms of spectral and impedance characteristics. If the coordination consistency parameter is higher than the preset coordination threshold, the corresponding node is determined to be the final key node and is added to the key node group.

4. The closed-loop control system for an AI energy platform used in urinary tract surgery according to claim 1, characterized in that, The generation of energy control commands that match the target node specifically includes: The received risk assessment parameters of the target node are compared with multiple preset consecutive security interval thresholds to determine the security level to which the risk assessment parameters belong. From the feature vector of the target node, the probability distribution representing the tissue type and the real-time rate of change of bioimpedance are extracted; The probability distribution of safety level, tissue type, and rate of change of bioimpedance are combined to form a multidimensional query index. Based on a multidimensional query index, a search and match is performed in a predefined energy parameter mapping table to directly map an energy control command composed of a specific power value, application time, and energy waveform pattern.

5. The closed-loop control system for an AI energy platform used in urinary tract surgery according to claim 4, characterized in that, The analytical process for the rate of change of bioimpedance is as follows: The numerical components in the feature vector that represent the spectral response of different tissues are compared with the preset spectral discrimination thresholds corresponding to various tissue types. Based on the comparison results, an initial probability value is calculated for each preset tissue type; Calculate the degree of dispersion among the initial probability values ​​corresponding to all organization types to obtain the overall confidence index; When the overall confidence index is lower than the preset confidence threshold, all initial probability values ​​are normalized and smoothed based on the corresponding confidence index to obtain the final organization type probability distribution. The bioimpedance values ​​at the current and previous times are extracted from the feature vector, and the change in bioimpedance values ​​per unit time is calculated to obtain the rate of change of bioimpedance.

6. The closed-loop control system for an AI energy platform used in urinary tract surgery according to claim 1, characterized in that, The adjustment of the output laser or plasma parameters, and its effect on the surgical area, specifically includes: Receive energy control commands, which include the final energy form identifier, power value, energy application time, and specific waveform pattern code determined after safety decisions. If the energy form is identified as laser, a set of wavelength tuning signals and pulse frequency modulation signals are generated based on the power value and waveform pattern encoding to control the laser generator and adjust the laser parameters. If the energy form is identified as plasma, a set of radio frequency power drive signals and working gas flow regulation signals are generated based on the power value and waveform pattern encoding to control the plasma generator and adjust the plasma parameters.

Images