An ultrasonic-guided integrated system for peritoneal perfusion and drainage

Abstract

The application discloses an ultrasonic guidance-based peritoneal perfusion and drainage integrated system, relates to the perfusion and drainage control technical field, and comprises an ultrasonic guidance module, a puncture and establishment module, a liquid medicine processing module, a perfusion and drainage execution module, a state determination module and a cooperative control module; first, a puncture target point, a puncture path and a peritoneal perfusion baseline image data before perfusion are determined based on real-time ultrasonic images, then puncture and establishment, peritoneal chemotherapy liquid preparation, peritoneal dialysis machine perfusion and drainage execution are completed, and the liquid diffusion state, the target area coverage state and the abnormal state are determined in combination with real-time ultrasonic images, execution parameters and vital signs, and then cooperative control instructions are generated; the accuracy of target area coverage determination in the peritoneal tumor targeted chemotherapy process can be improved, and the safety and adaptability of perfusion and drainage control can be enhanced.

Description

Technical Field

[0001] This invention relates to the field of perfusion and drainage control technology, and more specifically, to an integrated system for intraperitoneal perfusion and drainage based on ultrasound guidance. Background Technology

[0002] In existing intraperitoneal tumor perfusion therapy techniques, the clinical pathway typically involves establishing a perfusion channel after ultrasound-guided puncture localization, and then administering medication according to a preset flow rate and dose. This is combined with intraperitoneal pressure monitoring to determine whether deceleration or pausing is necessary. While this approach has a certain foundation in terms of equipment and operational procedures, its control logic is still primarily based on the premise that effective drug administration is achieved upon successful puncture and that lesion coverage is achieved upon reaching the required total perfusion volume. However, the peritoneum is not a regular, stable, and homogeneous space. The distribution of intestinal loops, omental obstruction, postoperative adhesions, peritoneal space septa, and differences in local compliance within the peritoneum cause the medication to exhibit significant non-uniform diffusion characteristics after entering the peritoneum, leading to a deviation between the actual diffusion path and the intended target path.

[0003] Therefore, in real-world treatment scenarios, even if ultrasound confirms accurate needle placement, the medication may still preferentially enter low-resistance non-lesion areas, or experience deflection, accumulation, or restricted diffusion on adjacent organ surfaces, adhesion gaps, and locally enclosed areas, resulting in insufficient tumor target coverage. Simultaneously, local areas may experience increased pressure, organ compression, or leakage risks. Furthermore, current abdominal pressure monitoring technologies tend to focus on threshold assessments of overall safety, indicating infusion risks but failing to reveal the spatial reasons for ineffective target coverage. While two-dimensional ultrasound can provide information on local fluid echo changes, it typically remains at the observational level and lacks a systematic approach to discriminating and controlling the direction of medication diffusion, target area coverage, and bypass diffusion in non-target areas.

[0004] Therefore, the deficiency of the existing technology is that, in the dynamic and non-uniform space of the abdominal cavity, there is a lack of a technical mechanism that can accurately determine whether the drug solution truly covers the target area of ​​the abdominal tumor based on real-time imaging results, and directly convert the determination result into the basis for perfusion correction control. Summary of the Invention

[0005] To overcome the above-mentioned deficiencies of the prior art, embodiments of the present invention provide an ultrasound-guided integrated peritoneal perfusion and drainage system to solve the problems that during peritoneal tumor perfusion treatment, only puncture positioning and drug infusion can be completed, but it is difficult to accurately determine whether the tumor target area is effectively covered based on the real-time diffusion of the drug in the peritoneal cavity, and it is difficult to link the coverage status, abnormal status and perfusion and drainage process for coordinated control.

[0006] To achieve the above objectives, the present invention provides the following technical solution: an integrated system for peritoneal perfusion and drainage based on ultrasound guidance, comprising an ultrasound guidance module, a puncture and perfusion module, a drug treatment module, a perfusion and drainage execution module, a status determination module, and a collaborative control module. The ultrasound guidance module is used to acquire real-time ultrasound images of the patient's abdominal cavity, and based on the real-time ultrasound images, identify the target area of ​​the abdominal tumor, the boundaries of surrounding organs, and the safe puncture channel, and output the puncture target point, puncture path, and pre-perfusion abdominal ultrasound baseline image data. The puncture and connection module is used to receive the puncture target and puncture path, complete the insertion of the puncture needle under real-time ultrasound guidance, establish an irrigation and drainage channel connecting the puncture needle and the peritoneal dialysis machine, and output the needle tip position confirmation result and the channel position confirmation result. The drug solution processing module is used to receive the preset chemotherapy regimen and channel location confirmation results, complete the preparation of intraperitoneal chemotherapy drug solution, record the dosage, concentration and preparation time of intraperitoneal chemotherapy drug solution, and output the intraperitoneal chemotherapy drug solution recording results. The perfusion and drainage execution module is used to receive the needle tip position confirmation result, channel position confirmation result, peritoneal chemotherapy drug solution recording result, and control commands output by the collaborative control module. It controls the peritoneal dialysis machine to perform trial injection, drug infusion, auxiliary drainage, and treatment end drainage through the perfusion and drainage channel. It also responds to control commands to perform pause and hold, flow rate adjustment, and infusion direction adjustment, and outputs execution status and execution parameters. The status determination module is used to receive real-time ultrasound images, pre-perfusion abdominal ultrasound baseline image data, execution status, execution parameters, and patient blood pressure, heart rate, respiratory rate and abdominal pressure, and to jointly determine the drug diffusion status, target coverage status and abnormal status, and output the target coverage status, abnormal status and status change trend. The collaborative control module is used to receive the recording results of intraperitoneal chemotherapy drug solution, execution status, execution parameters, target coverage status, abnormal status and status change trend, generate control instructions for the drug infusion stage and drainage stage, and send the control instructions to the perfusion and drainage execution module to achieve collaborative control of perfusion and drainage.

[0007] By adopting the above technical solutions, it is possible to complete the identification of the target area of ​​abdominal tumor, puncture and establishment, drug preparation, perfusion and drainage execution, status judgment and collaborative control under real-time ultrasound guidance, thereby realizing the integrated linkage control of perfusion and drainage during targeted chemotherapy for abdominal tumors.

[0008] In a preferred embodiment, the ultrasound guidance module is used to: take the real-time ultrasound image sequence of the patient's abdominal cavity as input, perform temporal registration and boundary enhancement processing on the ultrasound images at consecutive time points, segment the abdominal tumor target area, the boundaries of surrounding organs and the abdominal wall access area, and output the outline of the abdominal tumor target area, the set of boundaries of surrounding organs and the candidate access area. The peritoneal tumor target area contour, the set of surrounding organ boundaries, and the candidate access route region are used as inputs to construct candidate connectivity paths from the candidate access route region to the peritoneal tumor target area contour. The organ avoidance distance, path curvature change, and cross-time path stability corresponding to each candidate connectivity path are calculated, and the safe puncture channel score and the candidate connectivity path with the highest score are output. Using the candidate connectivity path with the highest score and the real-time ultrasound image sequence as input, the intersection of the candidate connectivity path and the peritoneal tumor target area contour is determined as the puncture target point, the candidate connectivity path is determined as the puncture path, and the ultrasound image corresponding to the moment with the highest path stability is extracted as the pre-perfusion peritoneal ultrasound baseline image data.

[0009] By adopting the above technical solution, it is possible to perform temporal registration, boundary enhancement, and candidate path scoring on puncture approaches based on real-time ultrasound images, thereby improving the rationality and stability of the determination of puncture target points and puncture paths.

[0010] In a preferred embodiment, the puncture and connection module is used to: take the puncture target point, puncture path and real-time ultrasound image as input, jointly track the needle body trajectory and needle tip echo position during the puncture needle advancement process, calculate the path deviation between the current position of the needle tip and the puncture path and the target deviation between the current position of the needle tip and the puncture target point, and output the needle tip alignment status and path tracking results. The needle tip alignment status, path tracking results, and real-time ultrasound images are used as inputs. When the needle tip reaches the puncture target at its current position and the path deviation is within the preset deviation range, the needle tip determines whether the end of the puncture needle has entered the target perfusion cavity based on the changes in the fluid echo in the vicinity of the needle tip and the displacement changes of the surrounding organ boundaries, and outputs the needle tip position confirmation result. The system takes the needle tip position confirmation result, real-time ultrasound image, and peritoneal dialysis machine connection status as inputs, controls the puncture needle to establish a perfusion drainage channel connected to the peritoneal dialysis machine, and determines whether the perfusion drainage channel is connected and effective based on the test fluid echo distribution, local fluid diffusion continuity, and backflow response results after the channel is established, and outputs the channel position confirmation result.

[0011] By adopting the above technical solution, needle trajectory tracking, needle tip alignment determination, and channel validity confirmation can be completed simultaneously during the advancement of the puncture needle, thereby improving the accuracy and reliability of establishing the perfusion drainage channel.

[0012] In a preferred embodiment, the drug solution processing module is used to: take the preset chemotherapy regimen and the channel location confirmation result as input, analyze the target drug solution type, target dose and target concentration in the preset chemotherapy regimen, and generate corresponding drug solution preparation parameters when the channel location confirmation result indicates that the perfusion drainage channel is effectively established, and output preparation task information; The preparation task information is used as input. The intraperitoneal chemotherapy solution is prepared according to the preparation task information. The amount of drug solution added, the amount of diluent added and the time of preparation completion are collected during the preparation process. The actual dose and actual concentration of the intraperitoneal chemotherapy solution are calculated and the drug solution preparation result is output. The drug preparation result and preparation completion time are used as inputs to record the actual dose, actual concentration and preparation time of the intraperitoneal chemotherapy drug solution, generate the intraperitoneal chemotherapy drug solution record result and output it to the perfusion and drainage execution module and the collaborative control module.

[0013] By adopting the above technical solution, after the perfusion drainage channel is effectively established, the drug solution can be prepared according to the preset chemotherapy plan, and the dosage, concentration and preparation time can be recorded in conjunction, thereby improving the standardization and traceability of the intraperitoneal chemotherapy drug solution processing process.

[0014] In a preferred embodiment, the perfusion and drainage execution module is used to: take the needle tip position confirmation result, channel position confirmation result, intraperitoneal chemotherapy drug recording result, and control command output by the collaborative control module as input, perform consistency verification on the needle tip position confirmation result, channel position confirmation result, and intraperitoneal chemotherapy drug recording result, and determine the corresponding trial injection mode, drug infusion mode, combined drainage mode, or treatment end drainage mode for the peritoneal dialysis machine based on the consistency verification result, and output mode execution command and initial execution parameters; The peritoneal dialysis machine is controlled to perform trial injection, drug infusion, drainage, or treatment termination drainage through the perfusion and drainage channel, using the mode execution command, initial execution parameters, and control command as inputs. During the execution process, the machine's pause and hold, flow rate, and infusion direction are dynamically adjusted according to the control command, and the real-time execution status and real-time execution parameters are output. The system takes the real-time execution status, real-time execution parameters, and peritoneal dialysis machine operation feedback results as inputs, calculates the infused volume, current flow rate, cumulative drainage volume, and channel response status in the current mode, and generates the execution status and execution parameters corresponding to the current mode, which are then output to the status determination module and the collaborative control module.

[0015] By adopting the above technical solution, the peritoneal dialysis machine can be controlled to complete the test injection, drug infusion and drainage in different execution modes according to the consistency verification results, and the execution parameters can be dynamically adjusted according to the control instructions, thereby improving the continuity and adaptability of the perfusion and drainage execution process.

[0016] In a preferred embodiment, the state determination module is used to: take real-time ultrasound images, pre-perfusion peritoneal ultrasound baseline image data and execution parameters as input, register and compare the real-time ultrasound images with the pre-perfusion peritoneal ultrasound baseline image data, extract changes in drug diffusion boundary, diffusion direction and diffusion range, and output the drug diffusion state. Using the drug diffusion state, real-time ultrasound images, and the abdominal tumor target area as inputs, the boundary encirclement relationship and regional overlap relationship between the drug diffusion area and the abdominal tumor target area are calculated, and the target area coverage state is output. Using the drug diffusion status, execution status, and patient blood pressure, heart rate, respiratory rate, and abdominal pressure as inputs, the system performs correlation analysis on drug leakage signs, organ compression signs, and vital sign fluctuations, and outputs abnormal states and state change trends.

[0017] By adopting the above technical solution, it is possible to jointly determine the drug diffusion status, target coverage status, and abnormal status based on real-time ultrasound images, baseline image data, execution status, and vital sign information, thereby improving the accuracy of coverage judgment and abnormal identification during targeted chemotherapy for abdominal tumors.

[0018] In a preferred embodiment, the state determination module is further configured to: take real-time ultrasound images, execution state, execution parameters and drug diffusion state as input, extract the position changes of the drug diffusion front, the migration direction of the diffusion center and the diffusion speed changes at continuous moments, construct the drug diffusion evolution trajectory, and output the diffusion offset result; Using diffusion offset results, target coverage status, and abdominal tumor target boundary as input, calculate the degree of convergence and deviation between the drug diffusion trajectory and the abdominal tumor target boundary, and output the coverage offset status. Using the coverage offset state, abnormal state, and continuous changes in the patient's abdominal pressure and vital signs as input, the risk enhancement trend of the current infusion process is calculated, and the updated state change trend is output.

[0019] By adopting the above technical solutions, it is possible to further obtain diffusion offset results, coverage offset status and risk enhancement trends through the analysis of drug diffusion evolution trajectory, thereby improving the ability to identify drug diffusion offset risk and coverage deviation risk in advance.

[0020] In a preferred embodiment, the state determination module is further configured to: take the execution parameters, diffusion offset results, and coverage offset states as inputs, construct the drug diffusion response prediction result corresponding to the current execution parameters, compare the drug diffusion response prediction result with the actual diffusion result represented by the real-time ultrasound image, and output the diffusion response deviation result; The diffusion response deviation result, abnormal state, and continuous changes in the patient's abdominal pressure and vital signs are used as inputs to determine whether the current abnormal state matches the current execution parameters, and the abnormality correction result is output. The anomaly correction results, target coverage status, and updated status change trend are used as inputs to correct the target coverage status and anomaly status, and the corrected target coverage status, corrected anomaly status, and corrected status change trend are output.

[0021] By adopting the above technical solution, it is possible to predict the drug diffusion response by combining execution parameters and correct abnormal states and target area coverage states, thereby improving the matching and stability between the state judgment results and the actual execution process.

[0022] In a preferred embodiment, the collaborative control module is specifically used to: take the intraperitoneal chemotherapy drug recording results, execution status, execution parameters, target coverage status, abnormal status and status change trend as input, determine the drug infusion stage or drainage stage to which the current treatment process belongs, generate a set of candidate control actions for the corresponding stage based on the dose information, concentration information and preparation time information in the intraperitoneal chemotherapy drug recording results, and output the stage determination result and the set of candidate control actions; The system takes the stage judgment result, candidate control action set, target coverage status, abnormal status and status change trend as input, calculates the comprehensive matching result of each candidate control action on target coverage maintenance, abnormal risk suppression and treatment continuity, and outputs the target control action. The target control action, execution status, and execution parameters are used as inputs to generate control commands corresponding to the perfusion and drainage execution module and send them to the perfusion and drainage execution module. The control commands include pause and hold commands, flow rate adjustment commands, and infusion direction adjustment commands for the drug infusion stage, as well as drainage start commands and drainage parameter adjustment commands for the drainage stage.

[0023] By adopting the above technical solution, control instructions that are adapted to the current drug infusion stage or drainage stage can be generated based on the stage determination results and the candidate control action set, thereby improving the synergistic control effect between target coverage maintenance, abnormal risk suppression and treatment continuity.

[0024] In a preferred embodiment, the collaborative control module is further configured to: take the target control action, execution state, execution parameters and state change trend as input, construct the stage response result corresponding to the target control action, compare the stage response result with the actual execution state and actual execution parameters fed back by the infusion and drainage execution module, and output the control response deviation result; The control response deviation, target coverage status and abnormal status are taken as inputs. The effect is written back and the priority is updated for each candidate control action in the candidate control action set. The updated candidate control action set is output. The updated set of candidate control actions, the stage determination results, and the intraperitoneal chemotherapy drug recording results are used as inputs to recalculate the suitability of each candidate control action for the current stage and generate iterative control instructions to be sent to the perfusion and drainage execution module, so that the control instructions for the drug infusion stage and the drainage stage are adaptively updated with the execution feedback.

[0025] By adopting the above technical solution, it is possible to perform effect write-back, priority update and iterative control on the candidate control action set based on the control response deviation results, thereby improving the adaptive optimization capability of control commands in the drug infusion stage and the drainage stage.

[0026] The technical effects and advantages of this invention are as follows: 1. By adopting a joint judgment mechanism that combines real-time ultrasound images, pre-perfusion abdominal ultrasound baseline image data, execution status, execution parameters and vital signs information, and directly inputting the target area coverage status and abnormal status into the collaborative control module to generate control commands, the judgment result of whether the drug solution truly covers the abdominal tumor target area can be directly transformed into the basis for perfusion correction, thereby improving the pertinence of target area coverage judgment and perfusion safety control in the dynamic non-uniform space of the abdominal cavity; 2. First, obtain the drug diffusion status, target area coverage status, and abnormal status. Then, further obtain the diffusion offset results, coverage offset status, and risk enhancement trend. Combine the execution parameters to correct the status judgment results. This can provide the collaborative control module with a control basis that is closer to the actual treatment process, thereby improving the accuracy of control command generation, updating, and execution adaptation. Attached Figure Description

[0027] Figure 1 This is a system block diagram of the present invention. Detailed Implementation

[0028] 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.

[0029] Refer to the instruction manual appendix Figure 1 An integrated system for intraperitoneal perfusion and drainage based on ultrasound guidance, as detailed below: The ultrasound guidance module is used to acquire real-time ultrasound images of the patient's abdominal cavity, and based on the real-time ultrasound images, identify the target area of ​​the abdominal tumor, the boundaries of surrounding organs, and the safe puncture channel, and output the puncture target point, puncture path, and pre-perfusion abdominal ultrasound baseline image data. In ultrasound-guided targeted chemotherapy for abdominal tumors, relying solely on single-frame ultrasound images for manual observation is susceptible to the effects of respiratory fluctuations, slight probe displacement, intestinal peristalsis, and uneven local echoes. This makes it difficult to stably determine the boundaries of the abdominal tumor target area, the boundaries of surrounding organs, and the available abdominal wall access area for puncture, further leading to fluctuations in the selection of puncture target points and puncture paths. To ensure that subsequent puncture patency is established on a repeatable, calculable, and verifiable basis, in this embodiment, the ultrasound guidance module first performs temporal registration and boundary enhancement processing on the real-time ultrasound image sequence of the patient's abdominal cavity, then completes the segmentation of the abdominal tumor target area, the boundaries of surrounding organs, and the abdominal wall access area. Based on this, candidate connectivity paths are constructed, and safe puncture channel scores are calculated. Finally, the puncture target point, puncture path, and pre-perfusion abdominal ultrasound baseline image data are output. This implementation process includes the following steps: When using a real-time ultrasound image sequence of the patient's abdominal cavity as input, it is preferable to continuously acquire ultrasound image sequences within at least one complete respiratory cycle to ensure that subsequent temporal registration can cover tissue displacement changes caused by respiration. Each frame in the real-time ultrasound image sequence undergoes grayscale normalization and speckle noise suppression. Grayscale normalization maps the grayscale values ​​of images acquired at different times to a uniform range, facilitating subsequent comparisons. Speckle noise suppression can employ median filtering, anisotropic diffusion filtering, or commonly used ultrasound denoising methods to reduce the interference of speckle noise on boundary extraction. A correspondence is established between ultrasound images at consecutive times through temporal registration. Specifically, a frame-by-frame registration method can be used, with the previous image as the reference image and the current image as the image to be registered. First, a coarse translation is determined through overall image correlation, and then detailed positions are corrected through local feature point or local grayscale block matching, thus obtaining an ultrasound image sequence under unified coordinates across time periods. After temporal registration, boundary enhancement processing is performed on the images. The purpose of enhancement processing is to improve the identifiability of the abdominal tumor target area contour, the outer edge of surrounding organs, and the location of echo abrupt changes in the abdominal wall. Gradient enhancement, Laplacian enhancement, or edge response operator enhancement methods can be used. Subsequently, the abdominal tumor target area, the boundaries of surrounding organs, and the abdominal wall access area are segmented on the enhanced ultrasound image. The abdominal tumor target area can be identified by the degree of closure between the hypoechoic or mixed echoic abnormal area and its surrounding boundary. The boundaries of surrounding organs are extracted based on continuous edges, echo levels, and typical organ morphology. The abdominal wall access area is defined as the abdominal wall area between the body surface and the abdominal cavity that is not occupied by bony structures, obvious vascular images, or high-risk tissues that cannot be penetrated. After segmentation, the abdominal tumor target area contour, the set of surrounding organ boundaries, and the access candidate area are output. The abdominal tumor target area contour is represented by a set of closed boundary points, the set of surrounding organ boundaries is represented by a set of multiple organ boundary curves, and the access candidate area is represented by a puncturable continuous segment on the abdominal wall surface. After taking the peritoneal tumor target area contour, the set of surrounding organ boundaries, and the candidate access area as input, the ultrasound guidance module selects multiple candidate starting points from the candidate access area and connects each candidate starting point with multiple candidate arrival points on the peritoneal tumor target area contour to construct candidate connection paths from the candidate access area to the peritoneal tumor target area contour. To avoid the candidate connection paths being too dense, candidate starting points and candidate arrival points can be selected at fixed intervals, such as one point every preset pixel distance along the boundary of the candidate access area and one point every preset arc length along the peritoneal tumor target area contour. After each candidate connection path is constructed, the organ avoidance distance, path curvature change, and cross-time path stability are calculated. The organ avoidance distance refers to the minimum distance from any point on the candidate connection path to the nearest surrounding organ boundary. This indicator reflects the ability to maintain a safe distance from organs during puncture. The minimum value of the entire path is taken as the organ avoidance distance of the candidate connection path. The path curvature change is used to characterize whether the candidate connection path is smooth. Specifically, the candidate connection path can be discretized into multiple continuous line segments, the angle change can be calculated segment by segment, and then the angle changes can be summed. Alternatively, the mean value can be calculated. The smaller the curvature change, the closer the path is to a straight line, which is more conducive to actual puncture. Cross-time path stability is used to characterize the degree to which the same candidate connected path is preserved in images at different registration times. Specifically, the candidate connected path can be mapped to the registered images at each time point, and the relative positional changes between the path and the boundaries of surrounding organs and the outline of the abdominal tumor target area can be statistically analyzed. The smaller the change, the higher the cross-time path stability. After obtaining the above three indicators, each candidate connected path is comprehensively scored. To ensure the feasibility of the scoring, the organ avoidance distance, path curvature change, and cross-time path stability can be normalized first. Among them, the larger the organ avoidance distance, the higher the score; the smaller the path curvature change, the higher the score; and the higher the cross-time path stability, the higher the score. Then, a weighted sum is performed according to preset weights to obtain the safe puncture channel score result. The preset weights can be set according to clinical experience or obtained through historical case statistics. For example, the weight of organ avoidance distance can be set to the highest to prioritize safety. Finally, the safe puncture channel score results corresponding to all candidate connected paths and the candidate connected path with the highest score are output. After inputting the candidate connectivity path with the highest score and the real-time ultrasound image sequence, the ultrasound guidance module performs endpoint localization and baseline image selection for the candidate connectivity path with the highest score. During endpoint localization, the candidate connectivity path with the highest score is extended to the outline of the abdominal tumor target area, and the point where the path first intersects the outline of the abdominal tumor target area is taken as the puncture target point. If multiple adjacent intersection points occur, the intersection point located at the outer edge of the abdominal tumor target area outline and with a large distance from the boundaries of surrounding organs is preferred as the puncture target point, so that the subsequent puncture needle can approach the abdominal tumor target area without easily approaching high-risk organs. After determining the puncture target point, the candidate connectivity path with the highest score is directly determined as the puncture path. Subsequently, pre-perfusion abdominal ultrasound baseline image data is selected from the real-time ultrasound image sequence. This baseline image is not arbitrarily selected, but rather based on the highest score... The stability of high-stability candidate connected paths in images at various time points is used for screening: First, the boundary clarity, the degree of occlusion around the path, and the path position offset are calculated in the image at each time point; the boundary clarity can be determined by the mean gray-level gradient of the path's neighborhood, with a larger gradient indicating a clearer boundary; the degree of occlusion can be determined by the proportion of high-echo occlusion areas in the path's neighborhood, with a lower proportion being more conducive to observation; the path position offset is determined by the deviation between the path position at that time point and the average path position, with a smaller deviation indicating greater stability; after comprehensively comparing the three, the ultrasound image corresponding to the time point with the highest path stability is selected as the pre-perfusion peritoneal ultrasound baseline image data; this pre-perfusion peritoneal ultrasound baseline image data is used for subsequent state determination module registration and comparison with real-time ultrasound images during the perfusion process to determine the drug diffusion state and target area coverage state; Through the above implementation process, the ultrasound-guided module does not simply output a single image observation result, but rather completes registration, enhancement, segmentation, path construction, path scoring, and baseline image selection based on real-time ultrasound image sequences. This ensures that the puncture target point, puncture path, and pre-perfusion abdominal ultrasound baseline image data all have clear calculation sources and connections, thus providing a consistent data foundation for subsequent puncture patency establishment, drug perfusion, and status assessment. It also helps improve the stability and repeatability of ultrasound-guided puncture path selection and pre-perfusion baseline establishment. In practical applications: for abdominal masses... For patients with mild peristalsis of the intestinal loops around the tumor, ultrasound image sequences within one respiratory cycle can be acquired continuously. After time registration, the outline of the abdominal tumor target area, the set of surrounding organ boundaries, and the candidate access area are segmented. Then, multiple candidate access paths are scored, and the candidate access path with the highest safe puncture channel score is selected as the puncture path. The intersection of this path and the outline of the abdominal tumor target area is taken as the puncture target point. At the same time, the ultrasound image corresponding to the end-expiratory stabilization moment of this path is selected as the pre-perfusion abdominal ultrasound baseline image data for use in subsequent puncture establishment and perfusion processes.

[0030] The puncture and connection module is used to receive the puncture target and puncture path, complete the insertion of the puncture needle under real-time ultrasound guidance, establish an irrigation and drainage channel connecting the puncture needle and the peritoneal dialysis machine, and output the needle tip position confirmation result and the channel position confirmation result. During targeted chemotherapy for abdominal tumors, even if the ultrasound-guided module provides the puncture target and puncture path, if it cannot be continuously confirmed during the puncture process whether the needle body is advancing along the predetermined puncture path, whether the needle tip has truly reached the puncture target, and whether the tip of the puncture needle has entered the target perfusion cavity, the subsequent establishment of the perfusion drainage channel may still encounter problems such as misalignment, entry into non-target cavities, or unstable channel connectivity. Especially when the abdominal cavity is affected by respiratory fluctuations, slight peristalsis of intestinal loops, and changes in probe posture, relying solely on manual observation can easily lead to inconsistent needle tip position judgments. To ensure that the puncture establishment process has the ability to continuously track, determine the position, and verify the establishment, this implementation process is based on the puncture target, puncture path, and real-time ultrasound images. First, the combined tracking of the needle body trajectory and needle tip echo position is completed, then the needle tip position is confirmed, and finally the perfusion drainage channel is established and its effectiveness is confirmed. This implementation process includes the following steps: When the puncture target point, puncture path, and real-time ultrasound image are used as input, the puncture path is first mapped to the coordinates of the current real-time ultrasound image to form a target path line for comparison. During the advancement of the puncture needle, continuous ultrasound images of the puncture area are acquired in real time, and the hyperechoic linear features of the puncture needle and the local strong echo features of the needle tip are extracted in each frame. The needle trajectory is determined by the position of the continuous bright line of the needle body in the image, and the position of the needle tip echo is determined by the echo enhancement point at the front end of the needle body combined with the continuity of the previous and subsequent frames. To avoid misjudgment caused by single-frame noise, the needle trajectory and needle tip echo position in adjacent frames are temporally correlated. If the detection result of the current frame is continuous with the detection result of the previous frame in spatial position and the change amplitude is within a preset range, it is determined to be a valid tracking in the same advancement process. Results; then the path deviation between the current position of the needle tip and the puncture path, and the target deviation between the current position of the needle tip and the puncture target point are calculated. The path deviation is the shortest distance from the current position of the needle tip to the target path line, and the target deviation is the straight-line distance from the current position of the needle tip to the puncture target point. The preset deviation range can be set according to the puncture needle specifications and clinical precision requirements. For example, the upper limit of the path deviation can be set to 1 to 2 times the outer diameter of the puncture needle to ensure that the needle tip does not deviate significantly from the predetermined path. When the path deviation continues to decrease and the target deviation gradually approaches zero, the needle tip alignment status is output as the alignment approaching state, and the corresponding path tracking result is output. When the path deviation continues to increase or the needle trajectory deviates significantly from the target path line, the needle tip alignment status is output as the deviation state, so that subsequent operations can correct the needle posture. After inputting the needle tip alignment, path tracking results, and real-time ultrasound images, the system confirms whether the needle tip has truly entered the target perfusion cavity. Specifically, when the needle tip reaches the puncture target and the path deviation is within a preset range, a local ultrasound image region near the needle tip is captured, and the changes in fluid echogenicity and displacement of surrounding organ boundaries within this region are analyzed. Fluid echogenicity changes are used to determine whether a hypoechoic or anechoic fluid region corresponding to the target perfusion cavity appears in front of the needle tip. This is determined by comparing the average grayscale value and echogenicity within the same neighborhood before and after the needle tip's arrival. If the hypoechoic area in front of the needle tip increases and the region boundary is continuous, it indicates that there is perfusionable space in front of the needle tip. Displacement of surrounding organ boundaries is used to rule out cases where the needle tip is pressing against the organ surface but has not entered the target perfusion cavity. Regarding the specific situation of the perfusion cavity, the displacement of the organ boundary near the needle tip before and after the needle tip is in place can be compared. If the boundary only undergoes local compressive displacement without the expansion of the fluid area in front of the needle tip, it is determined that the needle tip has not entered the target perfusion cavity. If the following conditions are met simultaneously: the current position of the needle tip reaches the puncture target point, the path deviation is within the preset deviation range, the fluid echo expansion occurs in the vicinity of the needle tip, and the boundary of the surrounding organ does not show obvious compression or displacement, it is determined that the tip of the puncture needle has entered the target perfusion cavity, and the output needle tip position confirmation result is confirmed. If only some conditions are met, the output needle tip position confirmation result is pending confirmation or unconfirmed. For example, in practical applications, when the needle tip reaches the puncture target point, if the ultrasound image shows a continuous low-echo area in front of the needle tip and the boundary of the adjacent intestinal loop is not significantly displaced, it can be determined that the needle tip has entered the target perfusion cavity. After inputting the needle tip location confirmation result, real-time ultrasound image, and peritoneal dialysis machine connection status, the perfusion drainage channel is established and its location confirmed. Once the needle tip location is confirmed, the end of the puncture needle is connected to the infusion and return ports of the peritoneal dialysis machine to form a connected perfusion drainage channel. A small dose of test fluid is injected first. The test fluid can be a sterile liquid compatible with the subsequent perfusion system, and its injection volume is a micro-injection method that is observable under ultrasound without creating a significant intra-abdominal pressure burden. After the test fluid is injected, the echo distribution changes around the needle tip are observed in the real-time ultrasound image. If the test fluid forms a continuously expanding low-echo distribution in the vicinity of the needle tip and diffuses along the target perfusion cavity, a normal echo distribution is determined to exist. Simultaneously, the continuity of local fluid diffusion is analyzed, i.e., whether the diffusion boundary of the test fluid is continuous and whether the diffusion direction is consistent with the target perfusion cavity. If the flow path is consistent, and only isolated fluid bags or fluid stagnation in unexpected locations are observed, diffusion continuity is deemed insufficient. The channel connectivity is then confirmed by combining the reflux response results. Specifically, a preset low negative pressure or an open reflux path is applied to the reflux end of the peritoneal dialysis machine to check for stable reflux. If repeatable fluid reflux occurs after injection, and the reflux volume meets the preset response relationship with the injection volume of the test fluid, the perfusion drainage channel is considered to be effectively bidirectionally connected. If reflux is insufficient, interrupted, or ultrasound shows abnormal fluid accumulation near the needle tip, the channel connectivity is deemed abnormal. Based on the test fluid echo distribution, local fluid diffusion continuity, and reflux response results, the output channel position is confirmed as either effectively connected, requiring adjustment, or ineffective. If the connectivity requires adjustment, the puncture needle depth or direction can be fine-tuned before repeating the test fluid verification. If the connectivity is ineffective, the perfusion drainage channel must be re-established. Through the above implementation process, the puncture and drainage module seamlessly connects puncture path execution, needle tip positioning confirmation, and channel effectiveness confirmation. This ensures that the puncture needle is not positioned solely based on a single observation, but rather that the irrigation and drainage channel is established through combined tracking, cavity determination, and trial verification. This helps improve the reliability of needle tip and channel positioning confirmation results and provides a stable foundation for subsequent drug treatment and irrigation / drainage execution. In practical applications: for patients with abdominal tumors adjacent to intestinal loops and experiencing slight displacement due to respiration, the puncture target and drainage path can be established first. The puncture needle is advanced along the puncture path, and the needle trajectory and needle tip echo position are continuously tracked under real-time ultrasound images. The path deviation and target deviation are calculated. After the needle tip reaches the puncture target point, the needle tip is confirmed to have entered the target perfusion cavity by combining the changes in the fluid echo in the vicinity of the needle tip and the displacement changes of the surrounding organ boundaries. Then, the puncture needle is connected to the peritoneal dialysis machine and a small amount of test fluid is injected. If the ultrasound shows that the test fluid continuously diffuses along the target perfusion cavity and there is a stable reflux at the return end, the output channel position is confirmed to be connected and effective, and the subsequent peritoneal chemotherapy drug processing and perfusion drainage process is carried out.

[0031] The drug solution processing module is used to receive the preset chemotherapy regimen and channel location confirmation results, complete the preparation of intraperitoneal chemotherapy drug solution, record the dosage, concentration and preparation time of intraperitoneal chemotherapy drug solution, and output the intraperitoneal chemotherapy drug solution recording results. In targeted chemotherapy for intraperitoneal tumors, if drug processing still relies on manual conversion, manual proportioning, and manual recording after the perfusion drainage channel is effectively established, problems such as incorrect selection of target drug type, inconsistency between target dose and target concentration conversion, and lack of traceability in the preparation process can easily occur. This will affect the subsequent invocation of drug infusion mode and execution parameters by the perfusion drainage execution module. Especially when different patients' intraperitoneal tumor treatment plans differ in drug type, dosage intensity, and dilution ratio, the lack of a unified drug processing procedure can easily lead to a disconnect between information before and after the process. To ensure that the intraperitoneal chemotherapy drug preparation process has a clear analysis, execution, and recording link, this implementation process takes the preset chemotherapy plan and channel location confirmation results as input. First, it generates preparation task information, then completes the intraperitoneal chemotherapy drug preparation and actual parameter calculation, and finally generates the intraperitoneal chemotherapy drug recording results and sends them to the perfusion drainage execution module and the collaborative control module. This implementation process includes the following steps: When using the pre-set chemotherapy regimen and channel location confirmation results as input, the target drug type, target dose, and target concentration in the pre-set chemotherapy regimen are first analyzed. The target drug type determines the name of the chemotherapy drug used in this treatment and the type of diluent used; the target dose determines the total amount of effective drug to be added; and the target concentration determines the drug content per unit volume of the final intraperitoneal chemotherapy solution. During analysis, the treatment items corresponding to this patient in the pre-set chemotherapy regimen can be read as structured fields, such as drug name, dose, concentration, and solvent volume. If the pre-set chemotherapy regimen only provides the dose and concentration, the target total volume can be obtained by dividing the target dose by the target concentration, which serves as the basis for subsequent preparation. The channel location confirmation results are used to confirm that the perfusion drainage channel is... Whether the channel has been established effectively is not determined. Drug preparation parameters will only be generated after the channel location confirmation indicates that the perfusion drainage channel has been effectively established, to avoid wasting drug preparation before the channel is confirmed. Drug preparation parameters must include at least the target drug type, target drug volume, diluent volume, and target total volume. The target drug volume is calculated based on the target dosage and the specifications of the original drug solution. The diluent volume is obtained by subtracting the target drug volume from the target total volume. When the target drug is a lyophilized powder or concentrate, the original solution can be prepared first according to the drug instructions, and then the volume of the original solution can be included in the target drug volume. After completing the above calculations, preparation task information is output. This preparation task information is saved as an executable record and must include at least the drug name, the volume to be added, the diluent volume, the target concentration, and the target total volume. After inputting the preparation task information, the intraperitoneal chemotherapy solution is prepared according to the task information, and the volume of drug solution added, the volume of diluent added, and the time of preparation completion are collected simultaneously. Specifically, first verify that the drug name in the preparation task information matches the label of the drug to be used, then add the target drug solution to the preparation container according to the volume specified in the task information. If volume measurement is used, the volume of drug solution added can be read through syringe graduations, electronic pipettes, or the output of a metering pump. If mass measurement is used, first weigh the empty container, the mass after adding the drug, and the mass after adding the diluent, then convert the volume based on density. After adding the drug solution, add the corresponding volume of diluent; the volume of diluent added is collected directly from the graduated volume or metering device. After all the drug solution and diluent have been added, the preparation container is mixed thoroughly. Mixing methods include gentle shaking, inverting, or low-speed stirring to ensure uniform distribution of the drug in the diluent. Preparation complete. The time point is recorded after mixing is completed and the liquid is visually confirmed to be homogeneous. Then, the actual dose and concentration of the intraperitoneal chemotherapy solution are calculated. The actual dose is the total amount of active ingredient corresponding to the amount of solution added. If the solution is a stock solution, it is obtained by multiplying the drug content per unit volume in the stock solution specification by the actual amount of solution added. The actual concentration is the actual dose divided by the sum of the amount of solution added and the amount of diluent added. If there is a small amount of residue or transfer loss, it can be corrected based on the difference in container volume before and after preparation. After the calculation is completed, the solution preparation result is output, which includes at least the actual dose, actual concentration, actual total volume, and preparation completion time. For example, when the target solution is a chemotherapy drug, the target dose is a certain number of milligrams, and the target concentration is a certain number of milligrams per milliliter, the target total volume can be calculated first, then the corresponding volume can be drawn according to the stock solution specification and added to the preparation container, and diluent can be added to the target total volume to obtain the final solution preparation result. After inputting the drug preparation result and preparation completion time, the actual dose, concentration, and preparation time of the intraperitoneal chemotherapy drug are linked and recorded, generating an intraperitoneal chemotherapy drug record result, which is then output to the perfusion and drainage execution module and the collaborative control module. Specifically, during recording, the actual dose, concentration, total volume, and preparation completion time from the drug preparation result are written into the same record, and a correspondence is established with the current patient identifier, the preset chemotherapy regimen identifier, and the channel location confirmation result, enabling subsequent modules to directly read drug information consistent with the current treatment process. To ensure the traceability of the linked records, the type of drug, diluent, operator identifier, or preparation equipment identifier can also be additionally recorded, but this is not applicable to the actual... During the treatment, the actual dose, actual concentration, and preparation time are the core recorded information. After generating the intraperitoneal chemotherapy drug solution record, the result is sent to the perfusion and drainage execution module. This module then determines the trial injection mode, drug infusion mode, or the handling method for any remaining drug solution before drainage at the end of treatment based on the actual dose and concentration. Simultaneously, the result is sent to the collaborative control module. This module uses the actual dose, actual concentration, and preparation time as one of the control criteria when generating control commands for the drug infusion and drainage stages. The preparation time can be used to determine whether the drug solution is within the permitted usage period. If the preset usage time limit is exceeded, the collaborative control module can restrict entry into the formal drug infusion stage. Through the above implementation process, the drug processing module continuously connects the analysis of the preset chemotherapy regimen, the execution of drug preparation, the calculation of actual parameters, and the recording of results. This ensures that the intraperitoneal chemotherapy drug record is no longer simply manually registered information, but an execution result with a clear calculation source and module transmission relationship. This helps improve the standardization, verifiability, and consistency of the intraperitoneal chemotherapy drug processing process with subsequent perfusion and drainage execution and collaborative control. In practical application: after the channel location confirmation result indicates that the perfusion and drainage channel has been effectively established, the drug processing module reads the preset chemotherapy regimen corresponding to the current patient, analyzes the target drug type, target dose, and target concentration, and converts them into preparation task information. Subsequently, the chemotherapy drug and diluent are added according to the preparation task information, and after mixing, the amount of drug added, the amount of diluent added, and the time of preparation completion are collected to calculate the actual dose and actual concentration. Finally, the actual dose, actual concentration, and preparation time are written into the intraperitoneal chemotherapy drug record result and sent to the perfusion and drainage execution module and the collaborative control module, so that the generation of subsequent drug infusion and control instructions is based on the same drug processing result.

[0032] The perfusion and drainage execution module is used to receive the needle tip position confirmation result, channel position confirmation result, peritoneal chemotherapy drug solution recording result, and control commands output by the collaborative control module. It controls the peritoneal dialysis machine to perform trial injection, drug infusion, auxiliary drainage, and treatment end drainage through the perfusion and drainage channel. It also responds to control commands to perform pause and hold, flow rate adjustment, and infusion direction adjustment, and outputs execution status and execution parameters. During targeted chemotherapy for peritoneal tumors, even after obtaining confirmation results for needle tip location, channel location, and peritoneal chemotherapy drug recordings, if the perfusion and drainage execution process simply starts the peritoneal dialysis machine according to a fixed procedure without uniformly addressing the consistency between input information, the adaptation to the current treatment stage, and the dynamic adjustment needs during execution, problems such as inappropriate mode selection, mismatch between drug infusion and drainage, and inconsistencies between execution parameters and actual status can easily arise. Especially when transitioning from trial injection to drug infusion, or inserting drainage during drug infusion, the lack of clear mode switching criteria and parameter transmission rules makes it difficult for the subsequent status determination module and collaborative control module to obtain accurate and continuous execution feedback. Therefore, this implementation process is based on the needle tip location confirmation results, channel location confirmation results, peritoneal chemotherapy drug recordings, and control commands output by the collaborative control module. It first completes consistency verification and mode determination, then completes peritoneal dialysis machine execution and dynamic adjustment, and finally generates the execution status and execution parameters corresponding to the current mode. This implementation process includes the following steps: When the needle tip position confirmation result, channel position confirmation result, intraperitoneal chemotherapy drug recording result, and control command output by the collaborative control module are used as inputs, a consistency check is performed first. The consistency check includes at least three parts: position consistency, channel consistency, and drug consistency. Position consistency is used to confirm that the needle tip position confirmation result indicates that the needle tip has entered the target perfusion cavity. If the needle tip position confirmation result is unconfirmed or pending confirmation, entering the drug infusion mode is not allowed. Channel consistency is used to confirm that the channel position confirmation result indicates that the perfusion drainage channel is effectively connected. If the channel position confirmation result is connected but needs adjustment or is ineffective, only entering the trial injection mode or adjustment waiting state is allowed. Drug consistency is used to confirm that the actual dose, actual concentration, and preparation time in the intraperitoneal chemotherapy drug recording result meet the current treatment requirements. Among them, the preparation time must be within the preset usage time limit, and the actual dose and actual concentration must match the treatment stage corresponding to the current control command of the collaborative control module. For example, when the control command output by the collaborative control module indicates that the formal drug infusion stage should be entered, the intraperitoneal chemotherapy drug recording result must have a valid actual dose and actual concentration. Concentration records are kept accurate, and preparation time does not exceed the allowable duration. After completing the above consistency verification, the execution mode corresponding to the peritoneal dialysis machine is determined based on the consistency verification results. If the needle tip position confirmation result and channel position confirmation result are both valid, but the intraperitoneal chemotherapy drug solution recording result is not yet ready, it is determined to be a trial injection mode. If all three are valid and the control command indicates that formal infusion has begun, it is determined to be a drug infusion mode. If the control command requires the simultaneous release of local intraperitoneal fluid during infusion, it is determined to be a combined drainage mode. If the control command indicates that the target area coverage has reached the requirement and the process has entered the termination phase... If a segment is selected, it is determined to be the drainage mode at the end of treatment. After the mode is determined, mode execution instructions and initial execution parameters are generated. The mode execution instructions are used to instruct the peritoneal dialysis machine to enter the corresponding mode. The initial execution parameters include at least the initial flow rate, target infusion volume, drainage opening / closing status, and infusion direction setting. The initial flow rate can be selected based on the actual concentration in the peritoneal chemotherapy drug recording results and the preset rate range of this treatment plan. The target infusion volume can be determined based on the amount of drug to be completed in the current stage. The drainage opening / closing status and infusion direction setting are determined based on the control instructions output by the collaborative control module. After inputting the mode execution command, initial execution parameters, and control commands, the peritoneal dialysis machine is controlled to perform trial infusion, drug infusion, auxiliary drainage, or post-treatment drainage via the perfusion and drainage channel. When entering trial infusion mode, the peritoneal dialysis machine outputs a small amount of test fluid or compatible fluid at a low flow rate to further observe the response of the perfusion and drainage channel to fluid entering the peritoneal cavity. When entering drug infusion mode, the peritoneal dialysis machine begins infusing peritoneal chemotherapy drugs according to the initial flow rate and target infusion volume in the initial execution parameters. When entering auxiliary drainage mode, the peritoneal dialysis machine maintains drug infusion while opening the drainage path or switching to a preset drainage ratio to achieve parallel execution of perfusion and drainage. When entering post-treatment drainage mode, the peritoneal dialysis machine stops increasing drug input and drains the remaining fluid at a preset drainage rate. During execution, the perfusion and drainage execution module continuously receives output from the collaborative control module. The system executes control commands and dynamically adjusts the pause / hold, flow rate, and infusion direction of the peritoneal dialysis machine based on these commands. Pause / hold is achieved by temporarily stopping pumping while maintaining the current channel connectivity, allowing the status assessment module to reassess the drug diffusion status. Flow rate adjustment is achieved by changing the pumping volume per unit time, and can be done in stages, such as increasing or decreasing the current flow rate by a preset percentage. Infusion direction adjustment is used to adapt to multi-channel or switchable outlet direction structures. When the peritoneal dialysis machine or connecting channel has different outlet directions, the perfusion and drainage execution module switches the output channel according to the control commands. During execution, the system synchronously records the real-time execution status and parameters. The real-time execution status includes at least the current execution mode, execution start time, and pause or run flags. The real-time execution parameters include at least the current pump speed, current cumulative infusion volume, and current drainage setting. After taking the real-time execution status, real-time execution parameters, and peritoneal dialysis machine operation feedback results as inputs, the infused volume, current flow rate, cumulative drainage volume, and channel response status in the current mode are calculated, and the execution status and execution parameters corresponding to the current mode are generated and output to the status determination module and the collaborative control module. The infused volume is calculated through the peritoneal dialysis machine pump count or time integration; specifically, the current flow rate can be multiplied by the continuous running time and the pause period subtracted to obtain the amount of liquid actually entering the perfusion drainage channel in the current mode. The current flow rate is directly read from the current pumping rate of the peritoneal dialysis machine; if the flow rate is adjusted during execution, the latest adjusted rate is used as the current flow rate. The cumulative drainage volume is obtained by accumulating the liquid metering results at the drainage end, which can be determined using drainage bag calibration, weighing conversion, or feedback values ​​from the peritoneal dialysis machine's internal metering unit. The channel response status is used to characterize whether the perfusion drainage channel's execution response to the current mode is normal; specifically, it can be combined with pump pressure fluctuations and flow rate data from the peritoneal dialysis machine operation feedback results. The stability of the flow rate and the continuity of the return flow are determined. For example, when the pump pressure is stable, the deviation between the actual output flow rate and the set flow rate is within the preset range, and the drainage feedback is continuous, the channel response is judged to be normal. If the pump pressure rises abnormally, the flow rate is significantly lower than the set value, or the return flow is intermittent, the channel response is judged to be abnormal. Based on this, the perfusion drainage execution module generates the corresponding execution status and execution parameters according to the current mode. When it is the test injection mode, the focus is on outputting the start time of the test injection, the test injection volume, and the test injection response status. When it is the drug infusion mode, the focus is on outputting the infused volume, the current flow rate, and the infusion continuity status. When it is the combined drainage mode or the treatment end drainage mode, the focus is on outputting the cumulative drainage volume, the current drainage status, and the channel response status. The generated execution status and execution parameters are sent to the status judgment module to judge the drug diffusion status, target area coverage status, and abnormal status in conjunction with real-time ultrasound images and vital sign information. At the same time, they are sent to the collaborative control module for subsequent control command updates. Through the above implementation process, the perfusion and drainage execution module seamlessly connects mode determination, peritoneal dialysis machine execution, dynamic adjustment, and feedback calculation. This ensures that the execution behavior of the peritoneal dialysis machine at different treatment stages has clear entry conditions, parameter sources, and feedback outputs. This helps improve the continuity of switching between trial injections, drug infusions, combined drainage, and post-treatment drainage, as well as the consistency of execution status and parameters in supporting subsequent status determination and coordinated control. In practical applications: when the needle tip position confirmation result indicates that the needle tip has entered the target perfusion cavity, the channel position confirmation result indicates that the perfusion and drainage channel is effectively connected, and the peritoneal chemotherapy drug recording result indicates… Once the medication solution has been prepared and is within its permissible usage period, the perfusion and drainage execution module first completes a consistency check and determines to enter the medication infusion mode. Subsequently, it controls the peritoneal dialysis machine to infuse the medication solution according to the initial execution parameters. If the co-control module issues a control command to reduce the flow rate and initiate co-drainage during the infusion process, the perfusion and drainage execution module switches to the co-drainage mode and synchronously updates the real-time execution status and real-time execution parameters. After the target area coverage meets the requirements, it enters the treatment end drainage mode according to the control command and outputs the infused volume, cumulative drainage volume, and channel response status to the status determination module and the co-control module for subsequent processing.

[0033] The status determination module is used to receive real-time ultrasound images, pre-perfusion abdominal ultrasound baseline image data, execution status, execution parameters, and patient blood pressure, heart rate, respiratory rate and abdominal pressure, and to jointly determine the drug diffusion status, target coverage status and abnormal status, and output the target coverage status, abnormal status and status change trend. In targeted chemotherapy for peritoneal tumors, status determination is not merely about observing changes in fluid echogenicity within the peritoneum. It requires answering three consecutive questions: how is the drug spreading within the peritoneum? Does this spreading truly cover the target area of ​​the peritoneal tumor? And does the current spreading status carry safety risks and require correction? Relying solely on a single ultrasound image or a single abdominal pressure value can easily lead to misjudging local fluid accumulation as effective coverage, misjudging short-term fluctuations as abnormalities, or continuing to use old status conclusions even after control commands have changed execution parameters. Therefore, this implementation constructs a progressively layered determination module. First, it extracts the drug spreading status, target coverage status, and abnormal status based on real-time ultrasound images and pre-perfusion peritoneal ultrasound baseline data. Then, it further obtains the spreading offset results, coverage offset status, and updated status change trends. Finally, it combines execution parameters to construct a drug spreading response prediction result, correcting the aforementioned statuses and providing the collaborative control module with a more accurate assessment basis for the actual treatment process. This implementation includes the following steps: After inputting real-time ultrasound images, pre-perfusion peritoneal ultrasound baseline image data, and execution parameters, the real-time ultrasound images and pre-perfusion peritoneal ultrasound baseline image data are first registered and compared. The pre-perfusion peritoneal ultrasound baseline image data is the ultrasound image corresponding to the stable puncture path in the non-perfusion state, used as a reference image for subsequent diffusion identification. During registration, rigid registration is first performed based on the abdominal wall layer, the outline of the abdominal tumor target area, and the boundaries of surrounding stable organs. Then, minor corrections are made to slightly deformed areas to ensure that the real-time ultrasound images and pre-perfusion peritoneal ultrasound baseline image data are in the same coordinate system. After registration, differences are extracted between the two, identifying newly added hypoechoic or anechoic fluid-filled areas in the difference regions. The drug diffusion region is selected as a candidate region for drug diffusion. Then, the changes in the drug diffusion boundary, diffusion direction, and diffusion range are determined by comparing consecutive frame images. The drug diffusion boundary is determined by the outer edge of the newly added liquid region; the diffusion direction is determined by the direction of centroid movement and boundary advancement of the diffusion region at consecutive time points; and the change in diffusion range is obtained by the area difference of the diffusion region at adjacent time points. Based on this, the drug diffusion state is output, which includes at least the current diffusion region, the main diffusion direction, and the change in diffusion range. Subsequently, the drug diffusion state, real-time ultrasound images, and the abdominal tumor target area are used as inputs to calculate the boundary encirclement relationship and regional overlap relationship between the drug diffusion region and the abdominal tumor target area. The boundary encirclement relationship is used to reflect the drug diffusion... The degree to which the fluid diffusion area surrounds the outer edge of the abdominal tumor target area can be determined by uniformly selecting multiple contour points along the outline of the abdominal tumor target area and counting the proportion of points located inside or near the boundary of the fluid diffusion area. The regional overlap relationship reflects the overlap ratio between the fluid diffusion area and the projected area of ​​the abdominal tumor target area, which can be obtained by dividing the overlapping area by the area of ​​the abdominal tumor target area. When both the boundary coverage and the regional overlap ratio reach a preset threshold, the target area coverage is considered sufficient; if only partially reached, it is considered insufficient coverage; if significantly deviated, it is considered ineffective coverage. The fluid diffusion status, execution status, and patient blood pressure, heart rate, respiratory rate, and abdominal pressure are then used as inputs to analyze signs of fluid leakage. Correlation analysis was performed between signs of organ compression and fluctuations in vital signs. Drug leakage was identified by the sudden appearance of new fluid-filled echoes outside the target perfusion area, accompanied by irregular outward expansion of the boundaries. Organ compression was identified by unilateral displacement, local deformation, or pressure adhesion of adjacent organ boundaries within a short period. Fluctuations in vital signs were determined by the changes in current blood pressure, heart rate, respiratory rate, and abdominal pressure relative to the pre-perfusion baseline. When image abnormalities and fluctuations in vital signs occurred simultaneously within the same time window, the abnormality level was increased. This resulted in the output of the abnormal state and its trend, where the trend reflected whether the abnormality or coverage change was strengthening, weakening, or remaining relatively stable compared to the previous judgment time. After obtaining the drug diffusion state, target coverage state, and abnormal state, real-time ultrasound images, execution status, execution parameters, and drug diffusion state are used as inputs to further extract the positional changes of the drug diffusion front, the migration direction of the diffusion center, and the diffusion velocity changes at continuous time intervals, thus constructing the drug diffusion evolution trajectory. The drug diffusion front refers to the outermost frontal boundary of the drug diffusion boundary towards the main diffusion direction, and its positional change can be determined by the distance difference between the frontal boundary and the same reference coordinate at adjacent time intervals. The diffusion center is represented by the geometric center or area-weighted center of the current diffusion region, and the migration direction of the diffusion center is determined by the continuous... The direction of the line connecting the centers at different times is determined; the change in diffusion rate is obtained by dividing the increase in diffusion range between adjacent times by the time interval; connecting the aforementioned changes in the leading edge position, the migration direction of the diffusion center, and the change in diffusion rate in chronological order forms the drug diffusion trajectory; in practical applications, if the drug propagates towards the periphery of the abdominal tumor target area along the expected direction, the drug diffusion trajectory shows a stable convergence; if the drug continuously deviates towards adjacent low-resistance cavities, it shows a significant deviation; subsequently, using the diffusion deviation result, target area coverage status, and the boundary of the abdominal tumor target area as inputs, the relationship between the drug diffusion trajectory and the boundary of the abdominal tumor target area is calculated. The degree of convergence and deviation between the target and the target area is calculated. The degree of convergence can be determined by the reduction in the closest distance from the trajectory endpoint to the boundary of the abdominal tumor target area; the more significant the reduction, the higher the degree of convergence. The degree of deviation can be determined by the angle or lateral offset distance between the trajectory direction and the reference direction pointing to the center of the abdominal tumor target area; the larger the angle or the greater the offset, the higher the degree of deviation. After combining the target area coverage status, the coverage offset status is output, which is at least divided into three categories: approaching coverage, local offset, and significant offset. Then, the coverage offset status, abnormal status, and continuous changes in the patient's abdominal pressure and vital signs are used as inputs to calculate the current... The risk enhancement trend during the infusion process is analyzed. The continuous changes in abdominal pressure and vital signs are represented by a time series consisting of the current moment and several previous judgment moments. If abdominal pressure continues to rise and the coverage shift changes from near coverage to local shift or significant deviation, while the abnormal state changes from low risk to medium-high risk, then the risk enhancement trend is considered significant. If abdominal pressure changes gradually and the coverage shift improves, then the risk enhancement trend is considered reduced. The updated state change trend is then output to replace the aforementioned coarser-grained trend results, enabling subsequent collaborative control modules to control based on diffusion evolution rather than single-point states. After obtaining the diffusion offset results, coverage offset status, and updated status change trends, the execution parameters, diffusion offset results, and coverage offset status are used as inputs to construct the drug diffusion response prediction result corresponding to the current execution parameters. The drug diffusion response prediction result is then compared with the actual diffusion result represented by real-time ultrasound images, and the diffusion response deviation result is output. The execution parameters include at least the current flow rate, infused volume, drainage opening / closing status, and current infusion direction. The drug diffusion response prediction result is not an abstract prediction, but rather a diffusion expectation for the next time period constructed based on the correspondence between the current execution parameters and historically observed diffusion results; specifically, a nearest-neighbor time period extrapolation method can be used. This method uses actual diffusion changes under similar flow rates, infused volumes, and drainage conditions within the most recent judgment periods as the basis for prediction. It then obtains the predicted diffusion front location, predicted diffusion center migration direction, and predicted diffusion range increase for the next judgment period. This prediction is compared with the actual diffusion results represented by real-time ultrasound images to obtain the diffusion response deviation. If the actual diffusion lags significantly behind the predicted diffusion, it indicates insufficient current infusion response; if the actual diffusion deviates significantly from the predicted direction, it indicates changes in current cavity conditions or channel response. Subsequently, the diffusion response deviation results, abnormal conditions, and continuous changes in the patient's abdominal pressure and vital signs are used as inputs to determine... The system checks whether the current abnormal state matches the current execution parameters and outputs the abnormality correction result. For example, if the execution parameter is only a low-flow-rate infusion, but the abnormal state is characterized by high-intensity organ compression accompanied by a rapid increase in abdominal pressure, it indicates that the abnormal state does not match the current execution parameters, and there may be localized unexpected accumulation or image recognition deviation. In this case, the output abnormality correction result is that the abnormality level needs to be increased and an execution deviation should be indicated. Conversely, if the execution parameter is at a higher flow rate but vital signs are stable, abdominal pressure has not increased, and the diffusion response deviation is small, the abnormal state can be maintained or appropriately reduced. Finally, the abnormality correction result, target area coverage status, and updated state change trend are used as inputs to adjust the target area coverage. The system corrects the target area coverage and abnormal states, outputting the corrected target area coverage, abnormal states, and state change trends. The correction logic is as follows: when the diffusion response deviation indicates that the actual diffusion cannot support the original coverage conclusion, the target area coverage is downgraded; when the abnormality correction indicates that the current abnormality is underestimated or overestimated, the abnormal state is adjusted accordingly; after both coverage and abnormality are corrected, the state change trend is recalibrated as enhanced, weakened, or stabilized. After this correction layer, the state determination result no longer depends solely on the current image observation, but introduces the matching relationship between the execution parameters and the diffusion response, making the output result more suitable as the input of the collaborative control module. Through the above implementation process, the state determination module sequentially connects registration comparison, coverage determination, anomaly identification, diffusion evolution analysis, and response deviation correction. This allows the drug diffusion state, target area coverage state, and abnormal state to not only be extracted from real-time ultrasound images but also to be corrected and updated in conjunction with execution parameters and continuous changes. This helps improve the accuracy of the determination of the true coverage of the peritoneal tumor target area and changes in abnormal risks, and provides a continuous and consistent state basis for subsequent control command generation. In practical applications: when the peritoneal dialysis machine enters the drug infusion mode, the state determination module first registers and compares the real-time ultrasound image with the pre-infusion peritoneal ultrasound baseline image data to obtain the drug diffusion status. The diffusion status is then assessed, and the target coverage status is calculated based on the contour of the abdominal tumor target area. Abnormal statuses are identified by combining abdominal pressure and vital signs. Subsequently, the positional changes of the drug diffusion front and the migration direction of the diffusion center are continuously tracked to obtain the coverage offset status and the updated status change trend. If the collaborative control module has issued a flow rate adjustment or drainage coordination command, the status determination module further constructs a drug diffusion response prediction result based on the latest execution parameters, and corrects the target coverage status and abnormal status by comparing it with the actual diffusion results. After this continuous determination, the collaborative control module can select a more appropriate flow rate adjustment, infusion direction adjustment, or drainage parameter adjustment action based on the corrected status results.

[0034] The collaborative control module is used to receive the recording results of intraperitoneal chemotherapy drug solution, execution status, execution parameters, target coverage status, abnormal status and status change trend, generate control instructions for the drug infusion stage and drainage stage, and send the control instructions to the perfusion and drainage execution module to achieve collaborative control of perfusion and drainage. In targeted chemotherapy for abdominal tumors, even if the status determination module has output the target coverage status, abnormal status, and status change trend, if the collaborative control still uses fixed rules or a single threshold triggering method, two problems are likely to occur: First, when the current treatment has transitioned from the drug infusion stage to the drainage stage, the control logic of the previous stage is still used, resulting in a mismatch between the control actions and the treatment stage; second, after the control command is issued, if the candidate control action set is not corrected based on execution feedback, subsequent control may still repeatedly select actions with poor effects. Therefore, this implementation process enables the collaborative control module to first complete the stage determination and candidate control action generation, then complete the target control action selection and control command generation, and then further perform effect write-back, priority update, and iterative control on the candidate control action set based on actual execution feedback, thereby forming a closed-loop adaptive collaborative control around the target coverage status and abnormal status; this implementation process includes the following steps: After inputting the intraperitoneal chemotherapy drug recording results, execution status, execution parameters, target coverage status, abnormal status, and status change trends, the current treatment process is first determined to belong to either the drug infusion stage or the drainage stage. This stage determination is not solely based on chronological order, but rather on the dosage, concentration, and preparation time information from the intraperitoneal chemotherapy drug recording results, combined with the execution status and execution parameters. When the intraperitoneal chemotherapy drug recording results indicate the availability of usable drug, the execution status indicates the peritoneal dialysis machine is in infusion-related mode, the infused volume in the execution parameters has not reached the planned value for the current stage, and the target coverage status has not yet met the preset requirements, the current treatment process is determined to be in the drug infusion stage. When the execution status indicates the peritoneal dialysis machine is in drainage-related mode, or the target coverage status has met the preset requirements and an abnormal status indicates the need for release of the peritoneal dialysis drug, the current treatment process is determined to be in the drug infusion stage. When there is local fluid overload in the cavity, the current treatment process is determined to be in the drainage phase. After the phase determination is completed, a set of candidate control actions for the corresponding phase is generated based on the dosage, concentration, and preparation time information in the intraperitoneal chemotherapy drug recording results. Among them, the dosage information is used to limit the range of infusion volume that can still be executed, the concentration information is used to limit the flow rate adjustment range, and the preparation time information is used to determine whether the drug solution is still within the allowable usage period. In the drug infusion phase, the set of candidate control actions includes at least pause and hold, flow rate increase, flow rate decrease, and infusion direction adjustment. In the drainage phase, the set of candidate control actions includes at least drainage initiation, drainage rate increase, drainage rate decrease, and drainage hold. If the preparation time is close to the allowable usage period, the candidate control actions that can maintain the continuity of treatment are prioritized and retained. Thus, the phase determination result and the set of candidate control actions are output. After taking the stage judgment results, candidate control action set, target coverage status, abnormal status, and status change trend as input, the comprehensive matching result of each candidate control action on target coverage maintenance, abnormal risk suppression, and treatment continuity is calculated, and the target control action is output. Among them, target coverage maintenance is used to evaluate the possibility of maintaining or improving the current target coverage status after the execution of a candidate control action. If the target coverage status is insufficient and the status change trend indicates that the coverage is improving, the matching result of slightly increasing the flow rate or adjusting the infusion direction is relatively high. If the target coverage status is close to sufficient coverage, the matching result of pausing and maintaining or maintaining the current parameters is relatively high. Abnormal risk suppression is used to evaluate the degree of relief of the current abnormal status by the candidate control action. If the abnormal status is manifested as increased abdominal pressure or increased organ compression, the matching result of decreasing the flow rate, pausing and maintaining, or initiating drainage is relatively high. Treatment continuity is used to evaluate the ability of the candidate control action to maintain treatment progress without interrupting the existing treatment chain. If the abnormal status is mild and the status change trend is stable, the matching result is relatively high. If the desired outcome is not achieved, then flow rate fine-tuning or drainage parameter adjustment actions that do not require interruption of treatment will be prioritized. The comprehensive matching result can be determined by weighting, that is, by assigning weights to target coverage maintenance, abnormal risk suppression, and treatment continuity respectively, and then summing and comparing them. The weights can be dynamically adjusted according to the stage judgment results. For example, the weight of target coverage maintenance can be increased during the drug infusion stage, and the weight of abnormal risk suppression can be increased during the drainage stage. The candidate control action with the highest comprehensive matching result is determined as the target control action. Then, the target control action, execution status, and execution parameters are used as inputs to generate control instructions corresponding to the perfusion and drainage execution module and send them to the perfusion and drainage execution module. When the target control action belongs to the drug infusion stage, the control instruction corresponds to a pause and hold instruction, flow rate adjustment instruction, or infusion direction adjustment instruction. When the target control action belongs to the drainage stage, the control instruction corresponds to a drainage start instruction or drainage parameter adjustment instruction. In addition to the action type, the control instruction can also carry the adjustment range, duration, and switching conditions so that the perfusion and drainage execution module can directly call it. After generating and issuing control commands, the target control action, execution status, execution parameters, and status change trends are used as inputs to construct the stage response result corresponding to the target control action. This stage response result is then compared with the actual execution status and parameters fed back by the infusion and drainage execution module, outputting the control response deviation result. The stage response result represents the expected intra-stage response brought about by the target control action under the current execution status, execution parameters, and status change trends. For example, a decrease in expected abdominal pressure increase after flow rate reduction, a decrease in expected diffusion offset after infusion direction adjustment, and a decrease in expected abnormal state after drainage initiation. The stage response result can be constructed based on the change patterns in historical adjacent control cycles under the current stage, i.e., reading the status change trends before and after the target control action execution to form an expectation for the next control cycle. When the difference between the actual execution status and parameters fed back by the infusion and drainage execution module and the stage response result is small, it indicates that the target control action execution effect meets expectations; when the difference is large, it indicates a control response deviation. For example, if the actual execution parameters have been correctly reduced after flow rate reduction, but the abnormal state has not improved and the status change trend continues to strengthen, then the control response deviation result is too large. Subsequently, the control response deviation... Taking the poor results, target coverage status, and abnormal status as input, the system performs effect write-back and priority update on each candidate control action in the candidate control action set, outputting an updated set of candidate control actions. Effect write-back records the actual effect of the current target control action back to the corresponding candidate control action entry as a reference for subsequent actions under similar conditions. Priority update adjusts based on the control response deviation results. If a candidate control action improves target coverage and suppresses abnormal status after multiple executions, its priority is increased; if a candidate control action repeatedly exhibits large control response deviations, its priority is decreased. Finally, the updated set of candidate control actions, the stage determination results, and the intraperitoneal chemotherapy drug recording results are used as inputs to recalculate the suitability of each candidate control action for the current stage and generate iterative control instructions to be sent to the perfusion and drainage execution module. The suitability is still calculated based on target coverage maintenance, abnormal risk suppression, and treatment continuity, but the updated candidate control action priority has been introduced at this time, so that the new control instructions no longer depend solely on the current single state, but can be adaptively updated based on recent execution feedback. This allows the control instructions for the drug infusion stage and the drainage stage to be continuously corrected with execution feedback. Through the above implementation process, the collaborative control module continuously connects stage determination, candidate control action generation, target control action selection, control command issuance, and effect write-back and iterative updates based on execution feedback. This allows the collaborative control of perfusion and drainage to move beyond static rule matching and instead form a feedback-enabled and correctable control chain around target area coverage status, abnormal states, and state change trends. This helps improve the matching degree between control action selection and actual state at different treatment stages and enhances the ability of control commands to adaptively update with execution feedback. In practical applications: when the state determination module outputs a target area coverage status of insufficient coverage, an abnormal state of moderate risk, and a state change trend of increasing... The collaborative control module first determines that it is currently in the drug infusion stage, then selects the flow rate reduction and subsequent directional adjustment as the target control action from the candidate control action set, and generates the corresponding control command and sends it to the perfusion and drainage execution module. If the perfusion and drainage execution module reports that the actual execution parameters have changed, but the subsequent state change trend has not improved, the collaborative control module writes back the effect of the target control action as inefficient and lowers its priority. In the next control cycle, it recalculates the adaptation degree and then generates iterative control commands for pausing and maintaining or starting drainage. When the target area coverage reaches the preset requirements, it switches to the drainage stage control logic according to the stage determination result and adaptively adjusts the drainage rate and duration.

[0035] Working principle: First, continuous ultrasound observation of the abdominal cavity is used to determine the location of the abdominal tumor, the boundaries of surrounding organs, and a suitable puncture channel. Based on this, the puncture target, puncture path, and pre-perfusion image baseline are determined. Then, the puncture is performed along the puncture path, continuously confirming whether the needle tip has truly entered the target perfusion area, and simultaneously checking whether the perfusion drainage channel has been successfully established. After the channel is confirmed to be effective, the chemotherapy solution is prepared according to the preset chemotherapy regimen, and the actual dose, actual concentration, and preparation time of the solution are recorded. Then, the peritoneal dialysis machine performs the trial infusion, formal infusion, drainage, and termination of drainage. Throughout the process, the system continuously combines ultrasound images, execution status, execution parameters, and patient vital signs to assess the diffusion of the solution in the abdominal cavity, the coverage of the tumor area, and whether leakage, organ compression, or abnormal abdominal pressure occurs. Finally, based on these assessment results, the infusion rate, infusion direction, pause / hold, or drainage parameters are dynamically adjusted to ensure that the solution covers the tumor area as much as possible while keeping the treatment risks within a safe range. For example, if a patient's abdominal tumor is close to an intestinal loop and the abdominal cavity is irregularly shaped, the medication may easily diffuse to areas with less resistance. In such cases, relying solely on experience for infusion may result in the medication entering the abdominal cavity without actually covering the tumor. Using this approach, the system first selects a safer puncture path using ultrasound, then continuously observes the needle tip position during the puncture. Once the needle tip is confirmed to be in the target perfusion area, a small amount of fluid is used to test the flow and confirm the perfusion drainage channel is effective. Subsequently, the chemotherapy medication is prepared according to the pre-set protocol and infused via peritoneal dialysis. During the infusion process, the system simultaneously monitors whether the medication is spreading around the tumor via ultrasound, and assesses the current status by combining information such as abdominal pressure, heart rate, and respiration. If the medication is found to be spreading in the wrong direction, the system will adjust the infusion rate or direction. If the abdominal pressure is found to be increasing or there is a tendency for leakage, the system will pause the infusion or initiate drainage. Once the medication has achieved the expected coverage of the tumor area, the drainage process will be terminated. In this way, the entire treatment process is not simply about injecting medication, but about observing, assessing, and adjusting simultaneously, making the treatment more closely aligned with the actual situation.

[0036] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An integrated system for peritoneal perfusion and drainage based on ultrasound guidance, comprising an ultrasound guidance module, a puncture and perfusion module, a drug treatment module, a perfusion and drainage execution module, a status determination module, and a collaborative control module, characterized in that: The ultrasound guidance module is used to acquire real-time ultrasound images of the patient's abdominal cavity, and based on the real-time ultrasound images, identify the target area of ​​the abdominal tumor, the boundaries of surrounding organs, and the safe puncture channel, and output the puncture target point, puncture path, and pre-perfusion abdominal ultrasound baseline image data. The puncture and connection module is used to receive the puncture target and puncture path, complete the insertion of the puncture needle under real-time ultrasound guidance, establish an irrigation and drainage channel connecting the puncture needle and the peritoneal dialysis machine, and output the needle tip position confirmation result and the channel position confirmation result. The drug solution processing module is used to receive the preset chemotherapy regimen and channel location confirmation results, complete the preparation of intraperitoneal chemotherapy drug solution, record the dosage, concentration and preparation time of intraperitoneal chemotherapy drug solution, and output the intraperitoneal chemotherapy drug solution recording results. The perfusion and drainage execution module is used to receive the needle tip position confirmation result, channel position confirmation result, peritoneal chemotherapy drug solution recording result, and control commands output by the collaborative control module. It controls the peritoneal dialysis machine to perform trial injection, drug infusion, auxiliary drainage, and treatment end drainage through the perfusion and drainage channel. It also responds to control commands to perform pause and hold, flow rate adjustment, and infusion direction adjustment, and outputs execution status and execution parameters. The status determination module is used to receive real-time ultrasound images, pre-perfusion abdominal ultrasound baseline image data, execution status, execution parameters, and patient blood pressure, heart rate, respiratory rate and abdominal pressure, and to jointly determine the drug diffusion status, target coverage status and abnormal status, and output the target coverage status, abnormal status and status change trend. The collaborative control module is used to receive the recording results of intraperitoneal chemotherapy drug solution, execution status, execution parameters, target coverage status, abnormal status and status change trend, generate control instructions for the drug infusion stage and drainage stage, and send the control instructions to the perfusion and drainage execution module to achieve collaborative control of perfusion and drainage.

2. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 1, characterized in that, The ultrasound guidance module is used to: take the real-time ultrasound image sequence of the patient's abdominal cavity as input, perform temporal registration and boundary enhancement processing on the ultrasound images at consecutive time moments, segment the abdominal tumor target area, the boundaries of surrounding organs and the abdominal wall access area, and output the outline of the abdominal tumor target area, the set of boundaries of surrounding organs and the candidate access area. The peritoneal tumor target area contour, the set of surrounding organ boundaries, and the candidate access route region are used as inputs to construct candidate connectivity paths from the candidate access route region to the peritoneal tumor target area contour. The organ avoidance distance, path curvature change, and cross-time path stability corresponding to each candidate connectivity path are calculated, and the safe puncture channel score and the candidate connectivity path with the highest score are output. Using the candidate connectivity path with the highest score and the real-time ultrasound image sequence as input, the intersection of the candidate connectivity path and the peritoneal tumor target area contour is determined as the puncture target point, the candidate connectivity path is determined as the puncture path, and the ultrasound image corresponding to the moment with the highest path stability is extracted as the pre-perfusion peritoneal ultrasound baseline image data.

3. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 2, characterized in that, The puncture and connection module is used to: take the puncture target point, puncture path and real-time ultrasound image as input, jointly track the needle body trajectory and needle tip echo position during the advancement of the puncture needle, calculate the path deviation between the current position of the needle tip and the puncture path and the target deviation between the current position of the needle tip and the puncture target point, and output the needle tip alignment status and path tracking results. The needle tip alignment status, path tracking results, and real-time ultrasound images are used as inputs. When the needle tip reaches the puncture target at its current position and the path deviation is within the preset deviation range, the needle tip determines whether the end of the puncture needle has entered the target perfusion cavity based on the changes in the fluid echo in the vicinity of the needle tip and the displacement changes of the surrounding organ boundaries, and outputs the needle tip position confirmation result. The system takes the needle tip position confirmation result, real-time ultrasound image, and peritoneal dialysis machine connection status as inputs, controls the puncture needle to establish a perfusion drainage channel connected to the peritoneal dialysis machine, and determines whether the perfusion drainage channel is connected and effective based on the test fluid echo distribution, local fluid diffusion continuity, and backflow response results after the channel is established, and outputs the channel position confirmation result.

4. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 3, characterized in that, The drug solution processing module is used to: take the preset chemotherapy regimen and the channel location confirmation result as input, analyze the target drug solution type, target dose and target concentration in the preset chemotherapy regimen, and generate the corresponding drug solution preparation parameters when the channel location confirmation result indicates that the perfusion drainage channel is effectively established, and output the preparation task information; The preparation task information is used as input. The intraperitoneal chemotherapy solution is prepared according to the preparation task information. The amount of drug solution added, the amount of diluent added and the time of preparation completion are collected during the preparation process. The actual dose and actual concentration of the intraperitoneal chemotherapy solution are calculated and the drug solution preparation result is output. The drug preparation result and preparation completion time are used as inputs to record the actual dose, actual concentration and preparation time of the intraperitoneal chemotherapy drug solution, generate the intraperitoneal chemotherapy drug solution record result and output it to the perfusion and drainage execution module and the collaborative control module.

5. The ultrasound-guided intraperitoneal perfusion and drainage integrated system according to claim 4, characterized in that, The perfusion and drainage execution module is used to: take the needle tip position confirmation result, channel position confirmation result, intraperitoneal chemotherapy drug recording result, and control command output by the collaborative control module as input, perform consistency verification on the needle tip position confirmation result, channel position confirmation result, and intraperitoneal chemotherapy drug recording result, and determine the corresponding trial injection mode, drug infusion mode, combined drainage mode, or treatment end drainage mode for the peritoneal dialysis machine based on the consistency verification result, and output the mode execution command and initial execution parameters; The peritoneal dialysis machine is controlled to perform trial injection, drug infusion, drainage, or treatment termination drainage through the perfusion and drainage channel, using the mode execution command, initial execution parameters, and control command as inputs. During the execution process, the machine's pause and hold, flow rate, and infusion direction are dynamically adjusted according to the control command, and the real-time execution status and real-time execution parameters are output. The system takes the real-time execution status, real-time execution parameters, and peritoneal dialysis machine operation feedback results as inputs, calculates the infused volume, current flow rate, cumulative drainage volume, and channel response status in the current mode, and generates the execution status and execution parameters corresponding to the current mode, which are then output to the status determination module and the collaborative control module.

6. The integrated system for peritoneal perfusion and drainage based on ultrasound guidance according to claim 5, characterized in that, The state determination module is used to: take real-time ultrasound images, pre-perfusion peritoneal ultrasound baseline image data and execution parameters as input, register and compare the real-time ultrasound images with the pre-perfusion peritoneal ultrasound baseline image data, extract changes in drug diffusion boundary, diffusion direction and diffusion range, and output the drug diffusion state. Using the drug diffusion state, real-time ultrasound images, and the abdominal tumor target area as inputs, the boundary encirclement relationship and regional overlap relationship between the drug diffusion area and the abdominal tumor target area are calculated, and the target area coverage state is output. Using the drug diffusion status, execution status, and patient blood pressure, heart rate, respiratory rate, and abdominal pressure as inputs, the system performs correlation analysis on drug leakage signs, organ compression signs, and vital sign fluctuations, and outputs abnormal states and state change trends.

7. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 6, characterized in that, The state determination module is also used to: take real-time ultrasound images, execution status, execution parameters and drug diffusion status as input, extract the position changes of the drug diffusion front, the migration direction of the diffusion center and the diffusion speed changes at continuous moments, construct the drug diffusion evolution trajectory, and output the diffusion offset result; Using diffusion offset results, target coverage status, and abdominal tumor target boundary as input, calculate the degree of convergence and deviation between the drug diffusion trajectory and the abdominal tumor target boundary, and output the coverage offset status. Using the coverage offset state, abnormal state, and continuous changes in the patient's abdominal pressure and vital signs as input, the risk enhancement trend of the current infusion process is calculated, and the updated state change trend is output.

8. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 7, characterized in that, The state determination module is also used to: take the execution parameters, diffusion offset results and coverage offset status as inputs, construct the drug diffusion response prediction result corresponding to the current execution parameters, compare the drug diffusion response prediction result with the actual diffusion result represented by the real-time ultrasound image, and output the diffusion response deviation result; The diffusion response deviation result, abnormal state, and continuous changes in the patient's abdominal pressure and vital signs are used as inputs to determine whether the current abnormal state matches the current execution parameters, and the abnormality correction result is output. The anomaly correction results, target coverage status, and updated status change trend are used as inputs to correct the target coverage status and anomaly status, and the corrected target coverage status, corrected anomaly status, and corrected status change trend are output.

9. The ultrasound-guided intraperitoneal perfusion and drainage integrated system according to claim 8, characterized in that, The collaborative control module is specifically used to: take the intraperitoneal chemotherapy drug recording results, execution status, execution parameters, target coverage status, abnormal status and status change trend as input, determine the drug infusion stage or drainage stage to which the current treatment process belongs, and generate a set of candidate control actions for the corresponding stage based on the dosage information, concentration information and preparation time information in the intraperitoneal chemotherapy drug recording results, and output the stage determination result and the set of candidate control actions. The system takes the stage judgment result, candidate control action set, target coverage status, abnormal status and status change trend as input, calculates the comprehensive matching result of each candidate control action on target coverage maintenance, abnormal risk suppression and treatment continuity, and outputs the target control action. The target control action, execution status, and execution parameters are used as inputs to generate control commands corresponding to the perfusion and drainage execution module and send them to the perfusion and drainage execution module. The control commands include pause and hold commands, flow rate adjustment commands, and infusion direction adjustment commands for the drug infusion stage, as well as drainage start commands and drainage parameter adjustment commands for the drainage stage.

10. The ultrasound-guided peritoneal perfusion and drainage integrated system according to claim 9, characterized in that, The collaborative control module is also used to: take the target control action, execution state, execution parameters and state change trend as input, construct the stage response result corresponding to the target control action, compare the stage response result with the actual execution state and actual execution parameters fed back by the infusion and drainage execution module, and output the control response deviation result; The control response deviation, target coverage status and abnormal status are taken as inputs. The effect is written back and the priority is updated for each candidate control action in the candidate control action set. The updated candidate control action set is output. The updated set of candidate control actions, the stage determination results, and the intraperitoneal chemotherapy drug recording results are used as inputs to recalculate the suitability of each candidate control action for the current stage and generate iterative control instructions to be sent to the perfusion and drainage execution module, so that the control instructions for the drug infusion stage and the drainage stage are adaptively updated with the execution feedback.

Images