Intelligent medical material transfer method and system based on operating room transfer robot

Abstract

The embodiment of the invention discloses an intelligent medical material transfer method and system based on an operating room transfer robot, and relates to the technical field of robots. The method comprises the following steps: carrying out environment perception processing on a surgical transfer robot to obtain an environment map, and carrying out path generation processing to obtain a transfer path; obstacle avoidance processing and decision processing are carried out on the motion state of the operation transfer robot, and a moving track is obtained; carrying out identity verification processing on the operation transfer robot arriving at the operating room to obtain a verification result, and carrying out docking processing on the operation transfer robot passing the verification to obtain a stopping state; the method comprises the following steps: releasing materials in an operation transfer robot to obtain a material array, and monitoring the release treatment to obtain a confirmation signal; and carrying out recovery processing and verification processing on the pollution materials in the used material array to obtain an inventory result. Through the technical scheme disclosed by the invention, the efficiency and accuracy of medical material transfer of the operation transfer robot are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of robotics technology, and in particular to an intelligent medical material transportation method and system based on an operating room transportation robot. Background Art

[0002] In the modern medical field, accurate and timely delivery of medical supplies is crucial to the smooth progress of surgery. Traditional methods of transporting medical supplies mainly rely on manual operations, including the preparation, transportation, handover and recovery of supplies. Medical staff need to manually select supplies according to surgical needs and transport them from the storage area to the operating room using carts or other simple tools. This method meets basic needs to a certain extent, but with the advancement of medical technology and the increase in the complexity of surgery, the limitations of traditional methods are becoming increasingly prominent.

[0003] However, traditional methods of transporting medical supplies have many shortcomings. First, manual selection and allocation of supplies are prone to omissions or errors, affecting surgical safety; second, manual transport is inefficient, especially in large hospitals, which may lead to delayed surgery; third, manual operation is difficult to ensure the sterility of supplies during transportation; finally, the tracking, inventory and management of supplies lack a systematic approach, increasing the risk of wasting medical resources. These problems not only affect the quality and efficiency of surgery, but may also endanger patient safety. Summary of the invention

[0004] In order to address the deficiencies of the prior art, the embodiments of the present disclosure provide a method and system for intelligent transportation of medical supplies based on an operating room transportation robot, which are used to improve the efficiency and accuracy of existing medical supply transportation.

[0005] In a first aspect, an embodiment of the present disclosure provides an intelligent medical material transportation method based on an operating room transportation robot, comprising: analyzing and processing surgical demand information to obtain a sterile material combination scheme, performing load balancing processing on the sterile material combination scheme to obtain a loading instruction, and performing loading processing on the sterile materials according to the loading instruction to obtain a surgical transportation robot in a loaded state;

[0006] Performing environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and performing path generation processing on the environmental map to obtain a transport path;

[0007] Based on the transport path, performing obstacle avoidance processing on the motion state of the surgical transport robot to obtain an obstacle avoidance strategy, and performing decision processing on the obstacle avoidance strategy to obtain an operation trajectory;

[0008] According to the running trajectory, identity authentication processing is performed on the surgical transport robot that arrives at the operating room to obtain a verification result, and docking processing is performed on the surgical transport robot that passes the verification to obtain a docking state;

[0009] In the docked state, releasing the materials in the surgical transport robot to obtain a material array, and monitoring the releasing process to obtain a confirmation signal;

[0010] Based on the confirmation signal, the contaminated materials in the material array after use are recycled to obtain a contaminated material group, and the contaminated material group is verified to obtain an inventory result.

[0011] In the present disclosure, the steps of analyzing and processing the surgical requirement information to obtain a sterile material combination scheme, performing load balancing processing on the sterile material combination scheme to obtain a loading instruction, and loading the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state include:

[0012] Classify the surgical demand information to obtain surgical categories, and perform statistical analysis on historical surgical data according to the surgical categories to obtain a material usage frequency list;

[0013] Sort the materials by importance according to the material usage frequency list to obtain a core sterile material list, and perform combination optimization on the core sterile material list to obtain an initial medical material plan;

[0014] Performing user preference analysis on the initial medical supplies plan to obtain personalized adjustment suggestions, and modifying the initial medical supplies plan according to the personalized adjustment suggestions to obtain a sterile supplies combination plan;

[0015] Calculating the weight distribution of materials in the sterile material combination scheme to obtain center of gravity data, and adjusting the positions of the materials according to the center of gravity data to obtain a balanced layout scheme;

[0016] Performing temperature sensitivity analysis on the balanced layout plan to obtain temperature control requirements, and optimizing material placement according to the temperature control requirements to obtain a temperature balance plan;

[0017] A three-dimensional visualization model is generated according to the temperature balance solution, a loading indication diagram is extracted from the three-dimensional visualization model, and the contaminated materials are positioned and placed according to the loading indication diagram to obtain a surgical transport robot in a loaded state.

[0018] In the present disclosure, the step of performing environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and performing path generation processing on the environmental map to obtain a transport path includes:

[0019] Performing fusion processing on multi-sensor data of the surgical transport robot in the loaded state to obtain original environmental data, and performing noise filtering on the original environmental data to obtain cleaned environmental data;

[0020] Performing feature extraction on the cleaned environmental data to obtain an environmental feature set, and performing semantic segmentation on the environmental feature set to obtain semantically labeled data;

[0021] Identifying static obstacles in the environment according to the semantic tag data to obtain a static obstacle distribution map, and performing rasterization processing on the static obstacle distribution map to obtain a raster map;

[0022] Constructing a spatial index for the grid map to obtain a fast query structure, and dynamically updating the environment based on the fast query structure to obtain a real-time environment map;

[0023] Analyzing the traversable area of ​​the real-time environment map to obtain a candidate path set, and evaluating the candidate path set to obtain a path score list;

[0024] An optimal path is selected according to the path scoring list to obtain a target transfer path, and the target transfer path is smoothed to obtain a transfer path.

[0025] In the present disclosure, the step of performing obstacle avoidance processing on the motion state of the surgical transport robot based on the transport path to obtain an obstacle avoidance strategy, and performing decision processing on the obstacle avoidance strategy to obtain a running trajectory includes:

[0026] Segmenting the transport path to obtain a path segment sequence, and performing risk assessment on the path segment sequence to obtain a risk level table;

[0027] The speed of the surgical transport robot is adjusted according to the risk level table to obtain a speed control parameter, and the motion state of the surgical transport robot is predicted based on the speed control parameter to obtain a predicted trajectory;

[0028] Scanning the environment around the predicted trajectory in real time to obtain dynamic obstacle data, and performing motion trend analysis on the dynamic obstacle data to obtain an obstacle motion prediction result;

[0029] Calculating the potential collision risk according to the obstacle motion prediction result to obtain a collision risk assessment report, and generating multiple alternative obstacle avoidance schemes based on the collision risk assessment report to obtain an obstacle avoidance strategy set;

[0030] Scoring the safety and efficiency of each strategy in the obstacle avoidance strategy set to obtain a strategy scoring table, and selecting the best obstacle avoidance strategy according to the strategy scoring table to obtain a target obstacle avoidance strategy;

[0031] The motion parameters of the surgical transport robot are adjusted based on the target obstacle avoidance strategy to obtain a corrected motion instruction, and a real-time control signal is generated according to the corrected motion instruction to obtain a running trajectory.

[0032] In the present disclosure, the steps of performing identity authentication processing on the surgical transport robot arriving at the operating room according to the running trajectory to obtain a verification result, and performing docking processing on the surgical transport robot that passes the verification to obtain a docking state include:

[0033] Performing precise positioning processing on the running trajectory to obtain the real-time position coordinates of the surgical transport robot, and triggering an identity authentication request according to the real-time position coordinates to obtain a verification task instruction;

[0034] Scanning the hardware identification of the surgical transfer robot to obtain a unique identification code of the robot, and encrypting the unique identification code of the robot to obtain an encrypted identification;

[0035] Querying the authorization database according to the encrypted identifier to obtain permission information, and performing comparison and analysis on the permission information to obtain a preliminary verification result;

[0036] Performing image acquisition processing on the appearance features of the surgical transfer robot to obtain a feature image, and performing pattern recognition processing on the feature image to obtain a visual verification result;

[0037] Comprehensively evaluating the preliminary verification result and the visual verification result to obtain a verification result, and generating an access token based on the verification result to obtain an identity credential;

[0038] Performing a validity check on the identity credential to obtain an access instruction, and controlling the operating room access control system according to the access instruction to obtain a door opening signal;

[0039] The surgical transport robot is accurately guided based on the door opening signal to obtain a docking path, and the surgical transport robot is controlled to move to a preset position according to the docking path to obtain a docking state.

[0040] In the present disclosure, in the docked state, releasing the materials in the surgical transport robot to obtain a material array, and monitoring the release process to obtain a confirmation signal step include:

[0041] Performing posture calibration processing on the surgical transport robot in the docked state to obtain a precise docking position, and unlocking the robot storage compartment according to the precise docking position to obtain an open state;

[0042] Scan and identify the materials in the storage compartment in the open state to obtain a list of sterile materials, and generate a release sequence according to the list of sterile materials to obtain a material release plan;

[0043] According to the material release plan, materials in the storage compartment are taken out one by one to obtain a release sequence, and each material in the release sequence is positioned to obtain material coordinates;

[0044] Arranging the released materials based on the material coordinates to obtain a material array, and performing integrity check on the material array to obtain a material status report;

[0045] Performing real-time video acquisition and processing on the entire release process to obtain a monitoring video stream, and performing intelligent analysis and processing on the monitoring video stream to obtain an abnormal behavior record;

[0046] The material status report and the abnormal behavior record are comprehensively evaluated and processed to obtain a release quality score, and operation feedback is generated according to the release quality score to obtain a confirmation signal.

[0047] In the present disclosure, the step of recycling the contaminated materials in the material array after use based on the confirmation signal to obtain a contaminated material group, and verifying the contaminated material group to obtain an inventory result includes:

[0048] Performing an environmental scan of the operating room according to the confirmation signal to obtain a distribution map of materials after use, and classifying and marking the distribution map of materials after use to obtain a material recovery list;

[0049] Prioritizing the material recovery list to obtain a recovery sequence table, and planning a path for the surgical transport robot according to the recovery sequence table to obtain a recovery path;

[0050] Collect and process the used materials one by one along the recycling path to obtain a polluted resource group, and perform disinfection status detection on the polluted resource group to obtain a disinfection requirement list;

[0051] According to the disinfection requirement list, the polluted resource groups are partitioned and stored to obtain classified polluted resource groups, and the quantity of the classified polluted resource groups is checked to obtain preliminary inventory results;

[0052] Comparing the preliminary inventory results with the original sterile material list to obtain a difference report, and performing a second scan of the operating room based on the difference report to obtain missing material information;

[0053] The missing material information is integrated with the classified pollutant resource group to obtain a complete pollutant resource group, and the complete pollutant resource group is finally verified to obtain an inventory result.

[0054] In a second aspect, the disclosed embodiment further provides an intelligent medical material transport system based on an operating room transport robot, comprising:

[0055] A processing module, used to analyze and process the surgical demand information to obtain a sterile material combination plan, perform load balancing processing on the sterile material combination plan to obtain a loading instruction, and perform loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state;

[0056] A perception module, used to perform environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and perform path generation processing on the environmental map to obtain a transport path;

[0057] An obstacle avoidance module, used to perform obstacle avoidance processing on the motion state of the surgical transport robot based on the transport path to obtain an obstacle avoidance strategy, and perform decision processing on the obstacle avoidance strategy to obtain a running trajectory;

[0058] A verification module, used to perform identity authentication processing on the surgical transport robot arriving at the operating room according to the running trajectory to obtain a verification result, and to perform docking processing on the surgical transport robot that has passed the verification to obtain a docking state;

[0059] A monitoring module, used for releasing the materials in the surgical transport robot in the docked state to obtain a material array, and monitoring the release process to obtain a confirmation signal;

[0060] The recycling module is used to recycle the polluted materials in the material array after use based on the confirmation signal to obtain a polluted material group, and verify the polluted material group to obtain an inventory result.

[0061] In the technical solution provided by the present disclosure, by analyzing and processing the surgical demand information and generating the sterile material combination scheme, the medical material configuration is realized to be precise and personalized, ensuring that the materials required for each operation can be optimally configured, thereby improving the efficiency and accuracy of surgical preparation. The introduction of load balancing processing not only optimizes the distribution of materials in the robot, but also improves the stability and safety during transportation, and reduces the risk of material damage or contamination. The environmental perception processing and path generation functions enable the robot to adapt to the complex and changeable hospital environment in real time, autonomously plan the optimal transfer path, greatly improve the efficiency of material transportation, and reduce the consumption of human resources. The obstacle avoidance processing and decision processing functions in the motion state further enhance the robot's adaptability in complex environments, ensure the safety and reliability of the material transportation process, and effectively reduce the incidence of collision accidents. The identity authentication processing and docking processing after arriving at the operating room not only improve the safety of material handover, but also realize accurate positioning and docking, and prepare for the subsequent release of materials. The materials are released in the docked state, and a confirmation signal is generated through monitoring processing. This design ensures the accuracy and traceability of the material handover process, and effectively reduces the risk of human error and material loss. Finally, the recycling and verification of contaminated materials after use realizes the full life cycle management of medical materials, which not only improves the efficiency and accuracy of material recycling, but also provides reliable data support for subsequent disinfection, reuse and replenishment. Through precise material management and tracking, this method helps to optimize the hospital's inventory management, reduce waste and reduce operating costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0063] Figure 1 This is a flow chart of a method for intelligently transporting medical supplies based on an operating room transport robot in an embodiment of the present disclosure;

[0064] Figure 2 It is a schematic diagram of an intelligent medical material transport system based on an operating room transport robot in an embodiment of the present disclosure. DETAILED DESCRIPTION

[0065] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0066] In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", "right", "vertical", "horizontal", "inner", "outer", etc., indicating the orientation or positional relationship, are based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and "third" are used for descriptive purposes only, and cannot be understood as indicating or implying relative importance.

[0067] In addition, the technical features involved in the different embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other. For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , Figure 1 is a flow chart of a method for intelligently transporting medical supplies based on an operating room transport robot according to an embodiment of the present disclosure. Figure 1 As shown, the following steps are included:

[0068] S101, analyzing and processing the surgical demand information to obtain a sterile material combination plan, performing load balancing processing on the sterile material combination plan to obtain a loading instruction, and performing loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state;

[0069] S102, performing environmental perception processing on the surgical transport robot in a loaded state to obtain an environmental map, and performing path generation processing on the environmental map to obtain a transport path;

[0070] S103, based on the transport path, performing obstacle avoidance processing on the motion state of the surgical transport robot to obtain an obstacle avoidance strategy, and performing decision processing on the obstacle avoidance strategy to obtain an operation trajectory;

[0071] S104, performing identity authentication processing on the surgical transport robot that has arrived at the operating room according to the running trajectory to obtain a verification result, and performing docking processing on the surgical transport robot that has passed the verification to obtain a docking status;

[0072] S105, in the docked state, releasing the materials in the surgical transport robot to obtain a material array, and monitoring the release process to obtain a confirmation signal;

[0073] S106. Based on the confirmation signal, the polluted materials in the used material array are recycled to obtain a polluted material group, and the polluted material group is verified to obtain an inventory result.

[0074] It should be noted that the surgical demand information is analyzed and processed, and the surgical category is obtained through classification processing, and the frequency list of instrument use is obtained by combining historical surgical data statistical analysis. Subsequently, a core sterile material list is generated according to the importance of the instruments, and then the initial sterile material plan is obtained through combination optimization. After considering user preferences, the plan is personalized and finally a sterile material combination plan is formed. Then the plan is load balanced, the weight distribution is calculated to obtain the center of gravity data, and the position of the instrument is adjusted to form a balanced layout plan. After considering temperature sensitivity, the placement is optimized, a three-dimensional visualization model is generated, and a loading indication map is extracted to guide the positioning and placement of sterile materials, and finally a surgical transport robot in a loaded state is obtained. In the environmental perception processing stage, multi-sensor data is integrated, and static obstacles are identified and a raster map is constructed after noise filtering, feature extraction and semantic segmentation. A real-time environmental map is formed through spatial index construction and dynamic update. Based on this map, the traversable area is analyzed, candidate paths are evaluated and the optimal path is selected, and the transfer path is obtained after smoothing. During the movement, the transfer path is segmented and the risk is evaluated, the speed parameters are adjusted, and the movement state is predicted. By scanning the environment in real time and analyzing the movement trends of dynamic obstacles, the potential collision risks are calculated and multiple alternative obstacle avoidance solutions are generated. After evaluating the safety and efficiency of each strategy, the best obstacle avoidance strategy is selected, and the motion parameters are adjusted accordingly and real-time control signals are generated to form the running trajectory. After arriving at the operating room, the robot is accurately located and identity authentication is triggered. The hardware logo is scanned to obtain a unique identification code, which is encrypted and queried in the authorization database. At the same time, image acquisition and pattern recognition of appearance features are performed, and the verification results are obtained through comprehensive evaluation. After verification, an access token is generated to control the opening of the access control system and guide the robot to accurately dock to the preset position.

[0075] In the docked state, the robot is calibrated and the storage compartment is unlocked. The equipment in the compartment is scanned and identified to generate a material release plan. The materials are taken out one by one according to the plan, and the materials are positioned and arranged to form a material array. Video monitoring and intelligent analysis are carried out throughout the process to evaluate the release quality and generate operational feedback. Finally, based on the confirmation signal, the used materials are scanned in the environment, marked and classified to obtain a recycling list. Recycling is carried out in order of priority, the disinfection status is checked and stored in different areas. The quantity is checked and compared with the original list, and a second scan is performed to deal with the missing materials. After integrating all the information, a final verification is carried out to obtain the inventory results.

[0076] For example, a complex spinal surgery is required in the orthopedic operating room of a hospital. The surgical transport robot first analyzes the surgical demand information and determines the category of spinal surgery. By analyzing the data of 200 similar surgeries in the past 6 months, a frequency list of instrument use is generated, such as the frequency of use of spinal fixators is 98%, and the frequency of use of microsurgical instruments is 85%. Based on this sorting, a list of core sterile supplies is formed, including spinal fixators, microsurgical instruments, nerve monitors, etc. Considering the preferences of the surgeon, a specific model of intervertebral fusion cage is added. The final sterile material combination plan contains 45 instruments with a total weight of 28.5 kg. During the load balancing process, it is calculated that the center of gravity is 2.3 cm to the right. By adjusting the position of heavy instruments, the center of gravity is returned to the center. Temperature sensitivity analysis shows that some drugs need to be kept at 4-8°C, so they are placed in the built-in refrigerated area. The generated 3D visualization model guides the staff to accurately place each material and finally complete the loading. During the transportation process, the robot's laser radar, ultrasonic sensor and other equipment fuse data to generate a real-time environmental map inside the hospital. The path planning algorithm calculated the optimal path from the central supply room to the operating room, a total of 280 meters, which is expected to take 5 minutes. Three dynamic obstacles were detected along the way, including two stretchers and a fast-moving medical staff. The robot adjusted its speed and direction in real time and successfully avoided all obstacles. After arriving at the operating room, the robot performed dual identity authentication through RFID tags and cameras, with a verification pass rate of 99.9%. It docked accurately to the preset position with a deviation of less than 1 cm.

[0077] During the material release phase, the robot released 45 instruments in a preset order, and the entire process was recorded with a 4K high-definition camera. The intelligent analysis system detects the integrity and cleanliness of each material in real time to confirm that all materials are in good condition. After the operation, the robot scanned the operating room again and identified 43 used instruments and 2 unused instruments. During the recycling process, it was found that one small clamp was not in the expected location by comparing the original list, and it was found in the trash can through a second scan. Finally, it was verified that all 46 materials (including disposable items added during the operation) were recovered, and the inventory accuracy rate reached 100%.

[0078] By executing the above steps, through the analysis and processing of surgical demand information and the generation of sterile material combination plans, the precision and personalization of medical material configuration are achieved, ensuring that the materials required for each operation can be optimally configured, thereby improving the efficiency and accuracy of surgical preparation. The introduction of load balancing processing not only optimizes the distribution of materials in the robot, but also improves the stability and safety during transportation, and reduces the risk of material damage or contamination. The environmental perception processing and path generation functions enable the robot to adapt to the complex and changing hospital environment in real time, and autonomously plan the optimal transfer path, which greatly improves the efficiency of material transportation and reduces the consumption of human resources. The obstacle avoidance processing and decision processing functions in the motion state further enhance the robot's adaptability in complex environments, ensure the safety and reliability of the material transportation process, and effectively reduce the incidence of collision accidents. The identity authentication processing and docking processing after arriving at the operating room not only improve the safety of material handover, but also realize accurate positioning and docking, preparing for the subsequent release of materials. The materials are released in the docked state, and the confirmation signal is generated through monitoring processing. This design ensures the accuracy and traceability of the material handover process, and effectively reduces the risk of human errors and material loss. Finally, the recycling and verification of contaminated materials after use realizes the full life cycle management of medical materials, which not only improves the efficiency and accuracy of material recycling, but also provides reliable data support for subsequent disinfection, reuse and replenishment. Through precise material management and tracking, this method helps to optimize the hospital's inventory management, reduce waste and reduce operating costs.

[0079] In a specific embodiment, the process of executing step S101 may specifically include the following steps:

[0080] (1) Classify and process the surgical demand information to obtain surgical categories, and perform statistical analysis on historical surgical data based on surgical categories to obtain a frequency list of material usage;

[0081] (2) Sort the materials by importance according to the material usage frequency list to obtain a core sterile material list, and then optimize the core sterile material list to obtain an initial medical material plan;

[0082] (3) Conducting user preference analysis on the initial medical supplies plan to obtain personalized adjustment suggestions, and modifying the initial medical supplies plan based on the personalized adjustment suggestions to obtain a sterile supplies combination plan;

[0083] (4) Calculating the weight distribution of materials in the sterile material combination plan to obtain the center of gravity data, and adjusting the position of the materials according to the center of gravity data to obtain a balanced layout plan;

[0084] (5) Conduct temperature sensitivity analysis on the balanced layout plan to obtain temperature control requirements, and optimize material placement according to the temperature control requirements to obtain a temperature balance plan;

[0085] (6) A three-dimensional visualization model is generated according to the temperature balance solution, a loading indication diagram is extracted from the three-dimensional visualization model, and the contaminated materials are positioned and placed according to the loading indication diagram to obtain a surgical transfer robot in a loaded state.

[0086] Specifically, the intelligent transportation method of medical supplies based on the operating room transfer robot first classifies the surgical demand information to obtain the surgical category. The surgical description is converted into a standardized surgical category. For example, "coronary artery bypass surgery" will be classified as "cardiac surgery". Subsequently, the historical surgical data is statistically analyzed according to the surgical category to obtain a list of material usage frequencies. Data mining technology is used to calculate the frequency of use of each material in a specific surgical category. The frequency calculation formula is as follows:

[0087]

[0088] Among them, F i represents the usage frequency of material i, N i Indicates the number of times material i is used in this type of surgery, N t Represents the total number of surgeries of this type.

[0089] Next, the materials are ranked by importance according to the material usage frequency list to obtain the core sterile material list. The ranking adopts a weighted scoring method, taking into account the material usage frequency and importance. The scoring formula is as follows:

[0090] S i =w1F i +w2I i ;

[0091] Among them, S i represents the total score of material i, w1 and w2 are the weights of frequency of use and importance, respectively, and F i is the frequency of use, I i The importance is rated by experts, ranging from 1 to 10.

[0092] The core sterile material list was combined and optimized to obtain the initial medical material plan. A genetic algorithm was used to analyze user preferences for the initial medical material plan with the objective function of minimizing the number of materials and maximizing surgical coverage, and personalized adjustment suggestions were obtained. A collaborative filtering algorithm was used to predict the doctor's preference for the current package based on his past equipment selection history. The initial medical material plan was modified according to the personalized adjustment suggestions to obtain a sterile material combination plan.

[0093] Calculate the weight distribution of materials in the sterile material combination plan to obtain the center of gravity data. The calculation adopts the center of gravity calculation method of discrete particle system:

[0094]

[0095] in, is the coordinate of the center of gravity of the system, m i is the mass of the i-th material, (x i ,y i ,z i ) is the coordinate of the ith material. The material position is adjusted according to the center of gravity data to obtain a balanced layout plan. The adjustment process uses a simulated annealing algorithm to minimize the deviation between the system center of gravity and the ideal center of gravity position by continuously adjusting the material position. A temperature sensitivity analysis is performed on the balanced layout plan to obtain the temperature control requirements. Considering the suitable temperature range of each material and the ambient temperature, the temperature distribution of each position is calculated using a heat conduction model. The material placement is optimized according to the temperature control requirements to obtain a temperature balance plan. The optimization process uses a multi-objective optimization algorithm, taking into account both the center of gravity balance and temperature control objectives.

[0096] Finally, a three-dimensional visualization model is generated according to the temperature balance scheme, and a loading instruction diagram is extracted from the three-dimensional visualization model. Using computer graphics technology, the optimized material placement plan is transformed into an intuitive 3D model. The contaminated materials are positioned and placed according to the loading instruction diagram to obtain a surgical transport robot in a loaded state.

[0097] For example, in a complex heart surgery, the surgical requirement information is classified as "open heart surgery". The system analyzed the data of the past 1,000 similar surgeries and calculated that the frequency of use of the heart-lung machine was 98% and the frequency of use of the special heart clamp was 85%. Considering the importance of the heart-lung machine (expert score 10 points), its total score was 0.6×98%+0.4×10=9.88, ranking first in the core sterile material list. Through combination optimization, the system generated an initial package plan containing 25 instruments. Based on the surgeon's past preferences, the system recommends adding a special stapler to form the final sterile material combination plan. In the weight distribution calculation, it is assumed that the heart-lung machine weighs 50kg and the coordinates are (0, 0, 0), and the other 24 instruments weigh a total of 30kg and are distributed in the space from (1, 1, 1) to (5, 5, 5). The calculation shows that the center of gravity of the system is approximately (0.75, 0.75, 0.75). In order to make the center of gravity closer to the ideal position (2.5, 2.5, 0), the position of the instruments was adjusted. Temperature sensitivity analysis showed that the heart-lung machine needs to be maintained at 18-22°C, while some drug-coated instruments require a low-temperature environment of 4-8°C. Based on this, a zoned temperature control scheme was designed to place low-temperature instruments in the refrigeration area. The final 3D loading instruction diagram clearly shows the precise position and placement angle of each instrument, guiding staff to load accurately and ensuring that the surgical transport robot achieves the best balance and temperature control state.

[0098] In a specific embodiment, the process of executing step S102 may specifically include the following steps:

[0099] (1) fusing the multi-sensor data of the surgical transport robot in a loaded state to obtain raw environmental data, and filtering the raw environmental data for noise to obtain cleaned environmental data;

[0100] (2) Extract features from the cleaned environmental data to obtain an environmental feature set, and perform semantic segmentation on the environmental feature set to obtain semantically labeled data;

[0101] (3) Identify static obstacles in the environment based on the semantically labeled data to obtain a static obstacle distribution map, and rasterize the static obstacle distribution map to obtain a raster map;

[0102] (4) constructing a spatial index for the raster map to obtain a fast query structure, and dynamically updating the environment based on the fast query structure to obtain a real-time environment map;

[0103] (5) Analyze the traversable area of ​​the real-time environment map to obtain a set of candidate paths, and evaluate the set of candidate paths to obtain a path scoring list;

[0104] (6) Select the optimal path according to the path score list to obtain the target transfer path, and smooth the target transfer path to obtain the transfer path.

[0105] Specifically, the multi-sensor data of the loaded surgical transport robot is fused to obtain the original environmental data. It involves the data integration of multiple sensors such as lidar, ultrasonic sensors and cameras. The data fusion uses the Kalman filter algorithm to merge the data of different sensors into a unified environmental representation. Subsequently, the original environmental data is noise filtered to obtain the cleaned environmental data. The noise filtering uses a combination of median filtering and Gaussian filtering to effectively remove outliers and random noise. Next, the cleaned environmental data is feature extracted to obtain an environmental feature set. The feature extraction process uses the SIFT (Scale Invariant Feature Transform) algorithm to identify key points and descriptors in the environment. These features include stable environmental signs such as wall edges, door frame outlines, and fixed equipment. Then, the environmental feature set is semantically segmented to obtain semantically labeled data. Semantic segmentation uses the U-Net network in deep learning to classify the various elements in the environment into different categories such as corridors, rooms, doors, and obstacles.

[0106] According to the semantically labeled data, static obstacles in the environment are identified to obtain a static obstacle distribution map. Static obstacles include immovable objects such as walls, fixed equipment, and furniture. The identification process uses a clustering algorithm to classify points with similar features and positions as the same obstacle. Subsequently, the static obstacle distribution map is rasterized to obtain a raster map. Rasterization divides the continuous space into discrete grids, and each grid is marked as occupied, free, or unknown. A spatial index is constructed for the raster map to obtain a fast query structure. The spatial index uses an octree data structure to recursively divide the three-dimensional space into smaller cubes to achieve efficient retrieval of spatial data. The environment is dynamically updated based on the fast query structure to obtain a real-time environment map. During the dynamic update process, new sensor data is continuously compared and fused with the existing map to ensure the real-time and accuracy of the map.

[0107] The real-time environment map is analyzed for traversable areas to obtain a set of candidate paths. The traversable area analysis uses an expansion algorithm to appropriately reduce the original free space considering the actual size of the robot to ensure the safety of the path. The candidate path set is generated using the Rapidly Expanding Random Tree (RRT) algorithm to quickly generate multiple feasible paths in the traversable area. Subsequently, the candidate path set is evaluated to obtain a path score list. The evaluation criteria include multiple factors such as path length, smoothness, and safety margin. Each factor is assigned a different weight to calculate a comprehensive score. Finally, the optimal path is selected based on the path score list to obtain the target transfer path. The selection process uses a multi-objective decision-making method to seek the best balance between efficiency and safety. The target transfer path is smoothed to obtain the final transfer path. The smoothing process uses Bezier curve interpolation to eliminate mutation points in the path and make the robot move more smoothly.

[0108] For example, in a contaminated material transport mission in a large general hospital, the surgical transport robot departs from the central supply room and is destined for the cardiac surgery operating room. The robot first fuses the lidar and camera data to generate raw point cloud data containing about 1 million spatial points. Through noise filtering, about 5% of the abnormal points are removed to obtain a clearer environmental representation. The feature extraction process identifies 2,000 key points, which are mainly distributed in corridor corners, door frames and fixed equipment. Semantic segmentation divides the environment into corridors (40%), rooms (30%), doors (5%) and other areas (25%). Static obstacle recognition finds 15 major obstacles, including 6 fixed counters, 4 seating areas and 5 large medical equipment. Rasterization processing divides the environment into 500x500 grids, each representing a 10cmx 10cm area in the actual space. Spatial index construction organizes these grids into a 5-layer octree structure, which greatly improves the efficiency of spatial queries. During the dynamic update process, the robot processes about 100,000 new sensor data points per second and updates the environmental map in real time.

[0109] The traversable area analysis takes into account the width (30cm) and length (120cm) of the robot, reducing the original free space by a safe distance of 40cm. The RRT algorithm generated 20 candidate paths in 5 seconds, with path lengths ranging from 50 meters to 70 meters. The path evaluation takes into account three factors: length (weight 0.4), number of turns (weight 0.3), and obstacle distance (weight 0.3). The optimal path finally selected was 58 meters long, including 4 turns, with an average obstacle distance of 1.5 meters. After Bezier curve smoothing, the maximum curvature of the path was reduced from the original 30 degrees / meter to 15 degrees / meter, ensuring the stability and safety of the robot during movement.

[0110] In a specific embodiment, the process of executing step S103 may specifically include the following steps:

[0111] (1) Segment the transport route to obtain a route segment sequence, and perform risk assessment on the route segment sequence to obtain a risk level table;

[0112] (2) adjusting the speed of the surgical transport robot according to the risk level table to obtain a speed control parameter, and predicting the motion state of the surgical transport robot based on the speed control parameter to obtain a predicted trajectory;

[0113] (3) Scan the environment around the predicted trajectory in real time to obtain dynamic obstacle data, and perform motion trend analysis on the dynamic obstacle data to obtain obstacle motion prediction results;

[0114] (4) Calculate the potential collision risk according to the obstacle motion prediction result to obtain a collision risk assessment report, and generate multiple alternative obstacle avoidance schemes based on the collision risk assessment report to obtain an obstacle avoidance strategy set;

[0115] (5) Score the safety and efficiency of each strategy in the obstacle avoidance strategy set to obtain a strategy score table, and select the best obstacle avoidance strategy based on the strategy score table to obtain the target obstacle avoidance strategy;

[0116] (6) Based on the target obstacle avoidance strategy, the motion parameters of the surgical transfer robot are adjusted to obtain a corrected motion instruction, and a real-time control signal is generated according to the corrected motion instruction to obtain the operation trajectory.

[0117] Specifically, the transfer route is divided into several path segments according to environmental features (such as corners, doorways, narrow passages, etc.) to obtain a path segment sequence. Risk assessment is performed on each path segment, taking into account factors such as segment length, width, corner angle, and crowd density, and the risk level is calculated using a weighted scoring method. The risk scoring formula is as follows:

[0118] R i =β1L i +β2W i +β3A i +β4D i ;

[0119] Among them, R i is the risk score of the ith path segment, L i , W i , A i , D iare the standardized scores of length, width, angle and crowd density, respectively, and β1, β2, β3, and β4 are the corresponding weights. The speed of the surgical transport robot is adjusted according to the risk level table to obtain the speed control parameters. The speed adjustment uses a fuzzy control algorithm to map the risk level to the speed range. The speed control parameters include linear speed and angular speed, and the calculation formula is:

[0120] V=V max (1-k v R i );ω=ω max (1-k ω R i );

[0121] Where v and ω are linear velocity and angular velocity respectively, v max and ω max is the maximum linear velocity and angular velocity, k v and k ω is the speed adjustment factor, R i The risk score for the current path segment.

[0122] The motion state of the surgical transport robot is predicted based on the speed control parameters to obtain the predicted trajectory. The prediction uses a kinematic model to consider the current position, speed, and acceleration of the robot and calculate the position sequence in the future. The environment around the predicted trajectory is scanned in real time to obtain dynamic obstacle data. The scanning uses a laser radar and a depth camera to obtain the position, speed, and size information of the obstacle. The dynamic obstacle data is analyzed for motion trends, and the Kalman filter algorithm is used to predict the future position and speed of the obstacle to obtain the obstacle motion prediction result.

[0123] The potential collision risk is calculated based on the obstacle motion prediction results to obtain a collision risk assessment report. The collision risk calculation takes into account the predicted trajectories of the robot and the obstacle, and uses the space-time volume intersection method to evaluate the collision probability. Based on the collision risk assessment report, multiple alternative obstacle avoidance solutions are generated to form an obstacle avoidance strategy set. The generation of obstacle avoidance solutions uses the dynamic window method to search for feasible motion commands in the speed-time space.

[0124] Each strategy in the obstacle avoidance strategy set is scored for safety and efficiency to obtain a strategy scoring table. The scoring criteria include the minimum distance to the obstacle, the time to reach the target point, energy consumption, etc. The scoring formula is:

[0125]

[0126] Among them, S is the strategy score, d min is the minimum obstacle distance, t is the arrival time, E is the energy consumption, w s 、w t 、we is the weight coefficient, α, t max 、E max is the normalization parameter.

[0127] The best obstacle avoidance strategy is selected according to the strategy scoring table to obtain the target obstacle avoidance strategy. The selection process uses a multi-objective decision-making method to seek the best balance between safety and efficiency. The motion parameters of the surgical transport robot are adjusted based on the target obstacle avoidance strategy to obtain the corrected motion instructions. The adjustment process considers the dynamic constraints of the robot to ensure the executability of the instructions. Finally, a real-time control signal is generated according to the corrected motion instructions to obtain the running trajectory.

[0128] For example, in a transport mission from a material warehouse to a cardiac surgery operating room, the total length of the transport path is 100 meters and is divided into 10 path segments. Among them, the third segment passes through a narrow corridor with a width of only 1.5 meters and a risk score of 0.8 (out of 1). According to the risk level, the maximum speed of the robot in this path is limited to 0.5m / s (normal maximum speed is 1m / s). The predicted trajectory shows that the robot will pass through this corridor within 20 seconds. Real-time scanning detects a moving medical cart, with an initial position of 5 meters away from the robot and a speed of 0.3m / s. Kalman filtering predicts that the cart will intersect with the robot path in 10 seconds. The collision risk calculation shows that if no avoidance measures are taken, the collision probability is 85%. The system generates three obstacle avoidance strategies: deceleration to 0.2m / s, waiting for the cart to pass in place, and fine-tuning the path to bypass. After the score calculation, the deceleration strategy scored the highest (0.85 points) and was selected as the target obstacle avoidance strategy. Finally, the robot's motion parameters were adjusted to a linear speed of 0.2 m / s, maintaining the original path. This decision minimized delays while ensuring safety. The entire obstacle avoidance process lasted 30 seconds, 10 seconds longer than originally planned, but successfully avoided potential collision risks and ensured the safe transfer of materials.

[0129] In a specific embodiment, the process of executing step S104 may specifically include the following steps:

[0130] (1) Accurately positioning the running trajectory to obtain the real-time position coordinates of the surgical transfer robot, and triggering an identity authentication request based on the real-time position coordinates to obtain a verification task instruction;

[0131] (2) Scanning the hardware identification of the surgical transfer robot to obtain a unique identification code of the robot, and encrypting the unique identification code of the robot to obtain an encrypted identification;

[0132] (3) Query the authorization database according to the encrypted identifier to obtain the permission information, and compare and analyze the permission information to obtain preliminary verification results;

[0133] (4) performing image acquisition and processing on the appearance features of the surgical transfer robot to obtain a feature image, and performing pattern recognition processing on the feature image to obtain a visual verification result;

[0134] (5) Comprehensively evaluate the preliminary verification result and the visual verification result to obtain a verification result, and generate an access token based on the verification result to obtain an identity credential;

[0135] (6) Check the validity of the identity credentials, obtain the access instruction, and control the operating room access control system according to the access instruction to obtain the door opening signal;

[0136] (7) Based on the door opening signal, the surgical transfer robot is precisely guided to obtain a docking path, and the surgical transfer robot is controlled to move to a preset position according to the docking path to obtain a docking state.

[0137] Specifically, the running trajectory is precisely positioned to obtain the real-time position coordinates of the surgical transport robot. Indoor positioning technologies, such as ultra-wideband (UWB) positioning or visual-inertial odometer (VIO), are used to achieve centimeter-level precision positioning. When the robot approaches the preset trigger point in the operating room, an identity authentication request is automatically triggered according to the real-time position coordinates, and a verification task instruction is generated. Subsequently, the hardware identification of the surgical transport robot is scanned to obtain the robot's unique identification code. This is accomplished by reading the robot's built-in RFID chip or scanning the QR code on the body. In order to improve security, the robot's unique identification code is encrypted to obtain an encrypted identification. The encryption process uses an asymmetric encryption algorithm, such as RSA, to ensure that the identification is not tampered with or stolen during transmission.

[0138] The authorization database is queried according to the encrypted identifier to obtain the permission information. The authorization database stores the identity information of all legal robots and the corresponding permission levels. The query process uses an efficient hash index to quickly retrieve the corresponding permission information. The permission information is compared and analyzed to obtain preliminary verification results. The comparison analysis includes checking multiple dimensions such as permission level and time validity. To further ensure safety, the appearance features of the surgical transport robot are imaged and processed to obtain feature images. Image acquisition uses a high-resolution camera to capture the robot's all-round appearance. The feature image is processed for pattern recognition to obtain visual verification results. Pattern recognition uses deep learning algorithms, such as convolutional neural networks (CNN), to identify the robot's unique appearance features, such as logos, colors, shapes, etc.

[0139] The preliminary verification results and the visual verification results are comprehensively evaluated to obtain the final verification results. The comprehensive evaluation adopts a weighted scoring method, taking into account the reliability of hardware recognition and the accuracy of visual recognition. An access token is generated based on the verification results to obtain the identity credential. The access token uses the JWT (JSON Web Token) format and contains information such as the robot ID, permission level, and validity period. The identity credential is checked for validity to obtain the access instruction. The validity check includes token signature verification, expiration time check, etc. The operating room access control system is controlled according to the access instruction to obtain the door opening signal. The access control system and the robot communicate using an encrypted wireless protocol, such as WPA3, to ensure the security of signal transmission.

[0140] Finally, the surgical transport robot is precisely guided based on the door opening signal to obtain the docking path. The generation of the docking path takes into account the spatial layout of the operating room and the location of other equipment, and the A* algorithm is used to calculate the optimal path. According to the docking path, the surgical transport robot is controlled to move to the preset position to obtain the docking state. The docking process uses visual servo control to achieve millimeter-level precise docking.

[0141] For example, in a cardiac surgery instrument transport mission, the surgical transport robot approached the operating room with an important artificial heart-lung machine. When the robot reached a position 5 meters away from the door of the operating room, the UWB positioning system accurately located its coordinates as (X: 15.23m, Y: 8.76m, Z: 0m). This coordinate triggered the identity authentication process and generated a verification task instruction. The robot's RFID chip was scanned and a unique identification code "STR-2023-0001" was obtained. This identification code was encrypted with RSA to generate a 128-bit encrypted identifier.

[0142] The authorization database quickly retrieved the robot's permission information, showing that it had "Class A" access rights, allowing access to all surgical areas. After the initial verification, the high-definition camera captured a 360-degree image of the robot with a resolution of 4K. The deep learning model completed the image analysis within 0.5 seconds, identifying the robot's unique blue shell and the "Jiusin" logo, and the visual verification result was "matched".

[0143] The comprehensive evaluation gave a high confidence level of 98 points (out of 100), and generated a JWT token containing the robot ID, Class A permissions, and a validity period of 30 minutes. After the token passes the validity check, it sends a door opening command to the operating room access control system. After the access control system verifies the command, the door opening operation is completed within 0.8 seconds. After the robot obtains entry permission, it generates a 7-meter-long precise docking path based on the internal layout of the operating room, avoiding the operating table and other medical equipment. The robot moves slowly along this path and eventually stops precisely at a preset position 1.5 meters away from the operating table and close to the power socket, with an error of less than 5 mm. The entire identity authentication and docking process takes about 45 seconds, saving 3 minutes compared to manual operation, while greatly improving safety and accuracy.

[0144] In a specific embodiment, the process of executing step S105 may specifically include the following steps:

[0145] (1) Performing posture calibration on the surgical transport robot in the docked state to obtain a precise docking position, and unlocking the robot storage compartment according to the precise docking position to obtain an open state;

[0146] (2) Scan and identify the materials in the open storage compartment to obtain a list of sterile materials, and generate a release sequence based on the list of sterile materials to obtain a material release plan;

[0147] (3) Taking out materials from the storage compartment one by one according to the material release plan to obtain a release sequence, and positioning each material in the release sequence to obtain material coordinates;

[0148] (4) Arranging the released materials based on the material coordinates to obtain a material array, and performing integrity check on the material array to obtain a material status report;

[0149] (5) Real-time video acquisition and processing of the entire release process to obtain a monitoring video stream, and intelligent analysis and processing of the monitoring video stream to obtain abnormal behavior records;

[0150] (6) Comprehensively evaluate and process the material status report and abnormal behavior record to obtain a release quality score, and generate operation feedback based on the release quality score to obtain a confirmation signal.

[0151] Specifically, the intelligent transportation method for medical supplies based on the operating room transfer robot first performs posture calibration on the surgical transfer robot after entering the docking state to ensure accurate docking. The high-precision gyroscope and accelerometer on the robot are used in combination with the visual positioning system to fine-tune the position and posture of the robot. The posture calibration adopts the Kalman filter algorithm and fuses multi-sensor data to achieve sub-millimeter accuracy. After obtaining the precise docking position, the robot storage compartment is unlocked according to the preset security protocol. The unlocking process adopts a multiple authentication mechanism, including position verification, timestamp check and digital signature verification, to ensure that the storage compartment can only be opened at the correct position and time. After the storage compartment enters the open state, the internal instruments are immediately scanned and identified. A method combining a high-resolution 3D scanner and an RFID reader is used to quickly and accurately identify each instrument. The scanned data is analyzed by an image processing algorithm and a deep learning model to generate a detailed list of sterile materials, including the name, model, quantity and location information of each instrument. According to the list of sterile materials and the preset surgical process, a priority sorting algorithm is used to generate the optimal instrument release sequence to form a material release plan. According to the material release plan, the robot begins to take out the materials in the storage compartment one by one. The removal process is performed by a precision robotic arm, and each action is accurately calculated to ensure the safety and sterility of the instruments. For each released instrument, a machine vision system is used for real-time tracking, and triangulation is used to calculate its spatial coordinates. These coordinate information is recorded to form the release sequence and position data of the instrument.

[0152] Based on the obtained material coordinates, the robot arranges the released materials to form a material array. The arrangement process takes into account the layout of the operating room, the order of instrument use and the doctor's preferences, and uses a space optimization algorithm to calculate the best placement. After the arrangement is completed, the material array is checked for integrity, which includes quantity verification, location verification and appearance inspection. The inspection results are compiled into a detailed material status report, recording the status of each instrument and any potential problems. The entire release process is recorded in real time by multiple high-definition cameras to form a monitoring video stream. The video data is encoded and compressed in real time to ensure high quality without taking up too much storage space. The monitoring video stream is intelligently analyzed and processed, and computer vision and machine learning algorithms are used to detect any abnormal behavior, such as instrument drop, contact with non-sterile areas, etc. The analysis results are recorded as abnormal behavior records, including timestamps, abnormality types and severity.

[0153] Finally, the material status report and abnormal behavior records are comprehensively evaluated. The evaluation process uses a weighted scoring system to consider multiple factors such as device integrity, release sequence accuracy, and sterility maintenance to calculate an overall release quality score. Based on this score, detailed operational feedback is generated, including successfully completed operations, issues that require attention, and suggestions for improvement. This information is integrated into a confirmation signal and transmitted to the medical team in the operating room and the central control system.

[0154] For example, in a complex cardiac surgical instrument transport mission, the surgical transport robot carried 50 precision instruments to the operating room. The robot first performed posture calibration, and by fusing data from the gyroscope, accelerometer, and vision system, the position accuracy was improved to ±0.2mm. Subsequently, the robot's storage compartment was unlocked after verifying the position and timestamp. The 3D scanner scanned the interior of the storage compartment at a rate of 1 million points per second, while the RFID reader identified the electronic tags of all 50 instruments. The deep learning model completed instrument identification within 0.5 seconds and generated a list containing the name, model, and location of each instrument. Based on the surgical process, the algorithm generated a material release plan, dividing the 50 instruments into 10 priority groups. The robotic arm began to release the instruments as planned, and each release was confirmed three times. For example, the coordinates of a pair of precision surgical scissors were recorded as (X: 120.5mm, Y: 85.3mm, Z: 75.1mm), relative to the position of the operating table. All instruments were arranged in a 5x10 array, with the most commonly used instruments located in the position that was easiest for doctors to reach. An integrity check found that all 50 instruments were released correctly and without damage.

[0155] Four 4K cameras recorded the entire process from different angles, generating a video stream of about 15 minutes and a total size of 5GB. The video analysis algorithm detected a minor anomaly: Instrument No. 37 slightly touched the storage compartment wall when released, but did not cause contamination. This was recorded in the abnormal behavior record with a severity rating of 1 (out of 10). The comprehensive evaluation gave a release quality score of 96 points (out of 100). Operational feedback suggested slightly adjusting the release angle the next time Instrument No. 37 was released. A confirmation signal was sent to the surgical team, indicating that the instrument was ready and the operation could begin. The entire process took 17 minutes, saving about 30 minutes compared to traditional manual methods, while greatly improving accuracy and safety.

[0156] In a specific embodiment, the process of executing step S106 may specifically include the following steps:

[0157] (1) Performing an environmental scan of the operating room according to the confirmation signal to obtain a distribution map of materials after use, and classifying and marking the distribution map of materials after use to obtain a material recovery list;

[0158] (2) Prioritizing the material recovery list to obtain a recovery sequence table, and planning the path of the surgical transfer robot according to the recovery sequence table to obtain a recovery path;

[0159] (3) Collect and process the used materials one by one along the recycling path to obtain a polluted material group, and perform disinfection status detection on the polluted material group to obtain a disinfection requirement list;

[0160] (4) According to the disinfection demand list, the pollutant resource groups are divided into zones and stored to obtain classified pollutant resource groups, and the quantity of the classified pollutant resource groups is checked to obtain preliminary inventory results;

[0161] (5) Compare the preliminary inventory results with the original sterile material list to obtain a difference report, and perform a second scan of the operating room based on the difference report to obtain information on missing materials;

[0162] (6) Integrate the missing material information with the classified pollutant resource group to obtain a complete pollutant resource group, and perform final verification on the complete pollutant resource group to obtain the inventory result.

[0163] Specifically, the operating room environment is scanned and processed according to the confirmation signal to obtain a distribution map of materials after use. A high-resolution three-dimensional laser scanner and a depth camera are used to perform a full-scale scan of the entire operating room. The scan data is analyzed by point cloud processing and image recognition algorithms to generate a detailed instrument distribution map, marking the precise location and status of each instrument. Subsequently, the distribution map of materials after use is classified and marked to obtain a material recovery list. The classification and marking process uses a machine learning algorithm to classify the instruments into different categories such as surgical knives, clamps, and sutures according to their shape, size, and position, and mark their usage status and cleanliness. Next, the material recovery list is prioritized to obtain a recovery sequence table. The sorting algorithm considers multiple factors, including the importance of the instrument, the degree of contamination, and the accessibility of the location. For example, sharp instruments and easily contaminated instruments are given a higher priority. The surgical transport robot is path-planned according to the recovery sequence table to obtain a recovery path. The path planning uses an improved A* algorithm to consider the layout of the operating room, the location of obstacles, and the distribution of instruments to generate an efficient and safe recovery path.

[0164] Along the recycling path, the surgical transport robot collects and processes the used materials one by one to obtain the contaminated material group. The collection process is performed by the robot's mechanical arm, and each grab is precisely positioned and controlled to ensure that the instrument is not damaged and the sterile state is maintained. The contaminated material group is tested for disinfection status to obtain a disinfection requirements list. The detection process uses ultraviolet spectroscopy and chemical sensors to quickly evaluate the degree of contamination and the required disinfection level of each instrument. The contaminated material group is partitioned and stored according to the disinfection requirements list to obtain a classified contaminated material group. The partitioned storage takes into account the material, degree of contamination and subsequent processing requirements of the instrument, and the instruments are classified and stored in different compartments of the robot. The classified contaminated material group is checked for quantity to obtain a preliminary inventory result. The verification process uses a combination of RFID technology and computer vision to quickly and accurately count the number of each type of instrument.

[0165] The preliminary inventory results are compared with the original sterile material list to obtain a discrepancy report. The comparison process uses database query and matching algorithms to identify missing or redundant instruments. Based on the discrepancy report, the operating room is scanned again to obtain the missing material information. The second scan uses higher-precision scanning equipment and more detailed search algorithms to focus on the problem areas pointed out in the discrepancy report. Finally, the missing material information is integrated with the classified contaminated material group to obtain the complete contaminated material group. The integration process uses a data fusion algorithm to merge the results of the second scan with the previous classification results to form the final sterile material list. The complete contaminated material group is finally verified to obtain the inventory results. The final verification includes quantity verification, status inspection and location confirmation to ensure that all instruments are properly recovered and recorded.

[0166] For example, after a complex heart surgery, the surgical transport robot receives a recovery confirmation signal. The robot first starts a 360-degree environmental scan, using a high-precision laser scanner to scan the entire operating room at a speed of 1 million points per second. After the scan data is processed by point cloud, a three-dimensional instrument distribution map with a resolution of 0.5mm is generated. The image recognition algorithm completes the identification and classification of 60 used instruments within 10 seconds, including 15 scalpels, 20 forceps, 10 staplers and 15 other instruments. The priority sorting algorithm considers the contamination risk and location of the instruments and divides the 60 instruments into 5 priority groups. The path planning module generates a recovery path with a total length of 35 meters, avoiding the operating table and other large equipment. The robot recovers the instruments along this path, and each grasping is controlled by visual positioning and force feedback to ensure safe grasping. During the recovery process, the UV spectrometer scans each instrument for 2 seconds to assess its contamination level. Based on the analysis results, the instruments were divided into three categories: those requiring deep disinfection (25 items), routine disinfection (30 items) and simple cleaning (5 items).

[0167] The robot stored the instruments in three separate sterilization level compartments. The stored instruments were double-checked by RFID readers and high-definition cameras, and the preliminary inventory results showed that a total of 59 instruments were recovered. After comparing with the original list, it was found that a pair of precision scissors was missing. The robot immediately started a secondary scan, using a higher-precision scanner to focus on the area around the operating table. The missing scissors were found in the shadows under the operating table. After integrating the information of this pair of scissors with the previously classified contaminated assets, it was finally confirmed that all 60 instruments had been recovered. During the final verification process, each instrument was triple-verified by RFID identification, visual inspection, and location confirmation. The entire recycling process took 15 minutes, saving about 40 minutes compared to traditional manual methods, while ensuring a 100% instrument recovery rate and accurate sterilization classification, greatly improving the efficiency and safety of post-operative instrument management.

[0168] The disclosed embodiment also provides an intelligent medical material transport system based on an operating room transport robot, such as Figure 2 As shown, the intelligent medical material transport system based on the operating room transport robot specifically includes:

[0169] The processing module 201 is used to analyze and process the surgical demand information to obtain a sterile material combination plan, perform load balancing processing on the sterile material combination plan to obtain a loading instruction, and perform loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state;

[0170] A perception module 202 is used to perform environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and perform path generation processing on the environmental map to obtain a transport path;

[0171] The obstacle avoidance module 203 is used to perform obstacle avoidance processing on the motion state of the surgical transport robot based on the transport path to obtain an obstacle avoidance strategy, and perform decision processing on the obstacle avoidance strategy to obtain a running trajectory;

[0172] The verification module 204 is used to perform identity authentication processing on the surgical transport robot that arrives at the operating room according to the running trajectory to obtain a verification result, and perform docking processing on the surgical transport robot that passes the verification to obtain a docking state;

[0173] The monitoring module 205 is used to release the materials in the surgical transport robot in the docked state to obtain a material array, and monitor the release process to obtain a confirmation signal;

[0174] The recycling module 206 is used to recycle the polluted materials in the material array after use based on the confirmation signal to obtain a polluted material group, and verify the polluted material group to obtain an inventory result.

[0175] Through the collaborative work of the above modules, through the analysis and processing of surgical demand information and the generation of sterile material combination plans, the precise and personalized configuration of medical materials is achieved, ensuring that the materials required for each operation can be optimally configured, thereby improving the efficiency and accuracy of surgical preparation. The introduction of load balancing processing not only optimizes the distribution of materials in the robot, but also improves the stability and safety during transportation, and reduces the risk of material damage or contamination. The environmental perception processing and path generation functions enable the robot to adapt to the complex and changing hospital environment in real time, and autonomously plan the optimal transfer path, which greatly improves the efficiency of material transportation and reduces the consumption of human resources. The obstacle avoidance processing and decision processing functions in the motion state further enhance the robot's adaptability in complex environments, ensure the safety and reliability of the material transportation process, and effectively reduce the incidence of collision accidents. The identity authentication processing and docking processing after arriving at the operating room not only improve the safety of material handover, but also realize accurate positioning and docking, preparing for the subsequent release of materials. The materials are released in the docked state, and the confirmation signal is generated through monitoring processing. This design ensures the accuracy and traceability of the material handover process, and effectively reduces the risk of human errors and material loss. Finally, the recycling and verification of contaminated materials after use realizes the full life cycle management of medical materials, which not only improves the efficiency and accuracy of material recycling, but also provides reliable data support for subsequent disinfection, reuse and replenishment. Through precise material management and tracking, this method helps to optimize the hospital's inventory management, reduce waste and reduce operating costs.

[0176] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps recorded in this disclosure can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and this document does not limit this.

[0177] The above specific implementations do not constitute a limitation on the protection scope of the present disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present disclosure shall be included in the protection scope of the present disclosure.

Claims

1. An intelligent medical material transportation method based on an operating room transportation robot, characterized in that: include: Analyze and process the surgical demand information to obtain a sterile material combination plan, perform load balancing on the sterile material combination plan to obtain a loading instruction, and perform loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state; Performing environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and performing path generation processing on the environmental map to obtain a transport path; Based on the transport path, performing obstacle avoidance processing on the motion state of the surgical transport robot to obtain an obstacle avoidance strategy, and performing decision processing on the obstacle avoidance strategy to obtain an operation trajectory; According to the running trajectory, identity authentication processing is performed on the surgical transport robot that arrives at the operating room to obtain a verification result, and docking processing is performed on the surgical transport robot that passes the verification to obtain a docking state; In the docked state, releasing the materials in the surgical transport robot to obtain a material array, and monitoring the releasing process to obtain a confirmation signal; Based on the confirmation signal, the contaminated materials in the material array after use are recycled to obtain a contaminated material group, and the contaminated material group is verified to obtain an inventory result.

2. The intelligent medical material transportation method based on the operating room transportation robot according to claim 1 is characterized in that: The step of analyzing and processing the surgical requirement information to obtain a sterile material combination scheme, performing load balancing processing on the sterile material combination scheme to obtain a loading instruction, and performing loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state includes: Classify the surgical demand information to obtain surgical categories, and perform statistical analysis on historical surgical data according to the surgical categories to obtain a material usage frequency list; Sort the materials by importance according to the material usage frequency list to obtain a core sterile material list, and perform combination optimization on the core sterile material list to obtain an initial sterile material plan; Performing user preference analysis on the initial sterile material plan to obtain personalized adjustment suggestions, and modifying the initial sterile material plan according to the personalized adjustment suggestions to obtain a sterile material combination plan; Calculating the weight distribution of materials in the sterile material combination scheme to obtain center of gravity data, and adjusting the positions of the materials according to the center of gravity data to obtain a balanced layout scheme; Performing temperature sensitivity analysis on the balanced layout plan to obtain temperature control requirements, and optimizing material placement according to the temperature control requirements to obtain a temperature balance plan; A three-dimensional visualization model is generated according to the temperature balance scheme, a loading indication diagram is extracted from the three-dimensional visualization model, and sterile materials are positioned and placed according to the loading indication diagram to obtain a surgical transport robot in a loaded state.

3. The intelligent medical material transportation method based on the operating room transportation robot according to claim 1 is characterized in that: The step of performing environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and performing path generation processing on the environmental map to obtain a transport path includes: Performing fusion processing on multi-sensor data of the surgical transport robot in the loaded state to obtain original environmental data, and performing noise filtering on the original environmental data to obtain cleaned environmental data; Performing feature extraction on the cleaned environmental data to obtain an environmental feature set, and performing semantic segmentation on the environmental feature set to obtain semantically labeled data; Identifying static obstacles in the environment according to the semantic tag data to obtain a static obstacle distribution map, and performing rasterization processing on the static obstacle distribution map to obtain a raster map; Constructing a spatial index for the grid map to obtain a fast query structure, and dynamically updating the environment based on the fast query structure to obtain a real-time environment map; Analyzing the traversable area of ​​the real-time environment map to obtain a candidate path set, and evaluating the candidate path set to obtain a path score list; An optimal path is selected according to the path scoring list to obtain a target transfer path, and the target transfer path is smoothed to obtain a transfer path.

4. The intelligent medical material transportation method based on the operating room transportation robot according to claim 1 is characterized in that: The step of performing obstacle avoidance processing on the motion state of the surgical transport robot based on the transport path to obtain an obstacle avoidance strategy, and performing decision processing on the obstacle avoidance strategy to obtain a running trajectory includes: Segmenting the transport path to obtain a path segment sequence, and performing risk assessment on the path segment sequence to obtain a risk level table; The speed of the surgical transport robot is adjusted according to the risk level table to obtain a speed control parameter, and the motion state of the surgical transport robot is predicted based on the speed control parameter to obtain a predicted trajectory; Scanning the environment around the predicted trajectory in real time to obtain dynamic obstacle data, and performing motion trend analysis on the dynamic obstacle data to obtain an obstacle motion prediction result; Calculating the potential collision risk according to the obstacle motion prediction result to obtain a collision risk assessment report, and generating multiple alternative obstacle avoidance schemes based on the collision risk assessment report to obtain an obstacle avoidance strategy set; Scoring the safety and efficiency of each strategy in the obstacle avoidance strategy set to obtain a strategy scoring table, and selecting the best obstacle avoidance strategy according to the strategy scoring table to obtain a target obstacle avoidance strategy; The motion parameters of the surgical transport robot are adjusted based on the target obstacle avoidance strategy to obtain a corrected motion instruction, and a real-time control signal is generated according to the corrected motion instruction to obtain a running trajectory.

5. The intelligent medical material transportation method based on the operating room transportation robot according to claim 1 is characterized in that: The step of performing identity authentication processing on the surgical transport robot arriving at the operating room according to the running trajectory to obtain a verification result, and performing docking processing on the surgical transport robot that passes the verification to obtain a docking state includes: Performing precise positioning processing on the running trajectory to obtain the real-time position coordinates of the surgical transport robot, and triggering an identity authentication request according to the real-time position coordinates to obtain a verification task instruction; Scanning the hardware identification of the surgical transfer robot to obtain a unique identification code of the robot, and encrypting the unique identification code of the robot to obtain an encrypted identification; Querying the authorization database according to the encrypted identifier to obtain permission information, and performing comparison and analysis on the permission information to obtain a preliminary verification result; Performing image acquisition processing on the appearance features of the surgical transfer robot to obtain a feature image, and performing pattern recognition processing on the feature image to obtain a visual verification result; Comprehensively evaluating the preliminary verification result and the visual verification result to obtain a verification result, and generating an access token based on the verification result to obtain an identity credential; Performing a validity check on the identity credential to obtain an access instruction, and controlling the operating room access control system according to the access instruction to obtain a door opening signal; The surgical transport robot is accurately guided based on the door opening signal to obtain a docking path, and the surgical transport robot is controlled to move to a preset position according to the docking path to obtain a docking state.

6. The intelligent medical material transportation method based on the operating room transportation robot according to claim 1 is characterized in that: The step of releasing the materials in the surgical transport robot in the docked state to obtain a material array, and monitoring the release process to obtain a confirmation signal includes: Performing posture calibration processing on the surgical transport robot in the docked state to obtain a precise docking position, and unlocking the robot storage compartment according to the precise docking position to obtain an open state; Scan and identify the materials in the storage compartment in the open state to obtain a list of sterile materials, and generate a release sequence according to the list of sterile materials to obtain a material release plan; According to the material release plan, materials in the storage compartment are taken out one by one to obtain a release sequence, and each material in the release sequence is positioned to obtain material coordinates; Arranging the released materials based on the material coordinates to obtain a material array, and performing integrity check on the material array to obtain a material status report; Performing real-time video acquisition and processing on the entire release process to obtain a monitoring video stream, and performing intelligent analysis and processing on the monitoring video stream to obtain an abnormal behavior record; The material status report and the abnormal behavior record are comprehensively evaluated and processed to obtain a release quality score, and operation feedback is generated according to the release quality score to obtain a confirmation signal.

7. The intelligent medical material transportation method based on the operating room transportation robot according to claim 6 is characterized in that: The step of recycling the contaminated materials in the material array after use based on the confirmation signal to obtain a contaminated material group, and verifying the contaminated material group to obtain an inventory result includes: Performing an environmental scan of the operating room according to the confirmation signal to obtain a distribution map of materials after use, and classifying and marking the distribution map of materials after use to obtain a material recovery list; Prioritizing the material recovery list to obtain a recovery sequence table, and planning a path for the surgical transport robot according to the recovery sequence table to obtain a recovery path; Collect and process the used materials one by one along the recycling path to obtain a polluted resource group, and perform disinfection status detection on the polluted resource group to obtain a disinfection requirement list; According to the disinfection requirement list, the polluted resource groups are partitioned and stored to obtain classified polluted resource groups, and the quantity of the classified polluted resource groups is checked to obtain preliminary inventory results; Comparing the preliminary inventory results with the original sterile material list to obtain a difference report, and performing a second scan of the operating room based on the difference report to obtain missing material information; The missing material information is integrated with the classified polluted material group to obtain a complete material group, and the complete material group is finally verified to obtain an inventory result.

8. An intelligent medical material transport system based on an operating room transport robot, used to execute the intelligent medical material transport method based on an operating room transport robot as claimed in any one of claims 1 to 7, characterized in that: include: A processing module, used to analyze and process the surgical demand information to obtain a sterile material combination plan, perform load balancing processing on the sterile material combination plan to obtain a loading instruction, and perform loading processing on the sterile materials according to the loading instruction to obtain a surgical transport robot in a loaded state; A perception module, used to perform environmental perception processing on the surgical transport robot in the loaded state to obtain an environmental map, and perform path generation processing on the environmental map to obtain a transport path; An obstacle avoidance module, used to perform obstacle avoidance processing on the motion state of the surgical transport robot based on the transport path to obtain an obstacle avoidance strategy, and perform decision processing on the obstacle avoidance strategy to obtain a running trajectory; A verification module, used to perform identity authentication processing on the surgical transport robot arriving at the operating room according to the running trajectory to obtain a verification result, and to perform docking processing on the surgical transport robot that has passed the verification to obtain a docking state; A monitoring module, used for releasing the materials in the surgical transport robot in the docked state to obtain a material array, and monitoring the release process to obtain a confirmation signal; The recycling module is used to recycle the polluted materials in the material array after use based on the confirmation signal to obtain a polluted material group, and verify the polluted material group to obtain an inventory result.

Images