AI intelligent movable type operation affected limb lifting auxiliary device and control method

Abstract

The invention discloses an AI intelligent movable type surgical affected limb lifting auxiliary device and a control method, and relates to the field of technical intelligent medical equipment.The AI intelligent movable type surgical affected limb lifting auxiliary device comprises a movable base, a rotating base is installed in the middle of the movable base, and the lower portion of the rotating base is fixedly installed on a top shell of the movable base; the four-degree-of-freedom mechanical arm is installed on the upper portion of the rotating base, the four-degree-of-freedom mechanical arm is connected with the movable base through the rotating base, and the four-degree-of-freedom mechanical arm is composed of a large mechanical arm, a small mechanical arm, a spherical joint, a telescopic sliding rail, a servo motor and a hydraulic cylinder. By installing the four-degree-of-freedom mechanical arm, the function of accurately lifting an affected limb is achieved, the four-degree-of-freedom mechanical arm has extremely high movement precision and bearing capacity, the affected limb can be rapidly and accurately positioned and maintained in the optimal posture needed by an operation, and fatigue errors caused by manual lifting are avoided.

Description

Technical Field

[0001] This invention relates to the field of intelligent medical device technology, specifically to an AI-powered intelligent mobile surgical limb lifting assist device and control method. Background Technology

[0002] In orthopedic, microsurgical and trauma repair surgeries, the stable lifting, precise posture maintenance and dynamic adjustment of the affected limb during surgery are key to ensuring surgical success.

[0003] Traditional limb elevation relies on manual lifting, which is prone to fatigue errors. Traditional fixed supports cannot achieve dynamic adjustment of the affected limb during surgery, requiring manual operation by medical staff, which increases the risk of infection. The positional maintenance accuracy is insufficient. Electric lifting devices lack intelligent obstacle avoidance and cross-operating room movement capabilities, relying on manual handling. Voice control devices have insufficient recognition accuracy in the complex sound field environment of the operating room.

[0004] Patent document CN114587897B discloses a limb fixation device for heart rate monitoring during emergency cardiac surgery. The patent achieves the function of fixing the patient's limb. When the limb height needs to be adjusted, the micro motor in the height adjustment mechanism is activated. The third spur gear, in conjunction with the first spur gear, drives the three second spur gears to rotate in the same direction. The lifting threaded rod will lift the limb fixation ring upwards, which can be adjusted as needed.

[0005] The aforementioned patent uses a motor to drive a threaded movable rod, which in turn moves a rectangular movable block. This block can be adjusted to a suitable position according to the patient's body shape, making it convenient to adjust according to the patient's size. However, the control over the lifting height of the affected limb is not precise enough.

[0006] Therefore, this application proposes an AI-powered intelligent mobile surgical limb lifting assist device and control method capable of precisely lifting the affected limb. Summary of the Invention

[0007] The purpose of this invention is to provide an AI-powered intelligent mobile surgical limb lifting assistance device and control method to solve the technical problem of insufficient limb lifting accuracy mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution: an AI-powered intelligent mobile surgical limb lifting assist device, comprising a mobile base, a rotating base mounted in the middle of the mobile base, a lower part of the rotating base fixedly mounted on the top shell of the mobile base, and a four-degree-of-freedom robotic arm mounted on the upper part of the rotating base. The four-degree-of-freedom robotic arm is connected to the mobile base via the rotating base. The four-degree-of-freedom robotic arm consists of a robotic upper arm, a robotic lower arm, a ball joint, a telescopic slide rail, a servo motor, and a hydraulic cylinder. One end of the robotic upper arm penetrates the upper surface of the rotating base and is fitted with a hydraulic cylinder at its end. The other end of the robotic upper arm is connected to the robotic lower arm via a ball joint. The ball joint has a built-in six-dimensional force sensor for detecting contact posture. The telescopic slide rail is located on the side of the robotic upper arm and, in conjunction with the hydraulic cylinder connected to the robotic upper arm, enables the extension and retraction of the robotic arm. The servo motor passes through the lower part of the rotating base and drives the rotation of the rotating base and the movement of the four-degree-of-freedom robotic arm.

[0009] Preferably, the end of the four-degree-of-freedom robotic arm is connected to an adaptive lifting platform. The adaptive lifting platform is equipped with a detachable sterile sheath interface and a gyroscope. A contact surface temperature and humidity sensor is installed at the center of the adaptive lifting platform. Twelve medical silicone airbags are arranged in a ring around the center of the adaptive lifting platform. Each medical silicone airbag is equipped with a pressure sensor and an LED indicator. The pressure of each medical silicone airbag is independently controlled and the pressure area is displayed by the LED indicator, realizing the visualization of pressure distribution. A miniature UVC disinfection lamp array is embedded in the adaptive lifting platform around the medical silicone airbags.

[0010] Preferably, one end of the adaptive lifting platform is equipped with an AR projection positioning component, which consists of a DLP projection module, a TOF depth camera, and a dynamic calibration target. The DLP projection module, coated with a medical anti-glare coating, directly projects surgical planning information and positioning marks onto the surface of the affected limb. The TOF depth camera acquires three-dimensional point cloud data of the affected limb in real time to calibrate the projection position. The dynamic calibration target uses computer vision algorithms to ensure that the projected content accurately matches the anatomical structure of the affected limb, accurately projecting virtual information onto the physical entity, and realizing visual guidance for intraoperative positioning.

[0011] Preferably, the side of the mobile base is equipped with a lidar array and an atomizing sterilizer. The atomizing sterilizer is linked with a miniature UVC disinfection lamp array to achieve dynamic disinfection. The upper surface of the mobile base is equipped with an OLED touch screen, a ring microphone array, and an emergency physical joystick. The OLED touch screen, equipped with an operating room-specific anti-fog panel, provides a graphical user interface and real-time status feedback. The ring microphone array with 8-channel noise suppression is used for 360° sound pickup and suppresses environmental noise based on beamforming technology. The emergency physical joystick with IP67 protection rating is used for manual redundant control when the voice system fails. The OLED touch screen, ring microphone array, and emergency physical joystick together form a human-machine interaction unit.

[0012] Preferably, the mobile base is equipped with a central control module. The central control module is connected to the hospital's smart logistics system and supports HL7 protocol compatibility. The central control module integrates an AI intelligent controller, a SLAM positioning unit, and a storage unit. The SLAM positioning unit has a built-in UWB positioning module and works with a lidar array to build a three-dimensional point cloud map. The AI ​​intelligent controller acquires patient position data in real time through the HL7 protocol, automatically matches the pre-stored templates in the storage unit, and generates a personalized support solution by combining AI algorithms. The storage unit saves detailed operation procedures and generates operation logs.

[0013] Preferably, the upper surface of the rear of the mobile base is provided with a speaker and an alarm light. The voice alarm of the speaker and the optical warning and vibration reminder of the alarm light constitute a three-level safety early warning mechanism for the device.

[0014] Preferably, the bottom of the mobile base is provided with an omnidirectional drive wheel set, which includes a built-in drive motor and a Mecanum wheel set. A power supply compartment is provided in the middle of the mobile base. An emergency physical disconnect switch is provided on the surface of the power supply compartment. A lithium battery pack is provided inside the power supply compartment. The lithium battery pack is connected to each component of the device through signal lines to provide power for the operation of each component of the device.

[0015] Preferably, the control method includes the following steps:

[0016] S1. Voiceprint Activation and Command Parsing: The device is activated by the voiceprint features of pre-registered medical staff, and the voice commands are parsed after verification.

[0017] S2. Environmental perception and modeling: Construct a 3D point cloud map of the operating room using a lidar array and update obstacle information in real time;

[0018] S3. Dynamic Path Planning and Navigation: Combining AI algorithms and artificial potential field methods, a collision-free path is planned, and the control device enables autonomous movement and real-time obstacle avoidance to reach the target location.

[0019] S4. Force-position hybrid precision control: Adjust the support torque of the four-degree-of-freedom robotic arm to complete the precise lifting and posture adjustment of the affected limb, and achieve millimeter-level precise positioning and adaptive support of the affected limb;

[0020] S5. Safety monitoring and emergency response: Real-time monitoring of data acquired by pressure sensors. When an abnormal situation occurs, the device automatically triggers a three-level safety warning mechanism and immediately locks the four-degree-of-freedom robotic arm.

[0021] Preferably, in step S3, a Kalman filter algorithm is used to predict the movement trajectory of moving obstacles, and an electronic fence is set up in the sterile area of ​​the surgery to maintain a minimum safe distance of 200mm.

[0022] Preferably, S4 specifically comprises:

[0023] S4.1 Setting targets and continuous monitoring: The system sets the target position and posture of the robotic arm according to voice commands;

[0024] S4.2 Calculate the deviation between the target angle and the actual angle and its rate of change, and calculate the precise torque to be applied in real time;

[0025] S4.3 Output execution and dynamic adjustment: The central control module controls the movement of the four-degree-of-freedom robotic arm according to the calculated torque value, eliminates angle deviation, and continuously fine-tunes the output torque based on the real-time feedback of the affected limb to achieve adaptive compliant support.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention achieves precise lifting of the affected limb by installing a four-degree-of-freedom robotic arm. The four-degree-of-freedom robotic arm has extremely high motion accuracy and load-bearing capacity, and can quickly and accurately position the affected limb and maintain it in the optimal posture required for surgery, avoiding fatigue errors caused by manual lifting.

[0028] 2. This invention, by installing an adaptive lifting platform, achieves the function of flexible and stable support for the affected limb, monitors and adaptively adjusts the pressure distribution in real time, avoids excessive local pressure, effectively prevents intraoperative complications such as nerve compression and poor blood circulation, and improves surgical safety and patient comfort.

[0029] 3. This invention, by installing a mobile base, enables autonomous movement across operating rooms. The equipment can be efficiently shared among multiple operating rooms, shortening the switching time between operating rooms. Combined with dynamic disinfection, it improves the disinfection qualification rate and significantly enhances the turnover efficiency of operating rooms and the level of hospital infection control.

[0030] 4. This invention, by installing a central control module, realizes the function of full-process automation, integrates voiceprint recognition, AI navigation and force-position hybrid control, realizes intraoperative voice interaction, reduces the risk of infection, reduces the workload of medical staff, and improves scheduling efficiency and surgical intelligence through AI algorithm real-time path planning and dynamic adjustment. Attached Figure Description

[0031] Figure 1 This is a front view structural diagram of the present invention;

[0032] Figure 2 This is a front view of the present invention.

[0033] Figure 3 This is a top view of the movable base structure of the present invention;

[0034] Figure 4 This is a schematic diagram of the robotic arm structure of the present invention;

[0035] Figure 5 This is a schematic diagram of the lifting platform structure of the present invention;

[0036] Figure 6 This is a schematic diagram of the AR projection positioning component structure of the present invention;

[0037] Figure 7 This is a schematic diagram of the central control module structure of the present invention;

[0038] Figure 8 This is a schematic diagram of the intelligent control process of the present invention.

[0039] In the diagram: 1. Mobile base; 2. Four-DOF robotic arm; 3. Rotating base; 4. Adaptive lifting platform; 5. AR projection positioning component; 6. Omnidirectional drive wheel assembly; 7. LiDAR array; 8. Power supply compartment; 9. Central control module; 10. Lithium battery pack; 11. OLED touchscreen; 12. Ring microphone array; 13. Emergency physical joystick; 14. Atomizing sterilizer; 15. Speaker; 16. Robotic upper arm; 17. Robotic lower arm; 18. Ball joint; 19. 20. Telescopic slide rail; 21. Servo motor; 22. Hydraulic cylinder; 23. Medical silicone airbag; 24. Pressure sensor; 25. LED indicator light; 26. Miniature UVC disinfection lamp array; 27. DLP projection module; 28. TOF depth camera; 29. ​​Dynamic calibration target; 30. AI intelligent controller; 31. SLAM positioning unit; 32. Storage unit; 33. Alarm light; 34. Emergency physical disconnect switch; 35. Contact surface temperature and humidity sensor; 36. Signal line. Detailed Implementation

[0040] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0041] Example 1: Please refer to Figure 1 , Figure 2 , Figure 4 , Figure 5 and Figure 7 An AI-powered mobile surgical limb lifting assist device includes a mobile base 1, a rotating base 3 mounted in the middle of the mobile base 1, a lower part of the rotating base 3 fixedly mounted on the top shell of the mobile base 1, and a four-degree-of-freedom robotic arm 2 mounted on the upper part of the rotating base 3. The four-degree-of-freedom robotic arm 2 is connected to the mobile base 1 via the rotating base 3. The four-degree-of-freedom robotic arm 2 consists of a large robotic arm 16, a small robotic arm 17, a ball joint 18, a telescopic slide rail 19, a servo motor 20, and a hydraulic cylinder 21. One end of the large robotic arm 16 penetrates the upper surface of the rotating base 3 and is fitted with a hydraulic cylinder 21 at its end. The other end of the large robotic arm 16 is connected to the small robotic arm 17 via the ball joint 18. The telescopic slide rail 19 is located on the side of the large robotic arm 16. The servo motor 20 is located in the lower part of the rotating base 3 and drives the rotation of the rotating base 3 and the movement of the four-degree-of-freedom robotic arm 2.

[0042] The end of the four-degree-of-freedom robotic arm 2 is connected to an adaptive lifting platform 4. The adaptive lifting platform 4 is equipped with a detachable sterile sheath interface and a gyroscope. A contact surface temperature and humidity sensor 34 is installed at the center of the adaptive lifting platform 4. Twelve medical silicone airbags 22 are arranged in a ring around the center of the adaptive lifting platform 4. Each medical silicone airbag 22 is equipped with a pressure sensor 23 and an LED indicator 24. The pressure of each medical silicone airbag 22 is independently controlled and the pressure area is displayed by the LED indicator 24. A miniature UVC disinfection lamp array 25 is embedded on the adaptive lifting platform 4 around the medical silicone airbags 22.

[0043] The side of the mobile base 1 is equipped with a lidar array 7 and an atomizing sterilizer 14. The atomizing sterilizer 14 is linked with the miniature UVC disinfection lamp array 25 to achieve dynamic disinfection.

[0044] Furthermore, in orthopedic surgery, especially tibial fracture internal fixation, stable lifting and precise posture maintenance of the affected limb are crucial to the success of the surgery. In this embodiment, the AI ​​intelligent mobile surgical limb lifting assist device is activated through the voiceprint recognition system. The surgeon wakes up the device by giving the voice command "start lower limb support mode". The limb lifting assist device constructs a three-dimensional map of the operating room in real time through the lidar array 7 and autonomously navigates to the predetermined position on the side of the operating table in combination with the SLAM positioning unit 30.

[0045] The four-degree-of-freedom robotic arm 2 automatically adjusts the angles of the telescopic slide rail 19 and the ball joint 18 based on preset patient weight and limb size parameters, relying on hydraulic drive and high-precision servo motor 20, to achieve precise extension and positioning of the four-degree-of-freedom robotic arm 2. The 12 sets of medical silicone airbags 22 on the adaptive lifting platform 4 obtain real-time pressure data based on the built-in pressure sensor 23, and display the pressure area through LED indicator 24 to achieve visualization of pressure distribution. The medical silicone airbags 22 dynamically adjust the air pressure distribution to ensure that the limb is subjected to uniform force during the lifting process and avoid excessive local pressure that could lead to nerve or blood vessel compression.

[0046] During the procedure, the surgeon can fine-tune the height and angle of the affected limb in real time via voice commands. The central control module 9 achieves millimeter-level position adjustment through a force-position hybrid control algorithm. The force-position hybrid control is based on the formula F=Kp×Δθ+Kd×dΔθ / dt, which adjusts the support torque. Kp and Kd are preset parameters, Kp=2.3N·m / rad, Kd=0.15N·m·s / rad, Δθ is the angle deviation, and dΔθ / dt is the rate of change. The four-degree-of-freedom robotic arm 2 has a real-time tremor compensation function, which can effectively filter out even the slight vibrations caused by the surgeon operating the instruments, ensuring the stability of the affected limb. The lifting assistance device continuously monitors pressure and temperature data throughout the procedure. Once an abnormality is detected, a three-level safety warning mechanism is immediately triggered to ensure surgical safety.

[0047] Postoperatively, the lifting assist device automatically sterilizes the operating room and itself using a miniature UVC disinfection lamp array 25 and an atomizing sterilizer 14. After disinfection, the lifting assist device automatically folds and stores itself and returns to its charging position to wait for use. The entire process requires no manual intervention, reducing the flow of personnel during surgery and lowering the risk of infection. It is suitable for complex orthopedic surgeries that take a long time to perform.

[0048] For example, in an internal fixation surgery for a comminuted tibial fracture, the patient needs to maintain stable elevation and precise positioning of the affected limb to ensure accurate implantation of the plate and screws after fracture reduction. With traditional manual elevation methods, prolonged support by medical staff can easily lead to arm fatigue, causing slight displacement of the affected limb and potentially resulting in screw placement errors, thus affecting the surgical outcome.

[0049] After adopting this device, medical staff issue the command "Activate lower limb support mode" via pre-registered voiceprints, and the device is quickly activated. A lidar array constructs a real-time 3D map of the operating room, and the SLAM positioning unit, in conjunction with omnidirectional drive wheels, autonomously navigates to a predetermined position on the side of the operating table. The four-degree-of-freedom robotic arm, based on preset parameters such as patient weight and affected limb size, drives a rotating base via a servo motor. Combined with the synergistic effect of hydraulic cylinders and telescopic rails, it precisely adjusts the angles of the upper and lower robotic arms, lifting the affected limb to the optimal position required for surgery. Simultaneously, the central control module uses a force-position hybrid control algorithm (based on formulas)... in It achieves millimeter-level position adjustment, and even if there is a slight vibration during surgery, it can compensate in real time through algorithms to maintain the stability of the affected limb.

[0050] During the surgery, the surgeon could fine-tune the height and angle of the affected limb in real time via voice commands, eliminating the need for manual operation and avoiding errors inherent in human intervention. Ultimately, the surgery was successfully completed, and postoperative imaging showed precise placement of the plate and screws. The patient recovered well postoperatively without any complications caused by limb positioning deviations. This case clearly demonstrates that the device, through the high-precision motion and force-position hybrid control of a four-degree-of-freedom robotic arm, significantly improves surgical accuracy and reduces the risk of surgical errors.

[0051] Example 2: Please refer to Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 6 , Figure 7 and Figure 8 An AI-powered mobile surgical limb lifting assist device includes a mobile base 1, a rotating base 3 mounted in the middle of the mobile base 1, a lower part of the rotating base 3 fixedly mounted on the top shell of the mobile base 1, and a four-degree-of-freedom robotic arm 2 mounted on the upper part of the rotating base 3. The four-degree-of-freedom robotic arm 2 is connected to the mobile base 1 through the rotating base 3. The four-degree-of-freedom robotic arm 2 consists of a robotic upper arm 16, a robotic lower arm 17, a ball joint 18, a telescopic slide rail 19, a servo motor 20, and a hydraulic cylinder 21.

[0052] An AR projection positioning component 5 is installed at one end of the adaptive lifting platform 4. The AR projection positioning component 5 consists of a DLP projection module 26, a TOF depth camera 27, and a dynamic calibration target 28, which accurately projects virtual information onto the physical entity to achieve visual guidance for intraoperative positioning.

[0053] A ring microphone array 12 with 8-channel noise suppression is used for 360° sound pickup and suppresses ambient noise based on beamforming technology;

[0054] The upper surface of the rear of the mobile base 1 is equipped with a speaker 15 and an alarm light 32. The voice alarm of the speaker 15 and the optical warning and vibration reminder of the alarm light 32 constitute a three-level safety early warning mechanism for the device.

[0055] Furthermore, in upper limb surgery, such as elbow joint replacement or fracture repair, precise positioning and stable support of the affected limb are equally crucial. In this embodiment, the doctor activates the AI ​​intelligent mobile surgical limb lifting assistance device by using the voice command "start upper limb mode". The ring microphone array 12, together with the deep learning noise reduction chip, realizes medical-grade voice interaction. The central control module 9 automatically recognizes the physiological curve of the patient's upper limb, obtains the three-dimensional point cloud data of the affected limb through the TOF depth camera 27, and calculates the optimal support point by combining the AI ​​algorithm.

[0056] The four-degree-of-freedom robotic arm 2 achieves flexible adjustment in multiple angles and directions through the coordinated control of hydraulic cylinder 21 and servo motor 20. The 12 sets of medical silicone airbags 22 on the adaptive lifting platform 4 are adjusted in zones according to real-time pressure feedback to ensure that the elbow and wrist joints are in a natural and relaxed state, avoiding muscle strain or nerve damage caused by improper positioning during the operation. The AR projection positioning component 5 projects the surgical path and key anatomical structures onto the surface of the affected limb to provide visual guidance.

[0057] The limb lifting assist device also has a real-time voice feedback function. During the execution of instructions, it continuously reports the current status to the doctor through the speaker 15, such as "left arm raised 10 cm, pressure distribution normal". If an emergency occurs during the operation, such as the patient's body position sliding or the four-degree-of-freedom robotic arm 2 being subjected to abnormal force, the central control module 9 immediately activates the emergency safety mechanism. It issues a warning through three methods: optical warning, voice alarm and vibration reminder, and automatically locks the movement of the four-degree-of-freedom robotic arm 2 to avoid accidents and ensure the safety of the operation.

[0058] For example, when an elderly patient with severe elbow osteoarthritis requires elbow replacement surgery, the surgery necessitates precise positioning of the prosthesis to ensure it matches the patient's own skeletal anatomy and restore elbow function. In traditional surgery, surgeons rely primarily on experience to determine the location of anatomical structures, which carries the risk of prosthesis placement errors and may lead to postoperative problems such as limited joint movement and pain.

[0059] After applying this device, the doctor activates it via voice command "Activate Upper Limb Mode". The ring microphone array, utilizing 8-channel noise suppression and beamforming technology, accurately recognizes the command in the complex acoustic environment of the operating room. The AR projection positioning component then activates, and the DLP projection module, coated with a medical anti-glare coating, projects the pre-planned prosthesis installation path and key anatomical structure markers directly onto the surface of the patient's affected limb. The TOF depth camera acquires real-time 3D point cloud data of the affected limb, dynamically calibrating the projection position. The dynamic calibration target, through computer vision algorithms, ensures that the projected content precisely matches the anatomical structure of the affected limb.

[0060] During the surgery, the surgeon accurately determined the precise position and angle of the prosthesis placement based on the projected markings on the surface of the affected limb, effectively avoiding anatomical structure identification errors. Simultaneously, the four-degree-of-freedom robotic arm, through the coordinated control of hydraulic cylinders and servo motors, achieved flexible adjustments of the affected limb at multiple angles and directions. The 12 sets of medical-grade silicone airbags on the adaptive lifting platform dynamically adjusted the air pressure distribution based on pressure sensor feedback, ensuring the elbow joint remained in a naturally relaxed state and preventing muscle strain during surgery. Ultimately, the prosthesis was precisely installed, and the patient experienced good recovery of elbow joint mobility and significant pain relief post-operatively. This case demonstrates that the device's AR projection positioning component provides precise visual guidance for complex surgeries, assisting surgeons in performing delicate operations and improving surgical success rates.

[0061] Example 3: Please refer to Figure 1 , Figure 2 , Figure 7 and Figure 8 An AI-powered mobile surgical limb lifting assist device includes a mobile base 1, a rotating base 3 mounted in the middle of the mobile base 1, a lower part of the rotating base 3 fixedly mounted on the top shell of the mobile base 1, a four-degree-of-freedom robotic arm 2 mounted on the upper part of the rotating base 3, the four-degree-of-freedom robotic arm 2 connected to the mobile base 1 via the rotating base 3, a miniature UVC disinfection lamp array 25 embedded in the adaptive lifting platform 4 surrounding the medical silicone airbag 22, and a laser radar array 7 and an atomizing disinfector 14 arranged on the side of the mobile base 1. The atomizing disinfector 14 is linked with the miniature UVC disinfection lamp array 25 to achieve dynamic disinfection.

[0062] The central control module 9 integrates an AI intelligent controller 29, a SLAM positioning unit 30, and a storage unit 31. The SLAM positioning unit 30 works with the lidar array 7 to build a three-dimensional point cloud map, and the storage unit 31 saves detailed operation procedures and generates operation logs.

[0063] The bottom of the mobile base 1 is provided with an omnidirectional drive wheel set 6, which includes a built-in drive motor and a Mecanum wheel set.

[0064] Furthermore, in this embodiment, the AI ​​intelligent mobile surgical limb lifting assistance device demonstrates its ability to move autonomously across operating rooms and collaborate with multiple devices. When the hospital HIS system issues a scheduling command, the central control module 9 of the lifting assistance device autonomously plans the optimal path, avoiding surgical carts, personnel, and other moving obstacles. It achieves smooth and precise movement through the omnidirectional drive wheel set 6. The lifting assistance device updates the environmental map in real time through the lidar array 7 and predicts the movement trajectory of dynamic obstacles by combining the Kalman filter algorithm, ensuring that a safe distance is always maintained during the movement. After arriving at the target operating room, the lifting assistance device performs precise positioning through the UWB positioning module built into the SLAM positioning unit 30, with an error of less than 3mm.

[0065] Between multiple operating rooms, multiple limb lifting assist devices can form a collaborative disinfection matrix. Through wireless communication, they coordinate their respective disinfection areas to achieve efficient and seamless inter-operation transition. During the movement of the lifting assist device, the miniature UVC disinfection lamp array 25 and the atomizing disinfector 14 are continuously running, which complies with the "Technical Specifications for Disinfection in Medical Institutions" and ensures that the surface of the lifting assist device and the operating room are thoroughly disinfected.

[0066] In addition, the limb lifting assist device supports seamless integration with the hospital's smart logistics system, and the storage unit 31 uploads operation logs and device status data in real time, providing data support for hospital management.

[0067] For example, our orthopedics department performs five different types of orthopedic surgeries daily: femoral fracture surgery, spinal fusion surgery, ankle fracture surgery, shoulder replacement surgery, and wrist fracture surgery, all performed in three different operating rooms. Traditionally, surgical equipment had to be manually moved, which was not only time-consuming and labor-intensive but could also cause surgical delays due to untimely equipment transfer. Furthermore, the postoperative disinfection process was cumbersome, impacting the turnover efficiency of the operating rooms.

[0068] After adopting this device, our hospital sends scheduling instructions to it through the intelligent logistics system. Upon receiving the instructions, the central control module combines a 3D point cloud map constructed using SLAM positioning units and a lidar array, and employs AI algorithms to plan the optimal movement path. The device uses omnidirectional drive wheels and a Kalman filter algorithm to predict the movement trajectories of moving obstacles (such as surgical carts and medical staff), allowing it to move autonomously between operating rooms without human intervention. After completing the first femoral fracture surgery, the device automatically activates a miniature UVC disinfection lamp array and atomizing disinfector to dynamically disinfect itself and the surgical area. After disinfection, it autonomously proceeds to the operating room for the second spinal fusion surgery.

[0069] Throughout the surgical scheduling process, the device seamlessly integrated autonomous movement across operating rooms with efficient disinfection. The transfer and disinfection time for each surgical instrument was reduced by 40% compared to traditional methods, preventing surgical delays. Simultaneously, three devices could work collaboratively, serving different operating rooms and achieving equipment sharing, reducing equipment idle time. All five surgeries were completed on time that day, significantly improving operating room turnover efficiency and fully utilizing medical resources. This case demonstrates how the device effectively improves operating room turnover efficiency and optimizes medical resource allocation in multi-operating-room scheduling scenarios.

[0070] Example 4: Please refer to Figure 1 , Figure 2 , Figure 5 , Figure 7 and Figure 8 An AI-powered mobile surgical limb lifting assist device includes a mobile base 1, a rotating base 3 mounted in the middle of the mobile base 1, the lower part of the rotating base 3 fixedly mounted on the top shell of the mobile base 1, a four-degree-of-freedom robotic arm 2 mounted on the upper part of the rotating base 3, and the four-degree-of-freedom robotic arm 2 connected to the mobile base 1 via the rotating base 3; an adaptive lifting platform 4 is connected to the end of the four-degree-of-freedom robotic arm 2, a contact surface temperature and humidity sensor 34 is mounted at the center of the adaptive lifting platform 4, and 12 sets of medical silicone airbags 22 are arranged in a ring around the center of the adaptive lifting platform 4;

[0071] The mobile base 1 is equipped with a central control module 9. The central control module 9 is connected to the hospital's smart logistics system and supports HL7 protocol compatibility. The central control module 9 integrates an AI intelligent controller 29, a SLAM positioning unit 30 and a storage unit 31. The SLAM positioning unit 30 works with the lidar array 7 to build a three-dimensional point cloud map. The AI ​​intelligent controller 29 acquires the patient's body position data in real time through the HL7 protocol, automatically matches the pre-stored template in the storage unit 31 and combines AI algorithms to generate a personalized support plan. The storage unit 31 saves detailed operation procedures and generates operation logs.

[0072] Furthermore, in complex surgeries such as pelvic fractures combined with multiple soft tissue injuries, the affected limb needs to be adjusted in posture multiple times to adapt to different surgical stages. This AI intelligent mobile surgical limb lifting assistance device achieves true adaptive support through multimodal sensor fusion and AI algorithm. The storage unit 31 contains more than 2,000 orthopedic surgical templates, and the AI ​​intelligent controller 29 automatically matches the best support strategy according to the current surgical type.

[0073] During the procedure, the TOF depth camera 27 and the six-dimensional force sensor built into the ball joint 18 capture the posture and force of the affected limb in real time. Combined with the biomechanical model, the pressure of the medical silicone airbag 22 and the angle of the four-degree-of-freedom robotic arm 2 are dynamically adjusted to avoid nerve compression and blood circulation disorders. The voice system supports bilingual commands in Chinese and English and has the ability to understand context. For example, when the doctor says "higher", the system automatically makes a fine adjustment to the height based on the previous command.

[0074] In addition, the lifting assist has learning capabilities and can record each doctor's operating preferences, automatically recommending personalized settings in subsequent surgeries. During long surgeries, the PT100 contact surface temperature and humidity sensor 34 continuously monitors the temperature and humidity of the affected limb, achieving ±0.1℃ accuracy monitoring to prevent pressure sores. After surgery, the central control module 9 automatically generates a surgical report, recording data such as the number of times the affected limb posture was adjusted and the pressure distribution curve, providing data support for postoperative review and scientific research.

[0075] For example, in cases of comminuted pelvic fractures combined with multiple soft tissue injuries in the lower limbs, surgery needs to be performed in stages, including pelvic reduction and fixation, soft tissue repair, and other procedures. The position of the affected limb needs to be adjusted multiple times during the operation to adapt to the needs of different surgical steps. Traditional fixation frames cannot meet the requirements of dynamic adjustment and personalized support, which can easily lead to problems such as nerve compression and blood circulation disorders during the operation.

[0076] With this device, the central control module obtains patient positioning data, fracture type, and soft tissue injury extent from the hospital information system via the HL7 protocol. The AI ​​intelligent controller, combined with over 2000 orthopedic surgical templates stored in the storage unit, automatically matches a support plan suitable for the patient's condition and optimizes it based on real-time data. During the surgery, the TOF depth camera and the six-dimensional force sensor built into the ball joint capture the posture and force of the affected limb in real time, dynamically adjusting the angle of the four-degree-of-freedom robotic arm and the pressure of the medical silicone airbag based on a biomechanical model.

[0077] During the pelvic reduction phase, the device elevates the affected limb to a specific angle, ensuring the pelvis is in a stable, reduced position. During the soft tissue repair phase, the device slowly adjusts the height and tilt of the affected limb according to the surgical requirements, while temperature and humidity sensors on the contact surface monitor the temperature and humidity of the contact area in real time to prevent pressure sores. Furthermore, the device supports bilingual (Chinese and English) voice commands and has contextual understanding capabilities; for example, if the doctor says "fine-tune the angle of the affected limb," the system can precisely adjust it based on the previous posture. Post-operatively, the device automatically generates a surgical report, recording data such as the number of limb posture adjustments and pressure distribution curves, providing a reference for post-operative rehabilitation planning. The patient did not experience complications such as nerve compression or pressure sores post-operatively, the pelvic fracture healed well, and the soft tissue injury recovered smoothly. This case demonstrates that the device can provide personalized support solutions for complex, multi-stage surgeries, effectively improving patient prognosis.

[0078] Example 5: Please refer to Figure 1 , Figure 2 , Figure 7 and Figure 8 An AI-powered mobile surgical limb lifting assist device includes a mobile base 1, a rotating base 3 mounted in the middle of the mobile base 1, a lower part of the rotating base 3 fixedly mounted on the top shell of the mobile base 1, a four-degree-of-freedom robotic arm 2 mounted on the upper part of the rotating base 3, the four-degree-of-freedom robotic arm 2 connected to the mobile base 1 via the rotating base 3, a miniature UVC disinfection lamp array 25 embedded in the adaptive lifting platform 4 surrounding the medical silicone airbag 22, and a laser radar array 7 and an atomizing disinfector 14 arranged on the side of the mobile base 1. The atomizing disinfector 14 is linked with the miniature UVC disinfection lamp array 25 to achieve dynamic disinfection.

[0079] The bottom of the mobile base 1 is provided with an omnidirectional drive wheel set 6, which includes a built-in drive motor and a Mecanum wheel set. The middle of the mobile base 1 is provided with a power compartment 8, and an emergency physical disconnect switch 33 is provided on the surface of the power compartment 8. The power compartment 8 is provided with a lithium battery pack 10, which is connected to each component of the device through a signal line 35 to provide power for the operation of each component of the device.

[0080] Furthermore, in extreme environments such as field medical care or disaster relief, this device has undergone adaptive modifications to possess stronger environmental adaptability and anti-interference capabilities. The device casing adopts an anti-electromagnetic interference design, conforms to the MIL-STD-461G standard, and can operate stably and maintain communication in strong interference environments. The voice system has been enhanced to an offline voice recognition mode, which can still respond to commands in environments without network coverage. The built-in gesture control module recognizes gesture commands through a Leap Motion sensor and a convolutional neural network, providing an alternative interaction method to voice control in noisy environments. The lithium battery pack 10 has been upgraded to a dual-battery thermal redundancy architecture, increasing the battery life to 12 hours and supporting emergency power replenishment by solar panels. The mobile base 1 has been modified to use all-terrain tires to adapt to rugged roads, with a maximum climbing angle of 15°. In field surgery, the device achieves autonomous navigation and positioning in unstructured environments through enhanced SLAM algorithms, supporting stable movement in rugged terrain.

[0081] The disinfection module is enhanced to a high-pressure atomization and UV-C composite mode, which can achieve rapid sterilization and disinfection in environments with limited water resources. The adaptive lifting platform 4 adopts a quick disassembly and assembly design, which can complete the sterile sheath within 10 minutes. The lifting assistance device also has offline AI reasoning capabilities, which can independently complete path planning, force control adjustment and abnormal handling in environments without network, ensuring that reliable limb support services can still be provided under extreme conditions, providing intelligent equipment support for field medicine.

[0082] The lifting assistance device has a high degree of environmental adaptability and operational reliability, significantly improving the success rate of battlefield surgery and the survival rate of the wounded, and meeting the high-efficiency needs of battlefield emergency care.

[0083] For example, a large number of wounded soldiers needed to receive surgical treatment at makeshift field medical points. One of the wounded soldiers suffered severe crush injuries to his lower limbs due to a building collapse and required emergency amputation. The field medical environment was harsh, with problems such as strong electromagnetic interference, lack of stable networks, and rugged terrain. Traditional surgical equipment could not function properly, and sterilization conditions were limited, resulting in a high risk of infection.

[0084] After being adapted and activated, this device's electromagnetic interference-resistant casing (compliant with MIL-STD-461G standards) ensures stable operation even in environments with strong interference. Its offline voice recognition mode accurately responds to doctors' commands to "activate amputation surgery support mode" even without a network connection. In noisy environments, doctors can adjust the device's status via gesture commands using the built-in Leap Motion gesture control module. Equipped with all-terrain tires, the device easily adapts to rough terrain, achieving a maximum climbing angle of 15°, ensuring a smooth arrival at the rescue point.

[0085] During the surgery, the dual-battery thermal redundancy lithium battery pack provides stable power, with a battery life of up to 12 hours, ensuring no power outage risk throughout the procedure. The adaptive lifting platform features a quick-assembly and disassembly design, allowing for sterile sheath replacement within 10 minutes, meeting the sterility requirements of field environments. High-pressure atomization combined with UV-C disinfection rapidly disinfects the device and surgical area even with limited water resources, reducing the risk of infection. The four-degree-of-freedom robotic arm precisely lifts the affected limb, providing stable support for amputation surgery and ensuring its successful completion. This case demonstrates the device's strong environmental adaptability in extreme field medical scenarios, ensuring medical service needs in special situations and saving patients' lives.

[0086] Working Principle: The workflow of this device begins with the voice wake-up stage. After medical staff issue commands through pre-recorded voiceprints, identity verification and command parsing are performed first. Then, the autonomous navigation function is activated. The environment is scanned and a 3D map is constructed through the LiDAR array 7. Combined with the SLAM positioning unit 30, the path is planned in real time. The central control module 9 controls the mobile base 1 to reach the designated position on the operating table. The four-degree-of-freedom robotic arm 2 automatically unfolds according to the patient's body position parameters. The adaptive lifting platform 4 achieves millimeter-level precise lifting of the affected limb through the pressure sensor 23 and force feedback mechanism. During the operation, the device supports real-time voice adjustment of posture and continuous monitoring of pressure distribution and stability. Once an abnormality is detected, a three-level safety alarm is immediately triggered and the mechanism is locked. After the operation, the device automatically folds and stores, activates the dual disinfection program of the micro UVC disinfection lamp array 25 and the atomizing disinfector 14, and returns to the charging position to stand by, completing the fully automated operation.

[0087] Technical terms and their specific meanings in the description of this invention:

[0088] 1. Four-DOF Robotic Arm: The robotic arm possesses four independent degrees of freedom, enabling flexible movement in multiple directions within space. In this patent, the four-DOF robotic arm comprises a large robotic arm, a small robotic arm, ball joints, a telescopic slide rail, a servo motor, and hydraulic cylinders. The servo motor drives the rotating base and the movement of each part of the robotic arm. The hydraulic cylinders, in conjunction with the telescopic slide rail, enable the extension and retraction of the robotic arm, allowing for precise adjustment of the affected limb's position and posture.

[0089] 2. Force-Position Hybrid Control Algorithm: This algorithm comprehensively considers position control and force control, aiming to achieve precise control of the robotic arm's movement. It allows the arm to meet position accuracy requirements while adjusting the output force in real time according to external forces. In this patent, the central control module uses the force-position hybrid control algorithm to achieve millimeter-level position adjustment. It is based on the formula... (in Δθ is the angular deviation. Adjust the support torque (for the rate of change).

[0090] 3. AR Projection Positioning Component: This component, consisting of a DLP projection module, a TOF depth camera, and a dynamic calibration target, is used for intraoperative positioning visualization guidance. The DLP projection module is coated with a medical-grade anti-glare coating, allowing surgical planning information and positioning markers to be projected directly onto the surface of the affected limb. The TOF depth camera acquires real-time 3D point cloud data of the affected limb to calibrate the projection position, ensuring that the projected content accurately corresponds to the actual anatomical structure of the affected limb. The dynamic calibration target further ensures the accuracy of the alignment between the projected information and the affected limb through computer vision algorithms.

[0091] 4. SLAM Positioning Unit: This unit, also known as the Simultaneous Localization and Mapping (SLAM) unit, collects real-time environmental data using sensors such as LiDAR arrays to construct a 3D point cloud map of the operating room. Combined with its own motion information, it enables real-time positioning and navigation within the operating room environment. Relying on the SLAM positioning unit and omnidirectional drive wheels, the device can move autonomously within the operating room, avoiding obstacles in real-time, and proceeding to the required surgical location. For example, in multi-operating-room scheduling scenarios, the device can use the SLAM positioning unit to plan the optimal movement path based on instructions from the hospital's intelligent logistics system, efficiently transferring between different operating rooms and avoiding collisions with surgical equipment, medical personnel, and other obstacles. Compared to traditional positioning methods based on preset tracks or marker points, the SLAM positioning unit gives the device stronger environmental adaptability and autonomous movement capabilities, enabling it to better adapt to the complex and ever-changing operating room environment.

[0092] 5. Medical-grade silicone airbags: These airbag structures, made of medical-grade silicone, are installed on the adaptive lifting platform. Each airbag contains a pressure sensor and an LED indicator, and its pressure can be independently controlled. During surgery, the pressure sensors monitor the pressure data of each airbag in real time, and the LED indicators display the pressure area, providing visualization of the pressure distribution. The central control module dynamically adjusts the pressure of each medical-grade silicone airbag based on the data from the pressure sensors, ensuring even force distribution on the affected limb during lifting and avoiding excessive local pressure that could lead to nerve compression, poor blood circulation, or other intraoperative complications. Unlike ordinary inflatable support pads, medical-grade silicone airbags not only flexibly adjust pressure distribution but also, due to their medical-grade silicone material, have excellent biocompatibility, ensuring comfortable contact with the patient's skin and reducing the risk of skin irritation and damage.

[0093] 6. HL7 Protocol: Health Level 7 is a standard data exchange protocol used in the medical field, designed to enable data transmission and sharing between different medical information systems. The device's central control module uses the HL7 protocol to obtain patient positioning data, fracture type, and soft tissue injury extent from the hospital information system. This allows the AI ​​intelligent controller to generate personalized support plans based on built-in surgical templates. For example, in surgery for a pelvic fracture with multiple soft tissue injuries, the device can use the HL7 protocol to obtain comprehensive patient information, providing more tailored support services for complex surgeries. Compared to other non-standardized data transmission methods, the HL7 protocol ensures the accuracy, standardization, and compatibility of medical data transmission between different systems, promoting deep integration of medical devices and hospital information systems, and improving the informatization and intelligentization of medical services.

[0094] 7. Voiceprint Activation and Command Parsing: This function activates the device by recognizing the user's voiceprint characteristics and parses and executes the user's voice commands. Medical staff can activate the device using pre-registered voiceprint features. After successful verification, the device's voice recognition system uses natural language processing technology to parse the voice commands and control the device to perform corresponding operations, such as controlling the four-degree-of-freedom robotic arm to lift the affected limb or adjust the posture of the affected limb. Compared with traditional manual button operation, the voiceprint activation and command parsing function eliminates the need for medical staff to manually touch the device during surgery, freeing their hands and making operation more convenient and efficient. This is especially suitable for scenarios where medical staff's hands are busy during surgery and they need to quickly adjust the device's status, while also reducing the risk of contamination on the device surface caused by frequent manual operation.

[0095] 8. Miniature UVC Disinfection Lamp Array and Atomizing Sterilizer Linked System: A dynamic disinfection method combining a miniature UVC disinfection lamp array and an atomizing sterilizer. The miniature UVC disinfection lamp array uses ultraviolet light to kill bacteria, viruses, and other microorganisms, while the atomizing sterilizer atomizes the disinfectant and diffuses it into the air and onto object surfaces, further enhancing the disinfection effect. Before and after surgery, and during the movement of the device between operating rooms, the miniature UVC disinfection lamp array and the atomizing sterilizer can be activated in conjunction to disinfect the device itself and the operating room environment. Compared to single disinfection methods, this linked disinfection mode has a wider coverage, more comprehensive and efficient disinfection, complies with the "Technical Specifications for Disinfection in Medical Institutions," effectively reduces the number of bacteria and viruses in the operating room, reduces the risk of surgical infection, and protects the health and safety of patients and medical staff.

[0096] 9. Leap Motion Sensor: A sensor for detecting hand movements and postures, it captures hand position and gesture information in real time by emitting and receiving infrared light. In the adaptive modification of this patented device, when in noisy environments such as field medical settings where voice commands are difficult to recognize accurately, doctors can use the Leap Motion sensor and its built-in convolutional neural network to recognize gesture commands, thereby controlling the device and adjusting the height and angle of the affected limb. Compared with traditional input devices such as mice and keyboards, the Leap Motion sensor provides a more natural and intuitive human-computer interaction method, especially suitable for the need for rapid control of devices in special environments, thus broadening the device's operation methods and application scenarios.

[0097] 10. Dual-Battery Thermal Redundancy Architecture: This architecture employs two batteries to power the device simultaneously. If one battery fails, the other seamlessly takes over, ensuring continuous and stable operation. In extreme environments such as field medical settings, the dual-battery thermal redundancy architecture provides stable power to the device, offering up to 12 hours of continuous operation. This design significantly improves power reliability in complex environments, preventing surgical interruptions due to battery failure and ensuring the continuity and stability of medical services in special scenarios. Compared to traditional single-battery power supply methods, it offers higher fault tolerance and reliability.

[0098] 11. Six-Dimensional Force Sensor: A high-precision sensor capable of simultaneously detecting force and torque information in six dimensions of space (force in the X, Y, and Z axes and torque around the three axes). In this patent, it is integrated into a ball joint and can capture real-time force changes when the affected limb comes into contact with the robotic arm, including minute forces generated by muscle tremors and surgical manipulations. For example, in pelvic fracture surgery, when the surgeon adjusts the reduction force, the six-dimensional force sensor can transmit the force data to the central control module in real time. Combined with a biomechanical model, it calculates the safe force threshold, preventing soft tissue damage to the affected limb due to excessive support force or displacement due to insufficient support force. It is the core sensing component for achieving "compliant support." It is important to distinguish this from ordinary unidirectional force sensors, which can only detect force in a single direction and cannot meet the monitoring needs of complex force scenarios during surgery.

[0099] 12. TOF Depth Camera: A 3D imaging device based on the "Time of Flight" principle. It emits modulated near-infrared light, measures the time difference between the emitted light signal and its return after reflection from the object, calculates the distance between the object and the camera, and generates 3D point cloud data. In the AR projection positioning component of this patent, the TOF depth camera can scan the surface contour of the affected limb in real time, dynamically correcting the projection angle and position of the DLP projection module. When the patient's chest rises and falls due to breathing, causing slight displacement of the affected limb, the TOF camera can capture the displacement change within 50ms, triggering dynamic calibration of the target to adjust the projection content, ensuring that the surgical marker is always precisely aligned with the anatomical structure. Unlike traditional structured light cameras, the TOF depth camera has stronger resistance to ambient light interference, can work stably even under strong operating room lighting, and does not require preset texture patterns, making it suitable for low-texture scenes such as skin surfaces.

[0100] 13. Convolutional Neural Network (CNN): A deep learning model adept at processing grid-like data (such as images and point clouds). It extracts local features from data through multiple layers of convolutional kernels, gradually constructing global features to achieve tasks such as classification and recognition. In this patent, the CNN is mainly used for two core functions: First, in the Leap Motion gesture control module, it extracts features from the captured key point data of the hand skeleton to recognize preset gestures such as "raising the affected limb" and "fine-tuning the angle," achieving an accuracy rate of over 98.5% and filtering out invalid actions caused by hand tremors. Second, in the SLAM localization unit, it performs semantic segmentation on the environmental point cloud data collected by the LiDAR to distinguish different obstacle types such as "operating table," "medical staff," and "instrument cart," providing more accurate environmental information for path planning. It is important to distinguish this from Recurrent Neural Networks (RNNs), which are more suitable for temporal data processing, while CNNs have significant advantages in spatial feature extraction and are more suitable for spatially related tasks such as real-time localization and gesture recognition during surgery.

[0101] 14. Omnidirectional Drive Wheelset (Mecanum Wheelset): A wheel system composed of Mecanum wheels, where the roller axis of each wheel forms a 45° angle with the wheel axis. By controlling the forward and reverse rotation combinations of the four wheels, the device can be moved and rotated in any direction without turning to adjust its position. In this patent, the omnidirectional drive wheelset, in conjunction with the SLAM positioning unit, enables the device to move flexibly in the narrow spaces of the operating room (e.g., the distance between the operating table and the instrument cart is only 80cm). For example, when moving from the left side of the operating table to the right side, there is no need to back up to adjust the direction; it can move directly laterally to the desired position, avoiding collisions with surrounding equipment. Compared with traditional differential drive wheelsets, the omnidirectional drive wheelset eliminates the need for a "turning radius," significantly improving the device's mobility in complex environments, especially suitable for operating room scenarios with multiple devices and dense personnel. It should be noted that "omnidirectional movement" is not a simple upgrade of "universal wheels." Universal wheels can only adjust the direction of movement with steering, while the Mecanum wheelset can achieve true "lateral translation," resulting in significantly greater operational flexibility.

[0102] 15. Biomechanical Model: A mathematical model constructed based on human anatomy and mechanical principles to simulate the mechanical response of human tissues (bones, muscles, blood vessels) under force and movement. In this patent, the central control module incorporates multiple biomechanical models for different body parts (upper limbs, lower limbs, pelvis). For example, the lower limb model combines parameters such as patient weight, lower limb length, and bone density to calculate the safe force range for each joint when the affected limb is lifted. When the six-dimensional force sensor detects that the force at the knee joint exceeds the model's preset "safety threshold" (e.g., the maximum supporting force of an adult knee joint does not exceed 1.2 times the body weight), the system automatically reduces the robotic arm's supporting force to avoid joint damage due to excessive force. Simultaneously, when adjusting the posture of the affected limb, the biomechanical model can also predict the blood circulation status at different angles. For example, controlling the lower limb lifting angle between 15° and 30° satisfies the surgical field requirements while avoiding obstruction of venous return due to excessive angle. This model differs from the ordinary "force threshold setting," which is based solely on empirical values. The biomechanical model, on the other hand, dynamically calculates based on individual patient parameters, making it more personalized and scientific.

[0103] 16. Sterile Sheath Interface (Detachable): A dedicated interface for the device's contact area with the patient (adaptive lifting platform), used to secure a disposable sterile sheath, ensuring a tight, seamless, and stable fit between the sheath and the platform, facilitating quick replacement. The sterile sheath interface in this patent employs a "buckle + sealing strip" design, allowing medical personnel to install the sheath in just 15 seconds. After installation, a pressure sensor detects the sheath's seal—if a gap exists causing pressure leakage, an LED indicator will issue a warning. It is important to note the difference between a "sterile sheath" and a "regular dust cover": a sterile sheath must meet medical sterility standards (such as ISO 10993 biocompatibility requirements), withstand ethylene oxide sterilization, and have a smooth surface without particle shedding to prevent intraoperative contamination; while a regular dust cover only provides dust protection and cannot meet the sterility requirements of the operating room, potentially leading to infection risks if misused. In addition, the core design principle of the "detachable interface" is "quick replacement + sterility guarantee", rather than simply "detachable". The focus should be on its sterile sealing and ease of operation.

[0104] 17. Artificial Potential Field Method: An algorithm for path planning that treats the device as a "point mass," the target location as a "source of attraction," and obstacles as "sources of repulsion." By calculating the direction of the resultant force of attraction and repulsion, the algorithm guides the device towards the target location while avoiding obstacles. In the SLAM positioning unit of this patent, the artificial potential field method is used for dynamic path planning—for example, when the device moves from the sterilization area to the operating table, the operating table generates "attraction," while the instrument cart and medical staff generate "repulsion." The algorithm calculates the direction of the resultant force in real time, causing the device to automatically decelerate and fine-tune its direction when approaching medical staff, maintaining a safe distance of more than 30cm. Compared with the traditional A* algorithm, the artificial potential field method does not require pre-generating a complete path, can dynamically adjust according to the real-time obstacle position, has a faster response speed (adjustment delay <100ms), and is more suitable for the dynamic environment of frequent movement of personnel and equipment in the operating room. It should be noted that the algorithm has a "local optimum" problem (such as the inability to find a path when the device is "sandwiched" between two obstacles). This patent can effectively avoid this defect by combining the global map information of SLAM and ensure the global optimum of path planning.

[0105] 18. Offline Speech Recognition Mode: This mode allows for speech command parsing directly on the device without relying on a network connection. In the field medical scenario modification of this patent, the offline speech recognition mode stores the speech feature models of preset commands (such as "activate amputation mode" and "increase airbag pressure") in the local storage unit, enabling recognition without connecting to a cloud server. The recognition process consists of three steps: ① A ring microphone array collects the speech signal, filtering environmental noise using a noise suppression algorithm; ② Mel-frequency cepstral coefficients (MFCC) features of the speech are extracted; ③ The signal is compared with the locally stored feature model to determine the command content. The entire process takes less than 500ms. Compared to online speech recognition, the offline mode is not limited by network signals and avoids the privacy risks of medical data leakage. Furthermore, in noisy environments (such as generator noise and patient shouts at field medical sites), the recognition accuracy can still be maintained above 95% through targeted noise training of the model. It should be noted that "offline mode" does not mean "no training updates." The device can periodically receive command model updates via the hospital's intranet to ensure the integrity of the recognized commands.

[0106] 19. Emergency Physical Joystick (IP67 Protection Rating): A physical operating component for manual control of the device in case of voice system failure, equipped with an IP67 protection rating (completely dustproof and can be submerged in 1 meter of water for 30 minutes without damage). The emergency physical joystick is located on the side of the mobile base and supports manual control of the robotic arm's lifting, translation, and rotation. For example, when strong electromagnetic interference in the operating room causes the voice recognition system to malfunction, medical staff can manually adjust the patient's posture using the joystick to ensure the surgery can proceed normally. The IP67 protection rating allows it to withstand potential liquid splashes (such as disinfectants and saline solution) and dust contamination during surgery, preventing joystick malfunction due to environmental factors. It is important to note that the "emergency joystick" is not a "regular operating component." Its design purpose is to provide a redundant control scheme to ensure surgical safety, forming a "dual guarantee" with voice control, rather than replacing voice control as the primary operating method.

[0107] 20. High-Pressure Atomization and UV-C Composite Disinfection Mode: This is a synergistic disinfection method combining high-pressure atomization disinfection and UV-C ultraviolet disinfection. High-pressure atomization uses a high-pressure pump to atomize disinfectant (such as chlorine-containing disinfectant) into 5-10μm microparticles, evenly covering the object surface and air. UV-C ultraviolet disinfection uses ultraviolet light with a wavelength of 200-280nm to destroy the DNA structure of microorganisms, killing bacteria and viruses. In field medical transformation, this composite disinfection mode can quickly achieve disinfection even with limited water resources. For example, at earthquake rescue sites, only 50ml of disinfectant is needed to cover the surface of the device through high-pressure atomization, combined with 10 minutes of UV-C lamp irradiation, achieving a disinfection pass rate of over 99.9%. Compared with single disinfection modes, the composite disinfection mode has both "surface disinfection" and "air disinfection" capabilities, with shorter disinfection time and less disinfectant usage, making it suitable for resource-limited scenarios in field environments. It should be noted that "UV-C disinfection" should avoid direct exposure to human skin and eyes. When the device is activated in this mode, it will automatically detect the surrounding environment and pause disinfection if anyone approaches to ensure safe use.

[0108] 21. Mel-frequency cepstral coefficients (MFCC): A parameter used to extract speech features. It simulates the human ear's perception of sound frequencies, converting the speech signal from the time domain to the Mel-frequency domain and obtaining cepstral coefficients through discrete cosine transform for speech recognition. In offline speech recognition mode, MFCC is used to extract features from medical personnel's voice commands. For example, for the voice signal "Activate upper limb support mode," MFCC processing yields a 39-dimensional feature vector, which is compared with the locally stored feature model to achieve command recognition. MFCC effectively highlights key speech features and filters background noise, maintaining high feature extraction accuracy even in noisy environments, making it one of the core technologies ensuring the accuracy of offline speech recognition. It's important to note that MFCC is not "direct sampling of the speech signal," but rather feature compression and optimization of the speech signal. It can reduce data volume while retaining key recognition information, adapting to scenarios with limited local storage and computing resources.

[0109] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered in all respects as exemplary and non-limiting, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. An AI-powered intelligent mobile surgical limb lifting assist device, characterized in that: The system includes a movable base (1), a rotating base (3) mounted in the middle of the movable base (1), the lower part of the rotating base (3) being fixedly mounted on the top shell of the movable base (1), and a four-degree-of-freedom robotic arm (2) mounted on the upper part of the rotating base (3). The four-degree-of-freedom robotic arm (2) is connected to the movable base (1) through the rotating base (3). The four-degree-of-freedom robotic arm (2) consists of a large robotic arm (16), a small robotic arm (17), a ball joint (18), and a telescopic slide rail (19). The mechanical arm (16) consists of a servo motor (20) and a hydraulic cylinder (21). One end of the mechanical arm (16) passes through the upper surface of the rotating base (3) and is equipped with a hydraulic cylinder (21) at the end. The other end of the mechanical arm (16) is connected to the mechanical arm (17) through a ball joint (18). The telescopic slide rail (19) is opened on the side of the mechanical arm (16). The servo motor (20) is installed in the lower part of the rotating base (3) to drive the rotation of the rotating base (3) and the movement of the four-degree-of-freedom mechanical arm (2).

2. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 1, characterized in that: The end of the four-degree-of-freedom robotic arm (2) is connected to an adaptive lifting platform (4). The adaptive lifting platform (4) is equipped with a detachable sterile sheath interface. A contact surface temperature and humidity sensor (34) is installed at the center of the adaptive lifting platform (4). Twelve medical silicone airbags (22) are arranged in a ring around the center of the adaptive lifting platform (4). Each medical silicone airbag (22) is equipped with a pressure sensor (23) and an LED indicator (24). Each medical silicone airbag (22) is independently pressure controlled and the pressure area is displayed by the LED indicator (24). A miniature UVC disinfection lamp array (25) is embedded on the adaptive lifting platform (4) around the medical silicone airbags (22).

3. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 2, characterized in that: An AR projection positioning component (5) is installed at one end of the adaptive lifting platform (4). The AR projection positioning component (5) consists of a DLP projection module (26), a TOF depth camera (27), and a dynamic calibration target (28). The DLP projection module (26), coated with a medical anti-glare coating, projects surgical planning information and positioning marks directly onto the surface of the affected limb. The TOF depth camera (27) acquires three-dimensional point cloud data of the affected limb in real time to calibrate the projection position. The dynamic calibration target (28) uses computer vision algorithms to ensure that the projection content is accurately aligned with the anatomical structure of the affected limb, accurately projecting virtual information onto the physical entity to achieve visual guidance for intraoperative positioning.

4. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 1, characterized in that: The side of the mobile base (1) is equipped with a lidar array (7) and an atomizing sterilizer (14). The atomizing sterilizer (14) is linked with the miniature UVC disinfection lamp array (25) to achieve dynamic disinfection. The upper surface of the mobile base (1) is equipped with an OLED touch screen (11), a ring microphone array (12), and an emergency physical joystick (13). The OLED touch screen (11) with an operating room-specific anti-fog panel is used to provide a graphical operation interface and real-time status feedback. The ring microphone array (12) with 8-channel noise suppression is used for 360° sound pickup and suppresses environmental noise based on beamforming technology. The emergency physical joystick (13) with IP67 protection level is used for manual redundant control when the voice system fails. The OLED touch screen (11), the ring microphone array (12), and the emergency physical joystick (13) together form a human-computer interaction unit.

5. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 1, characterized in that: The mobile base (1) is equipped with a central control module (9). The central control module (9) is connected to the hospital's smart logistics system and supports HL7 protocol compatibility. The central control module (9) integrates an AI intelligent controller (29), a SLAM positioning unit (30), and a storage unit (31). The SLAM positioning unit (30) works with the lidar array (7) to build a three-dimensional point cloud map. The AI ​​intelligent controller (29) obtains the patient's position data in real time through the HL7 protocol, automatically matches the pre-stored template in the storage unit (31), and generates a personalized support solution in combination with AI algorithms. The storage unit (31) saves detailed operation procedures and generates operation logs.

6. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 2, characterized in that: The upper surface of the tail of the mobile base (1) is provided with a speaker (15) and an alarm light (32). The voice alarm of the speaker (15) and the optical warning and vibration reminder of the alarm light (32) constitute the three-level safety early warning mechanism of the device.

7. The AI-powered intelligent mobile surgical limb lifting assist device according to claim 1, characterized in that: The bottom of the mobile base (1) is provided with an omnidirectional drive wheel set (6), which includes a built-in drive motor and a Mecanum wheel set. The center of the mobile base (1) is provided with a power compartment (8), and the surface of the power compartment (8) is provided with an emergency physical disconnect switch (33). The power compartment (8) is provided with a lithium battery pack (10), which is connected to each component of the device through a signal line (35) to provide power for the operation of each component of the device.

8. A control method for an AI-powered intelligent mobile surgical limb lifting assist device, applicable to the AI-powered intelligent mobile surgical limb lifting assist device as described in any one of claims 1-7, characterized in that: The control method includes the following steps: S1. Voiceprint Activation and Command Parsing: The device is activated by the voiceprint features of pre-registered medical staff, and the voice commands are parsed after verification. S2. Environmental perception and modeling: A three-dimensional point cloud map of the operating room is constructed using a lidar array (7), and obstacle information is updated in real time; S3. Dynamic Path Planning and Navigation: Combining AI algorithms and artificial potential field methods, a collision-free path is planned, and the control device enables autonomous movement and real-time obstacle avoidance to reach the target location. S4, Force-position hybrid precision control: Adjust the supporting torque of the four-degree-of-freedom robotic arm (2) to complete the precise lifting and posture adjustment of the affected limb, and realize the millimeter-level precise positioning and adaptive support of the affected limb; S5. Safety monitoring and emergency response: Real-time monitoring of data acquired by pressure sensor (23). When an abnormal situation occurs, the device’s three-level safety warning mechanism is automatically triggered, and the four-degree-of-freedom robotic arm (2) is immediately locked.

9. The control method for the AI ​​intelligent mobile surgical limb lifting assist device according to claim 8, characterized in that: In S3, a Kalman filter algorithm is used to predict the movement trajectory of moving obstacles, and an electronic fence is set up in the sterile area of ​​the surgery to maintain a minimum safe distance of 200mm.

10. The control method for the AI ​​intelligent mobile surgical limb lifting assist device according to claim 8, characterized in that: Specifically, S4 is: S4.1 Setting targets and continuous monitoring: The system sets the target position and posture of the robotic arm according to voice commands; S4.2 Calculate the deviation between the target angle and the actual angle and its rate of change, and calculate the precise torque to be applied in real time; S4.3 Output execution and dynamic adjustment: The central control module (9) controls the movement of the four-degree-of-freedom robotic arm (2) according to the calculated torque value to eliminate angular deviation. The central control module 9 achieves millimeter-level position adjustment through the force-position hybrid control algorithm. The force-position hybrid control is based on the formula F=Kp×Δθ+Kd×dΔθ / dt to adjust the support torque, where Kp and Kd are preset parameters, Kp=2.3N·m / rad, Kd=0.15N·m·s / rad, Δθ is the angular deviation, and dΔθ / dt is the rate of change. According to the real-time feedback of the affected limb, the output torque is continuously fine-tuned to achieve adaptive compliant support.

Images