Heart valve operation recommendation system and method based on robot assistance

Abstract

The invention provides a heart valve surgery recommendation system and method.The system comprises a robot operation platform, a multi-mode sensing module, an intelligent control and decision module and a man-machine interaction interface, and the robot operation platform comprises a robot arm and an end effector; the multi-modal sensing module comprises a touch sensor and a medical image system, the touch sensor is installed at the tail end of a robot arm, and the intelligent control and decision module comprises a data processing module, a Gaussian process model module, a kernel function fusion module, a Bayesian optimization module, an energy constraint optimization module and an incision recommendation module. According to the robot-assisted heart valve operation recommendation system and method, the operation precision can be improved, the operation risk can be reduced, and a surgeon can be assisted to make a decision.

Description

Technical Field

[0001] The present invention relates to the technical field of medical assistance robots, and specifically, to a robot-assisted cardiac valve surgery recommendation system and method. Background Art

[0002] Cardiac valve disease is a common cardiovascular disease worldwide, seriously threatening human health. Cardiac valve replacement or repair is an effective means of treating cardiac valve disease. Traditional cardiac valve surgery mainly relies on the experience and palpation skills of surgeons, and is performed through open-chest or minimally invasive methods. However, this traditional surgical method has many limitations:

[0003] 1. Limited surgical field of view: Traditional open-chest surgery has large trauma and limited exposure of the surgical field, especially for the observation and operation of deep structures. Although minimally invasive surgery has less trauma, the operating space is even smaller, and higher requirements are placed on the operating skills and experience of doctors.

[0004] 2. Insufficient operating accuracy: The cardiac valve structure is delicate and complex. Traditional surgery mainly relies on the doctor's hand feeling and experience for positioning and operation, which is prone to errors, increasing the surgical risk and the incidence of complications.

[0005] 3. Long surgical time: The operation steps of traditional surgery are cumbersome and require a long time, increasing the intraoperative risk of patients and the postoperative recovery time.

[0006] 4. Steep learning curve: Traditional cardiac valve surgery has extremely high requirements for the experience and skills of surgeons. It takes a long time of training and practice to master proficiently, resulting in a steep learning curve and being difficult to popularize.

[0007] In recent years, with the development of robot-assisted surgery systems, new possibilities have been provided for improving the accuracy and efficiency of cardiac valve surgery. Robot-assisted surgery systems have advantages such as three-dimensional high-definition vision, flexible robotic arms, and precise operation control, and can overcome many limitations of traditional surgery. However, the existing robot-assisted cardiac valve surgery systems still have the following deficiencies:

[0008] 1. Lack of precise positioning means: The existing robot-assisted surgery systems mainly rely on preoperative imaging data for navigation, lacking real-time and precise intraoperative positioning means, and it is difficult to ensure the accuracy of surgical operations.

[0009] 2. Lack of force feedback mechanism: The robot-assisted surgery system lacks a force feedback mechanism, and doctors cannot sense the interaction force between the surgical instrument and the tissue, which is prone to causing tissue damage.

[0010] 3. High cost: The cost of the robot-assisted surgery system is high, restricting its wide clinical application.

[0011] Therefore, there is an urgent need to develop a new robot-assisted cardiac valve surgery recommendation system to improve surgical accuracy, shorten the operation time, reduce surgical risks, and promote the popularization of cardiac valve surgery. Summary of the Invention

[0012] Aiming at the defects in the prior art, the purpose of the present invention is to provide a robot-assisted cardiac valve surgery recommendation system and method to improve surgical accuracy, reduce surgical risks, and assist surgeons in making decisions.

[0013] To solve the above problems, the technical solution of the present invention is as follows:

[0014] A robot-assisted cardiac valve surgery recommendation system includes a robot operation platform, a multi-modal perception module, an intelligent control and decision-making module, and a human-computer interaction interface. The robot operation platform includes a robot arm and an end effector. The multi-modal perception module includes a tactile sensor and a medical imaging system. The tactile sensor is installed at the end of the robot arm. The intelligent control and decision-making module includes a data processing module, a Gaussian process model module, a kernel function fusion module, a Bayesian optimization module, an energy constraint optimization module, and an incision recommendation module. The data processing module is responsible for collecting, filtering, and preprocessing the data of the tactile sensor, and registering and fusing it with the medical imaging data. The Gaussian process model module is used to construct a hardness distribution map of the cardiac surface tissue based on the Gaussian process regression algorithm to quantitatively describe the characteristics of the cardiac tissue. The kernel function fusion module is used to optimize the kernel function parameters of the Gaussian process model. The Bayesian optimization module, based on the Bayesian optimization algorithm, intelligently selects the next optimal tactile sensor sampling point to obtain the optimal hardness distribution map with the least number of samplings. The energy constraint optimization module is used to optimize the movement path of the robot arm. The incision recommendation module recommends the best incision position according to the constructed hardness distribution map.

[0015] Preferably, the robot arm includes at least one robot arm with high precision and high flexibility, which is used to control the movement and positioning of the tactile sensor and perform surgical operations. The end effector is installed at the end of the robot arm and is used to fix and operate the tactile sensor.

[0016] Preferably, the medical imaging system is used to integrate the patient's CT or MRI images obtained before the operation, provide three-dimensional anatomical structure information of the heart, and fuse it with the tactile sensor data.

[0017] Preferably, the incision recommendation module, according to the constructed hardness distribution map, uses image segmentation and edge detection algorithms to identify the soft tissue area around the cardiac valve and recommends the best incision position.

[0018] Preferably, the human-machine interaction interface includes a display module and a control module. The display module is used to display tactile sensor data, hardness distribution maps, and surgical path planning in real time, providing intuitive visual feedback to surgeons. The control module is used to allow surgeons to control the movement of the robotic arm through a joystick and gesture recognition, and to adjust system parameters.

[0019] Furthermore, the present invention also provides a method for recommending robotic-assisted heart valve surgery, including the following steps:

[0020] Import the patient's preoperative CT or MRI images into the system, and the system automatically identifies the heart valve area;

[0021] Control the robotic arm to drive the tactile sensor to perform a preliminary scan of the heart valve area to obtain initial hardness data;

[0022] Perform data processing on the preliminary scan data based on the Gaussian process model to generate an initial hardness distribution map;

[0023] Based on the Bayesian optimization algorithm, automatically select the next optimal scan point and control the robotic arm to perform a fine scan;

[0024] Based on the final hardness distribution map, identify the soft tissue area around the heart valve suitable for incision. The system recommends multiple optimal incision positions to the surgeon and provides an analysis of the advantages and disadvantages of each position;

[0025] The system plans the optimal surgical path according to the final incision position selected by the surgeon and performs robotic-assisted heart valve surgery operations.

[0026] Preferably, the step of identifying the soft tissue area around the heart valve suitable for incision based on the final hardness distribution map, where the system recommends multiple optimal incision positions to the surgeon and provides an analysis of the advantages and disadvantages of each position, specifically includes: using an incision recommendation module to analyze the final hardness distribution map, identifying the soft tissue area around the heart valve, marking potential incision positions, sorting the potential incision positions, recommending the top three incision positions to the surgeon, and providing an analysis of the advantages and disadvantages of each position.

[0027] Preferably, the step of the system planning the optimal surgical path according to the final incision position selected by the surgeon and performing robotic-assisted heart valve surgery operations specifically further includes: during the entire surgical process, the system continuously monitors the hardness change of the tissue around the incision and alarms in time if any abnormality is found; the system records the hardness data and incision selection basis during the entire surgical process for subsequent analysis and system optimization.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0029] 1. Through the high-precision tactile sensor and optimized scanning strategy, the present invention can accurately locate the soft tissue area around the heart valve, improving the accuracy of the incision position.

[0030] 2. The system of the present invention automatically recommends the best incision position and surgical path, reducing the dependence on the experience of surgeons, improving the standardization of the surgery, and facilitating the training and skill improvement of novice doctors.

[0031] 3. Through the energy constraint optimization algorithm, the present invention optimizes the movement path of the robotic arm, reducing energy consumption and improving system efficiency.

[0032] 4. The present invention provides surgeons with a visual hardness distribution map and incision suggestions to assist in decision-making, improving surgical efficiency and safety.

[0033] 5. Through accurate positioning and optimized path planning, the present invention can shorten the operation time, reducing the pain and complication risk of patients.

[0034] 6. By accurately positioning the best incision position, the present invention can minimize surgical trauma, significantly improve the success rate of heart valve surgery, and accelerate the postoperative recovery of patients. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Other features, objectives, and advantages of the present invention will become more apparent by reading the detailed description of the non-limiting embodiments with reference to the following drawings:

[0036] Figure 1 It is a module architecture diagram of a robot-assisted heart valve surgery recommendation system of the present invention;

[0037] Figure 2 It is a flowchart of a robot-assisted heart valve surgery recommendation method of the present invention; DETAILED DESCRIPTION OF THE INVENTION

[0038] The present invention will be described in detail below in conjunction with specific embodiments. The following embodiments will help those skilled in the art to further understand the present invention, but do not limit the present invention in any form. It should be noted that those of ordinary skill in the art can make several changes and improvements without departing from the concept of the present invention. These all fall within the protection scope of the present invention.

[0039] Specifically, the present invention provides a robot-assisted heart valve surgery recommendation system, as Figure 1 shown, the system includes a robotic operation platform, a multi-modal perception module, an intelligent control and decision-making module, and a human-machine interaction interface.

[0040] The robot operation platform includes a robot arm and an end effector. The robot arm includes at least one robot arm with high precision and high flexibility, which is used to control the movement and positioning of the tactile sensor and perform surgical operations. The robot arm should have sufficient degrees of freedom to cover the surface of the heart and be able to perform delicate operations in a narrow space. The end effector is installed at the end of the robot arm, which is used to fix and operate the tactile sensor and can replace other surgical tools as needed, such as a needle holder, scissors, etc.

[0041] The multi-modal perception module includes a tactile sensor and a medical imaging system. The tactile sensor is installed at the end of the robot arm and is used to measure the mechanical properties of the heart surface tissue in real time, such as hardness and elasticity. The tactile sensor should have high sensitivity, high resolution and good biocompatibility.

[0042] The medical imaging system integrates the CT or MRI images of the patient obtained before surgery, provides three-dimensional anatomical structure information of the heart, and fuses with the tactile sensor data to achieve more accurate positioning and navigation.

[0043] The intelligent control and decision-making module includes a data processing module, a Gaussian process model module, a kernel function fusion module, a Bayesian optimization module, an energy constraint optimization module, and an incision recommendation module.

[0044] The data processing module is responsible for collecting, filtering and preprocessing the data of the tactile sensor, and registering and fusing it with the medical imaging data.

[0045] The Gaussian process model module is used to construct a hardness distribution map of the heart surface tissue based on the Gaussian process regression algorithm, and realize the quantitative description of the heart tissue characteristics.

[0046] The kernel function fusion module adopts the multi-kernel learning method to optimize the kernel function parameters of the Gaussian process model and improve the prediction accuracy and generalization ability of the model.

[0047] Combining the SE (Squared Exponential) kernel and the Matérn kernel to better capture the complex characteristics of the heart tissue, the kernel function expression is as follows:

[0048] k(x,x')=σ2_SE*exp(-||x-x'||2 / (2l2_SE))+σ2_Matérn*(1+√

[0049] 3||x-x'|| / l_Matérn)*exp(-√3||x-x'|| / l_Matérn)

[0050] Among them, σ2_SE, l_SE, σ2_Matérn, and l_Matérn are hyperparameters, which are optimized by maximum likelihood estimation.

[0051] The Bayesian optimization module, based on the Bayesian optimization algorithm, intelligently selects the next optimal tactile sensor sampling point to obtain the optimal hardness distribution map with the least number of samplings.

[0052] Using the UCB (Upper Confidence Bound) acquisition function, the next optimal scanning point is selected, and the acquisition function expression is:

[0053] α_t(x) = μ_t(x) + β_t^(1 / 2) * σ_t(x)

[0054] Where μ_t(x) and σ_t(x) are the mean and standard deviation at point x at time t respectively, and β_t is a parameter that balances exploration and exploitation.

[0055] The energy constraint optimization module optimizes the motion path of the robotic arm, minimizes energy consumption, and improves system efficiency while ensuring positioning accuracy.

[0056] Considering the kinematic model of the robotic arm, minimizing the joint motion and the moving distance of the end effector, the optimization objective function is: min E = w1 * Σ|θ_i| + w2 * ||Δp||, where θ_i is the joint angle change, Δp is the end effector position change, and w1 and w2 are weight coefficients.

[0057] The incision recommendation module, based on the constructed hardness distribution map, uses image segmentation and edge detection algorithms to identify the soft tissue area around the heart valve and recommends the best incision position.

[0058] The human-machine interaction interface includes a display module and a control module. The display module is used to display information such as tactile sensor data, hardness distribution map, surgical path planning, etc. in real time, providing intuitive visual feedback for surgeons. The control module is used to allow surgeons to control the motion of the robotic arm and adjust system parameters through joysticks, gesture recognition, etc.

[0059] Furthermore, the present invention also provides a robotic-assisted heart valve surgery recommendation method, as Figure 2 shown, the method includes the following steps:

[0060] S1: Import the patient's preoperative CT or MRI images into the system, and the system automatically identifies the heart valve area;

[0061] Specifically, the system is initialized, the CT or MRI images of the patient are input, the approximate position and shape information of the heart valve area are extracted, and the initial scanning area and sampling density are set according to the image information.

[0062] S2: Control the robotic arm to drive the tactile sensor to perform a preliminary scan of the heart valve area to obtain initial hardness data;

[0063] Specifically, control the robotic arm to drive the tactile sensor to perform a preliminary scan of the set scanning area in a grid-like path, and the sampling interval can be set to 5 mm to quickly obtain the overall hardness distribution.

[0064] S3: Perform data processing on the preliminary scan data based on the Gaussian process model to generate an initial hardness distribution map;

[0065] S4: Based on the Bayesian optimization algorithm, automatically select the next optimal scan point and control the robotic arm to perform a fine scan;

[0066] Specifically, use the Bayesian optimization module to calculate the position of the next optimal scan point, plan the optimal movement path of the robotic arm, control the robotic arm to move to the target position, and perform a fine scan, and the sampling interval can be reduced to 1 mm.

[0067] Repeat steps S3 and S4, add the newly collected data to the Gaussian process model, and continuously update and optimize the hardness distribution map until the preset accuracy or the number of scans is reached.

[0068] S5: Based on the final hardness distribution map, identify the soft tissue area around the heart valve suitable for the incision, and the system recommends multiple optimal incision positions to the surgeon and provides an analysis of the advantages and disadvantages of each position;

[0069] Specifically, use the incision recommendation module to analyze the final hardness distribution map, identify the soft tissue area around the heart valve, mark the potential incision positions, sort the potential incision positions, recommend the top three incision positions to the surgeon, and provide an analysis of the advantages and disadvantages of each position.

[0070] S6: The system plans the optimal surgical path according to the final incision position selected by the surgeon and performs robotic-assisted heart valve surgery operations.

[0071] Specifically, the surgeon selects the final incision position according to the system recommendation and his own experience, the system plans the optimal surgical path according to the selected incision position, and performs robotic-assisted heart valve surgery operations according to the selected incision position and surgical path.

[0072] In addition, during the entire surgical procedure, the system continuously monitors the change in the hardness of the tissue around the incision, and issues an alarm in a timely manner if any abnormality is detected; the system records the hardness data and the basis for incision selection during the entire surgical procedure for subsequent analysis and system optimization.

[0073] The specific embodiments of the present invention have been described above. It should be understood that the present invention is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which do not affect the essence of the present invention. Without conflict, the embodiments of the present application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. A robot-assisted heart valve surgery recommendation system, characterized in that: The system includes a robot operation platform, a multimodal perception module, an intelligent control and decision module and a human-computer interaction interface. The robot operation platform includes a robot arm and an end effector. The multimodal perception module includes a tactile sensor and a medical imaging system. The tactile sensor is installed at the end of the robot arm. The intelligent control and decision module includes a data processing module, a Gaussian process model module, a kernel function fusion module, a Bayesian optimization module, an energy constraint optimization module and an incision recommendation module. The data processing module is responsible for collecting, filtering and preprocessing the data of the tactile sensor, and aligning and fusing it with the medical imaging data; the Gaussian process model module is used to construct a heart surface tissue hardness distribution map based on a Gaussian process regression algorithm to achieve a quantitative description of the characteristics of the heart tissue; the kernel function fusion module is used to optimize the kernel function parameters of the Gaussian process model; the Bayesian optimization module is based on the Bayesian optimization algorithm, intelligently selects the next optimal tactile sensor sampling point, and obtains the optimal hardness distribution map with the least number of sampling times; the energy constraint optimization module is used to optimize the motion path of the robot arm; the incision recommendation module recommends the best incision position according to the constructed hardness distribution map.

2. The robot-assisted heart valve surgery recommendation system according to claim 1, characterized in that: The robot arm includes at least one robot arm with high precision and high flexibility, which is used to control the movement and positioning of the tactile sensor and perform surgical operations; the end effector is installed at the end of the robot arm and is used to fix and operate the tactile sensor.

3. The robot-assisted heart valve surgery recommendation system according to claim 1, characterized in that: The medical imaging system is used to integrate CT or MRI images of patients acquired before surgery, provide three-dimensional anatomical structure information of the heart, and fuse it with tactile sensor data.

4. The robot-assisted heart valve surgery recommendation system according to claim 1, characterized in that: The incision recommendation module uses image segmentation and edge detection algorithms based on the constructed hardness distribution map to identify the soft tissue area around the heart valve and recommend the best incision position.

5. The robot-assisted heart valve surgery recommendation system according to claim 1, characterized in that: The human-computer interaction interface includes a display module and a control module. The display module is used to display tactile sensor data, hardness distribution map, and surgical path planning in real time to provide surgeons with intuitive visual feedback; the control module is used to allow surgeons to control the movement of the robot arm through a joystick and gesture recognition, and to adjust system parameters.

6. A robot-assisted heart valve surgery recommendation method, characterized in that: The method comprises the following steps: The patient's preoperative CT or MRI images are imported into the system, which automatically identifies the heart valve area; Control the robot arm to drive the tactile sensor to perform a preliminary scan of the heart valve area to obtain initial hardness data; The preliminary scanning data is processed based on the Gaussian process model to generate an initial hardness distribution map; Based on the Bayesian optimization algorithm, the next optimal scanning point is automatically selected and the robot arm is controlled to perform fine scanning; Based on the final hardness distribution map, the system identifies the soft tissue area around the heart valve that is suitable for incision. The system recommends multiple optimal incision locations to the surgeon and provides an analysis of the advantages and disadvantages of each location. The system plans the optimal surgical path based on the final incision location selected by the surgeon and performs robot-assisted heart valve surgery.

7. The robot-assisted heart valve surgery recommendation method according to claim 6, characterized in that: The method identifies the soft tissue area around the heart valve that is suitable for incision based on the final hardness distribution map, recommends multiple optimal incision positions to the surgeon, and provides an analysis of the advantages and disadvantages of each position. The steps specifically include: using an incision recommendation module to analyze the final hardness distribution map, identifying the soft tissue area around the heart valve, marking potential incision positions, sorting the potential incision positions, recommending the top three incision positions to the surgeon, and providing an analysis of the advantages and disadvantages of each position.

8. The robot-assisted heart valve surgery recommendation method according to claim 6, characterized in that: The system plans the optimal surgical path according to the final incision position selected by the surgeon and performs the steps of robot-assisted heart valve surgery. Specifically, the system continuously monitors the hardness changes of the tissue around the incision during the entire operation and promptly issues an alarm if any abnormality is found. The system records the hardness data and incision selection basis during the entire operation for subsequent analysis and system optimization.

Images