Surgical navigation and quality control system based on multi-source information fusion

Abstract

The invention relates to the technical field of medical information systems, in particular to a surgical navigation and quality control system based on multi-source information fusion, and the system comprises an image collection module which is used for obtaining a preoperative CT image of a patient; collecting an endoscope video in the operation; the surgical navigation module is in communication connection with the image acquisition module and is used for receiving the CT image and the endoscope video sent by the image acquisition module; acquiring a real-time position of a surgical instrument in a patient coordinate system based on the CT image and the endoscope video; the relative position of the virtual scalpel and the focus is displayed in real time; the operation quality scoring module is in communication connection with the operation navigation module and is used for receiving the position information of the operation instrument sent by the operation navigation module; the deviation between the operation of the doctor and the operation target is evaluated and scored, and the preoperative CT image, the intraoperative endoscope video and other key information are fused, so that high-precision real-time operation navigation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of medical information systems, and in particular to a surgical navigation and quality control system based on multi-source information fusion. Background Art

[0002] In recent years, with the continuous advancement of medical technology, surgical navigation systems have played an increasingly important role in clinical practice. Traditional surgical navigation systems mainly rely on a single image source, such as preoperative CT or MRI images, to guide surgical operations. However, this method often seems to be inadequate in the face of complex surgical environments. Especially in soft tissue surgery, due to the deformation and movement of tissues, a single image source is difficult to provide real-time and accurate navigation information.

[0003] Some advanced systems attempt to introduce intraoperative images, such as endoscopic videos, to improve the real-time nature of navigation. However, these systems often have difficulty effectively aligning preoperative planning with intraoperative reality, resulting in insufficient navigation accuracy. In addition, most existing surgical navigation systems focus on spatial positioning, but pay insufficient attention to the evaluation and control of surgical quality. This makes it difficult for the surgical team to objectively evaluate the progress of the operation, and is unable to detect and correct potential problems in a timely manner.

[0004] In terms of surgical quality control, current methods mostly rely on post-operative manual evaluation and lack a real-time, objective evaluation mechanism. Some systems attempt to introduce automated evaluation, but they often only focus on a single indicator, such as instrument trajectory or operation time, which makes it difficult to fully reflect the quality and risk of the surgery.

[0005] In addition, existing systems generally have information island problems. Navigation systems, imaging systems, and quality control systems are often independent of each other, making it difficult to effectively integrate and utilize data. This not only increases the cognitive burden on doctors, but also limits the system's maximum effectiveness.

[0006] In the face of these challenges, there is an urgent need for a surgical assistance system that can integrate multi-source information to achieve precise navigation and comprehensive quality control. The present invention is aimed at the above problems and proposes a surgical navigation and quality control system based on multi-source information fusion. Summary of the invention

[0007] The present invention proposes a surgical navigation and quality control system based on multi-source information fusion, aiming to solve the conflicting technical problem between accurate navigation and real-time performance of the existing system.

[0008] The present invention proposes a surgical navigation and quality control system based on multi-source information fusion, comprising:

[0009] Image acquisition module, used for:

[0010] Obtain the patient's preoperative CT images;

[0011] Acquisition of intraoperative endoscopic video;

[0012] The surgical navigation module is connected to the image acquisition module for:

[0013] Receiving CT images and endoscopic videos sent by the image acquisition module;

[0014] Based on the CT image and the endoscopic video, obtaining the real-time position of the surgical instrument in the patient coordinate system;

[0015] Real-time display of the relative position of the virtual scalpel and the lesion;

[0016] The surgery quality scoring module is in communication with the surgery navigation module and is used to:

[0017] Receiving surgical instrument position information sent by the surgical navigation module;

[0018] The deviation between the doctor's operation and the surgical goal was evaluated and scored.

[0019] As a preference, it also includes:

[0020] A management system is connected to the surgical navigation module and the surgical quality scoring module in communication, and the management system includes:

[0021] Operation evaluation module, used to evaluate the standardization of surgical procedures and intraoperative operations;

[0022] Remote communication module, used for image and data transmission between systems.

[0023] Preferably, the remote communication module adopts a communication module with medical safety encryption function.

[0024] Preferably, the image acquisition module is further used for:

[0025] acquiring a plurality of endoscopic video images through a plurality of endoscopes;

[0026] Before the operation, the patient is scanned by a magnetic resonance imaging or CT scanning device to obtain a CT image of the patient;

[0027] During the operation, multiple sets of endoscopes are placed inside the patient's body to collect endoscopic video images.

[0028] Preferably, the image acquisition module is further used for:

[0029] By adopting multi-source information fusion and image registration strategy based on multimodal navigation, the preoperative CT images, intraoperative endoscopic images, fluorescent images and intraoperative anatomical landmarks are integrated to accurately locate the position of surgical instruments in three-dimensional space.

[0030] Preferably, the surgical navigation module is also used for:

[0031] Registration is performed based on the image acquisition module, and the registration adopts a virtual coordinate system registration method to locate the real-time positions of surgical instruments, endoscopes and optical fibers in the patient space.

[0032] Preferably, the surgery quality scoring module is also used for:

[0033] A surgical quality standard evaluation system is established based on surgical process data and patient information. In the surgical quality standard evaluation system, quantitative scores are given to standardization, accuracy, and risk avoidance evaluation items.

[0034] Preferably, the surgery quality scoring module is also used for:

[0035] Obtain preoperative CT image data, extract the lesion area from the CT image using multi-voxel segmentation technology, and establish a three-dimensional lesion model, using the three-dimensional coordinate information of the surgical target as a reference;

[0036] The position information of the surgical instrument relative to the surgical target is tracked in real time, and the surgical quality is scored based on the real-time tracked position of the surgical instrument.

[0037] As a preference, it also includes:

[0038] A management module for real-time remote display of surgical quality assessment results;

[0039] Statistics module, used to evaluate surgical quality based on surgical data.

[0040] Preferably, the image acquisition module is further used for:

[0041] Acquire images of the surgical target area and surrounding blood vessels and tissues through the multiple endoscopes to achieve real-time tracking;

[0042] According to the positional relationship between the real-time endoscope shooting points and the surgical target, the closed surface surrounded by the multiple endoscope shooting points is tracked, and the coordinates of the center point of the closed surface are calculated as the surgical target point;

[0043] The offset position of the surgical target is acquired according to the center point coordinates.

[0044] The beneficial effects of the present invention are mainly reflected in the following aspects:

[0045] The system of the present invention realizes high-precision real-time surgical navigation by innovatively integrating preoperative CT images, intraoperative endoscopic videos and other key information. At the same time, the system introduces a comprehensive quality assessment mechanism to perform real-time scoring on the standardization, accuracy and safety of the surgery, providing timely and objective feedback to the surgical team.

[0046] From a macro perspective, the system of the present invention realizes information integration and intelligent assistance for the entire surgical process. It breaks the situation in which the functional modules in the traditional system are separated from each other and builds a complete closed loop from preoperative planning, intraoperative navigation to postoperative evaluation. This comprehensive solution greatly improves the overall efficiency and safety of surgery.

[0047] At the micro level, the various modules of the present invention achieve deep synergy. For example, the image acquisition module not only provides basic data for the navigation module, but also works closely with the quality scoring module to achieve real-time evaluation of surgical operations. This synergy between modules significantly enhances the overall performance of the system, far exceeding the simple superposition of the effects of each module alone.

[0048] It is worth mentioning that the present invention has cleverly solved the contradiction between precise navigation and real-time performance. Through multi-source information fusion and advanced image registration technology, the system not only ensures high precision, but also achieves a millisecond-level response speed. This is particularly important for rapidly changing surgical environments.

[0049] In addition, the quality control mechanism of the present invention not only provides objective scoring, but also identifies potential risk factors. This forward-looking risk management greatly improves the safety of surgery and provides a strong guarantee for preventing medical accidents.

[0050] In general, the system of the present invention significantly improves the accuracy, safety and efficiency of surgery through innovative technologies such as multi-source information fusion, real-time navigation, and comprehensive quality control. It not only provides doctors with a powerful auxiliary tool, but also brings patients a better medical experience and treatment effect. The application of this system is expected to promote the development of precision medicine and make an important contribution to improving the overall quality of medical care. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a system logic block diagram of the present invention.

[0052] Figure 2 It is a logic block diagram of the image acquisition module of the present invention.

[0053] Figure 3 It is a logic block diagram of the surgical navigation module of the present invention.

[0054] Figure 4 It is a logic block diagram of the surgical quality scoring module of the present invention. DETAILED DESCRIPTION

[0055] See also Figure 1-4 The present invention provides a surgical navigation and quality control system based on multi-source information fusion. By fusing multiple information sources, the system achieves accurate surgical navigation and comprehensive quality control, significantly improving the accuracy and safety of surgery.

[0056] Specifically, the system of the present invention includes an image acquisition module 1, a surgical navigation module 2 and a surgical quality scoring module 3. The image acquisition module 1 is used to obtain the patient's preoperative CT image and the intraoperative endoscopic video. These image data provide a basis for subsequent surgical navigation and quality assessment. Preferably, the resolution of the CT image can reach 0.5-1mm, and the frame rate of the endoscopic video is usually 30-60fps to ensure image quality and real-time performance.

[0057] The surgical navigation module 2 is connected to the image acquisition module 1 in communication to receive CT images and endoscopic videos. Based on these data, the module can obtain the real-time position of the surgical instrument in the patient's coordinate system. In one embodiment of the present invention, the surgical navigation module 2 uses optical tracking technology, and the positioning accuracy can reach 0.1-0.3mm. In addition, the module also displays the relative position of the virtual scalpel and the lesion in real time, providing intuitive visual guidance for doctors.

[0058] The surgical quality scoring module 3 is connected to the surgical navigation module 2 to receive the surgical instrument position information. The core function of this module is to evaluate the deviation between the doctor's operation and the surgical goal and score it. The scoring is based on a percentage system, where 90 points or more are excellent, 80 to 90 points are good, 70 to 80 points are qualified, and below 70 points are unqualified. The scoring algorithm can be expressed as:

[0059]

[0060] in, is the actual surgical position vector, is the target surgical position vector, k is the correction coefficient, which is usually 50-100.

[0061] In one embodiment of the present invention, the system of the present invention further includes a management system 4. The system is connected to the surgical navigation module 2 and the surgical quality scoring module 3 in communication, further enhancing the functionality of the system. The management system 4 includes an operation evaluation module and a remote communication module. The operation evaluation module evaluates the standardization of the surgical process and intraoperative operations, and uses deep learning-based video analysis technology to identify the execution of standard operating procedures. The remote communication module is responsible for image and data transmission between systems, supporting real-time remote consultation and expert guidance.

[0062] In one embodiment of the present invention, the security of the remote communication module is further optimized. The module adopts a communication module with medical security encryption function to ensure the privacy of patients and the security of medical data. The present invention preferably adopts AES-256 encryption algorithm with a key length of 256 bits, which can effectively prevent data leakage and tampering.

[0063] The system of the present invention realizes accurate navigation and quality control of the entire surgical process through multi-source information fusion. The image acquisition module 1 provides high-quality preoperative CT and intraoperative endoscopic images, the surgical navigation module 2 realizes precise positioning and visual navigation based on these data, and the surgical quality scoring module 3 evaluates the surgical quality in real time. The introduction of the management system 4 further enhances the evaluation and remote collaboration capabilities of the system. This comprehensive technical solution significantly improves the accuracy and safety of the operation, reduces the risk of complications, and provides doctors with objective skill evaluation and remote expert support, which is of great significance to improving the overall medical quality. The system of the present invention further optimizes the function of the image acquisition module 1.

[0064] In one embodiment of the present invention, the module can not only acquire a single endoscope video, but also simultaneously acquire multiple video images through multiple endoscopes. This multi-view acquisition method significantly improves the visualization of the surgical area and provides doctors with a more comprehensive surgical field of view.

[0065] Preferably, the present invention uses 3-5 endoscopes to observe the surgical area from different angles. These endoscopes can be flexible, with a diameter of usually between 3-5 mm, and can flexibly enter various parts of the human body. The field of view of each endoscope can reach 120-140 degrees, and the resolution can reach 1920x1080 or higher, ensuring the clarity and detail of the image.

[0066] In the preoperative stage, the system of the present invention uses a magnetic resonance or CT scanning device to scan the patient and obtain high-precision CT images. The layer thickness of the CT scan is usually set to 0.5-1mm to ensure sufficient spatial resolution. Magnetic resonance imaging can provide better soft tissue contrast, which is particularly suitable for surgical planning of parts such as the brain and joints.

[0067] During the operation, multiple endoscopes are placed inside the patient to collect real-time video images of the surgical area. These endoscopes can use different imaging modes, such as white light mode and narrow band imaging (NBI) mode, to enhance the contrast of different tissues. The collected video is usually transmitted at a frame rate of 30-60fps, ensuring the smoothness and real-time nature of the image.

[0068] In one embodiment of the present invention, the advanced function of the image acquisition module 1 is that the system of the present invention adopts multi-source information fusion technology and image registration strategy based on multi-modal navigation. This advanced method integrates the CT images of preoperative planning, intraoperative endoscopic images, fluorescent images and intraoperative anatomical landmarks to achieve accurate positioning of surgical instruments in three-dimensional space.

[0069] Specifically, the image registration process of the present invention can be divided into the following steps:

[0070] 1. Feature extraction: Extract significant feature points from CT images and endoscopic images, such as vascular bifurcations, bone structures, etc.

[0071] 2. Feature matching: Use a robust matching algorithm, such as RANSAC (Random Sample Conse nsus), to match the extracted feature points.

[0072] 3. Transformation calculation: Based on the matching results, the transformation matrix from the CT coordinate system to the endoscope coordinate system is calculated. Usually, a rigid body transformation or affine transformation model is used.

[0073] 4. Image fusion: The registered CT image is fused with the real-time endoscopic image to generate an augmented reality view.

[0074] In a preferred embodiment of the present invention, a registration method based on deep learning is used, such as using a convolutional neural network (CNN) to automatically extract and match features. This method has higher robustness and accuracy than traditional methods, especially when dealing with soft tissue deformation.

[0075] In one embodiment of the present invention, the present invention describes in detail the registration process of the surgical navigation module 2. The module is registered on the basis of the image acquisition module 1, and the real-time position of the surgical instruments, endoscope and optical fiber in the patient space is located by a virtual coordinate system registration method. The advantage of this registration method is that a unified reference coordinate system can be established, so that spatial information from different sources can be integrated and analyzed under the same framework.

[0076] The system of the present invention uses an optical tracking system to locate the instrument, and the typical tracking accuracy can reach 0.1-0.3mm. Optical markers are usually installed on surgical instruments, endoscopes and optical fibers, and are tracked in real time by infrared cameras. The system establishes the corresponding relationship between the optical tracking coordinate system and the patient coordinate system through a pre-calibrated method.

[0077] Preferably, the present invention also uses an inertial measurement unit (IMU) to assist in positioning, so as to cope with situations such as optical occlusion, etc. The IMU data is fused with the optical tracking data through a Kalman filter algorithm, which further improves the stability and continuity of positioning.

[0078] In one embodiment of the present invention, the core function of the surgical quality scoring module 3 is that the module establishes a comprehensive surgical quality standard evaluation system based on surgical process data and patient information. This evaluation system quantitatively scores multiple aspects such as the standardization, accuracy and risk avoidance of the surgery, providing a reliable basis for the objective evaluation of surgical quality.

[0079] Specifically, the evaluation system of the present invention includes the following main indicators:

[0080] 1. Normative scoring (S norm ): Evaluate whether the surgical operation complies with standard procedures, with a full score of 100 points.

[0081] 2. Accuracy score (S prec ): Evaluate the deviation of the surgical instrument position from the predetermined target, with a full score of 100 points.

[0082] 3. Risk aversion score (S risk ): Assess the degree of protection of key structures during surgery, with a full score of 100.

[0083] The final surgical quality score (Q) was calculated by weighted average:

[0084] Q=w1·S norm +w2·S prec +w3·S risk

[0085] Wherein, w1, w2, and w3 are weights of various indicators respectively, and satisfy w1+w2+w3=1. In a preferred embodiment of the present invention, w1=0.3, w2=0.4, and w3=0.3 to balance the importance of various indicators.

[0086] This multi-dimensional scoring method can not only comprehensively reflect the quality of surgery, but also help doctors identify specific aspects that need improvement. Through long-term data accumulation and analysis, the system of the present invention can provide valuable quality management and training guidance for medical institutions, thereby continuously improving surgical standards and patient safety.

[0087] In one embodiment of the present invention, the function of the surgical quality scoring module 3 is to first obtain preoperative CT image data, then process the CT image using advanced multi-voxel segmentation technology, extract the lesion area and establish a three-dimensional lesion model. This process lays the foundation for subsequent surgical navigation and quality assessment.

[0088] The multi-voxel segmentation technology used in the present invention combines traditional image processing methods and deep learning algorithms. Specifically, the system first uses a grayscale threshold and region growing-based method for preliminary segmentation, and then applies a deep learning network such as 3D U-Net for fine segmentation. This hybrid method not only ensures the accuracy of segmentation, but also improves the robustness and generalization ability of the algorithm. Preferably, the segmentation accuracy of the present invention can reach a Dice coefficient of more than 95%.

[0089] The present invention adopts an improved 3D U-Net structure for lesion segmentation. Its loss function combines Dice loss and cross entropy loss:

[0090] L=αL Dice +(1-α)L CE

[0091] Among them, L Dice is the Dice loss, L CE is the cross entropy loss, α is the balance factor (usually 0.5). Dice loss is defined as:

[0092]

[0093] Among them, p i is the predicted probability map, g i is the true label, and N is the total number of voxels.

[0094] Based on the segmentation results, the system builds a high-precision three-dimensional lesion model. This model not only contains the geometric information of the lesion, but also integrates the metabolic activity information obtained from functional imaging such as PET-CT. This multi-modal fusion lesion model provides comprehensive information support for surgical planning. The spatial resolution of the model can usually reach 0.5mm, which is sufficient to support precise surgical navigation.

[0095] During surgery, the system of the present invention can track the position information of surgical instruments relative to the surgical target in real time. This function is achieved by combining the optical tracking system and the electromagnetic tracking system, ensuring stability in various surgical environments. The refresh rate of the optical tracking system can reach 60Hz, and the refresh rate of the electromagnetic tracking system can reach more than 100Hz, ensuring the real-time nature of the position information.

[0096] Based on the real-time tracking of the position of surgical instruments, the system dynamically calculates and scores the surgical quality.

[0097] The present invention uses an extended Kalman filter (EKF) to estimate the position of the device. The state vector x contains the position and velocity:

[0098]

[0099] State transfer equation:

[0100] x k =Fx k-1 +w k

[0101] Among them, F is the state transfer matrix, w k is the process noise.

[0102] Observation equation:

[0103] z k =Hx k +v k

[0104] Among them, H is the observation matrix, v k is the observation noise.

[0105] The scoring algorithm takes into account multiple factors, including the accuracy of the instrument position, the smoothness of the motion trajectory, and the safe distance from key anatomical structures. The scoring function can be expressed as:

[0106] Score=w1·f accuracy (d)+w2·f smoothness (v)+w3·f safety (s)

[0107] Where d is the distance between the instrument and the target position, v is the velocity vector of the instrument, and s is the minimum distance to the critical structure. accuracy 、f smoothness and f safety They are the evaluation functions for accuracy, smoothness and safety respectively, and w1, w2 and w3 are the corresponding weight coefficients.

[0108] In one embodiment of the present invention, the present invention also proposes a comprehensive scoring function:

[0109] Q=w1·f p (d)+w2·f s (v)+w3·f r (s

[0110] Among them, f p (d) is the position accuracy function, d is the distance vector between the instrument and the target, and f s (v) is the motion smoothness function, v is the device velocity vector, f r (s) is the risk assessment function, s is the safety distance vector from the critical structure, and w1, w2, w3 are weight coefficients.

[0111] Among them, the position accuracy function is defined as:

[0112]

[0113] Among them, σ is the standard deviation parameter, which controls the strictness of the accuracy requirement.

[0114] In one embodiment of the present invention, the management module 5 and the statistical module 6 are introduced to further enhance the functionality of the system of the present invention. The management module 5 is responsible for real-time remote display of the surgical quality assessment results. This function enables the surgical team and remote experts to understand the surgical progress and quality status in a timely manner, providing a basis for necessary intervention.

[0115] In a preferred embodiment of the present invention, the management module 5 adopts a Web-based real-time data visualization technology. The system uses the WebSocket protocol to achieve low-latency data transmission, and the front end uses libraries such as D3.js for dynamic data rendering. This solution not only ensures the real-time display (delay is usually <100ms), but also provides rich interactive functions, such as data scaling, filtering, etc.

[0116] Statistics module 6 comprehensively evaluates surgical quality based on accumulated surgical data. This module uses advanced data mining and machine learning techniques to extract valuable patterns and insights from massive surgical data. For example, the system can identify key factors affecting surgical quality, predict surgical risks, or develop personalized quality standards for different types of surgeries.

[0117] In one embodiment of the present invention, the statistical module 6 uses the random forest algorithm to perform feature importance analysis to identify the factors that have the greatest impact on surgical quality. At the same time, the module also uses the long short-term memory network (LSTM) to perform time series prediction of surgical quality and provide doctors with real-time quality trend warnings. These advanced analysis functions provide strong data support for the quality management and continuous improvement of medical institutions.

[0118] Finally, in one embodiment of the present invention, the functions of the image acquisition module 1 are further optimized, especially the image processing and analysis capabilities in a multi-endoscope scenario. The module acquires images of the surgical target area and surrounding blood vessels and tissues through multiple endoscopes, achieving all-round real-time tracking.

[0119] The present invention adopts an innovative algorithm to track the closed surface surrounded by multiple endoscopic shooting points based on the positional relationship between the real-time shooting points of the endoscope and the surgical target. This method not only improves the accuracy of positioning, but also can adapt to complex situations such as soft tissue deformation. Specifically, the system uses SLAM (Simultaneous Localization and Mapping) technology to reconstruct the motion trajectory of the endoscopic camera, and uses a stereoscopic vision algorithm to restore the 3D structure of the scene.

[0120] The coordinates of the center point of the closed surface are calculated using the weighted centroid method, namely:

[0121]

[0122] in, is the center point coordinate, is the coordinate of the i-th shooting point, w i is the corresponding weight (which can be determined based on image clarity or confidence).

[0123] The present invention adopts an improved SLAM algorithm to perform multi-view fusion. Define the objective function

[0124]

[0125] Among them, X is the camera pose set, P is the 3D landmark point set, z ij is the observed value, h(·) is the observation model, ρ(·) is the robust kernel function, Ω ij is the information matrix. By minimizing the above objective function, the camera pose and 3D structure can be optimized simultaneously.

[0126] The advantage of this method is that it can dynamically adapt to the movement and deformation of the surgical target. The system obtains the offset position of the surgical target by calculating the displacement of the center point between consecutive frames. In order to filter out noise and instantaneous fluctuations, the present invention uses a Kalman filter to smooth the offset trajectory. This processing not only ensures the real-time tracking, but also improves the stability and reliability of position estimation.

[0127] In summary, the surgical navigation and quality control system based on multi-source information fusion of the present invention realizes accurate navigation and all-round quality control of the whole surgical process through a series of innovative technical solutions. The various modules of the system work closely together to form a complete closed loop from image acquisition, surgical navigation to quality assessment and management. This not only significantly improves the accuracy and safety of surgery, but also provides strong technical support for the continuous improvement of medical quality.

[0128] Through the above method, the system of the present invention can realize the security analysis of sensitive data in a dynamic and intelligent manner, taking into account both privacy protection and performance optimization, and has strong practicality and innovation.

Claims

1. A surgical navigation and quality control system based on multi-source information fusion, characterized in that: include: Image acquisition module, used for: Obtain the patient's preoperative CT images; Acquisition of intraoperative endoscopic video; The surgical navigation module is connected to the image acquisition module for: Receiving CT images and endoscopic videos sent by the image acquisition module; Based on the CT image and the endoscopic video, obtaining the real-time position of the surgical instrument in the patient coordinate system; Real-time display of the relative position of the virtual scalpel and the lesion; The surgery quality scoring module is in communication with the surgery navigation module and is used to: Receiving surgical instrument position information sent by the surgical navigation module; The deviation between the doctor's operation and the surgical goal was evaluated and scored.

2. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: Also includes: A management system is connected to the surgical navigation module and the surgical quality scoring module in communication, and the management system includes: Operation evaluation module, used to evaluate the standardization of surgical procedures and intraoperative operations; Remote communication module, used for image and data transmission between systems.

3. The surgical navigation and quality control system based on multi-source information fusion according to claim 2, characterized in that: The remote communication module adopts a communication module with medical safety encryption function.

4. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: The image acquisition module is also used for: acquiring a plurality of endoscopic video images through a plurality of endoscopes; Before the operation, the patient is scanned by a magnetic resonance imaging or CT scanning device to obtain a CT image of the patient; During the operation, multiple sets of endoscopes are placed inside the patient's body to collect endoscopic video images.

5. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: The image acquisition module is also used for: By adopting multi-source information fusion and image registration strategy based on multimodal navigation, the preoperative CT images, intraoperative endoscopic images, fluorescent images and intraoperative anatomical landmarks are integrated to accurately locate the position of surgical instruments in three-dimensional space.

6. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: The surgical navigation module is also used for: Registration is performed based on the image acquisition module, and the registration adopts a virtual coordinate system registration method to locate the real-time positions of surgical instruments, endoscopes and optical fibers in the patient space.

7. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: The surgical quality scoring module is also used to: A surgical quality standard evaluation system is established based on surgical process data and patient information. In the surgical quality standard evaluation system, quantitative scores are given to standardization, accuracy, and risk avoidance evaluation items.

8. The surgical navigation and quality control system based on multi-source information fusion according to claim 7, characterized in that: The surgical quality scoring module is also used to: Obtain preoperative CT image data, extract the lesion area from the CT image using multi-voxel segmentation technology, and establish a three-dimensional lesion model, using the three-dimensional coordinate information of the surgical target as a reference; The position information of the surgical instrument relative to the surgical target is tracked in real time, and the surgical quality is scored based on the real-time tracked position of the surgical instrument.

9. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: Also includes: A management module for real-time remote display of surgical quality assessment results; Statistics module, used to evaluate surgical quality based on surgical data.

10. The surgical navigation and quality control system based on multi-source information fusion according to claim 1, characterized in that: The image acquisition module is also used for: Acquire images of the surgical target area and surrounding blood vessels and tissues through the multiple endoscopes to achieve real-time tracking; According to the positional relationship between the real-time endoscope shooting points and the surgical target, the closed surface surrounded by the multiple endoscope shooting points is tracked, and the coordinates of the center point of the closed surface are calculated as the surgical target point; The offset position of the surgical target is acquired according to the center point coordinates.

Images