Laparoscope and digestive endoscope combined positioning measurement system and method

Abstract

The invention relates to the technical field of medical instruments, and discloses a laparoscope and digestive endoscope combined positioning measurement system and method.The method comprises the steps of positioning mark initialization, image feature analysis, instrument path planning, dynamic positioning calibration, operation integrity evaluation and the like. Spatial matching is realized by setting multiple types of positioning marks and utilizing optical reflection signals, noise reduction and feature extraction are performed on endoscope imaging data to generate a measurement area boundary, an operation path is planned in combination with anatomical path information, the tail end position of an instrument is calibrated in real time, and operation integrity is evaluated. The system comprises a positioning module, an analysis module, a correction module and an integration module which are respectively used for realizing the functions of positioning mark setting, image processing and path generation, offset correction and operation evaluation. According to the system, the operation positioning precision and the operation safety are improved, intellectualization and precision of minimally invasive surgery are achieved, and the system is suitable for minimally invasive treatment of digestive system diseases.

Description

Technical Field

[0001] The present invention relates to the technical field of medical devices, and in particular to a laparoscopic-digestive-endoscopic combined positioning and measurement system and method. Background Art

[0002] In modern medicine, laparoscopy and digestive endoscopy are commonly used minimally invasive surgical procedures, widely used in the diagnosis and treatment of digestive system diseases. However, with the increasing demands for surgical sophistication, traditional technologies are increasingly exposed to limitations in positioning accuracy, operational safety, and postoperative evaluation, making them difficult to meet the demands of complex surgical scenarios.

[0003] From the perspective of positioning accuracy, traditional laparoscopes and digestive endoscopes rely mainly on the doctor's experience for visual positioning during surgery, and lack a real-time, accurate spatial coordinate reference system. For example, during gastrointestinal tumor resection surgery, due to the physiological activities of the organs (such as peristalsis) and differences in individual anatomical structures, it is difficult for doctors to quickly and accurately determine the spatial position and boundaries of the diseased tissue, which can easily lead to insufficient or excessive resection, affecting the surgical effect. At the same time, traditional positioning methods cannot achieve coordinate association between different instruments, resulting in errors in the matching of the field of view of the endoscopic device and the laparoscope, increasing the difficulty of operation.

[0004] In terms of operation path planning, existing technologies lack intelligent path generation methods based on image features. Doctors usually judge the operation path subjectively based on two-dimensional images, without fully considering the morphological characteristics (such as surface texture and contour curvature) and mechanical properties (such as elastic modulus) of the tissue. This empirical planning method may cause the instrument path to not match the actual anatomical structure, increasing the risk of tissue damage, especially when dealing with areas with rich blood vessels or complex structures, which can easily lead to complications such as bleeding and perforation. In addition, traditional methods do not quantitatively evaluate the safety factor of the path, and cannot predict the operation risk in advance, which limits the safety and efficiency of the operation.

[0005] The lack of a dynamic positioning and calibration mechanism is another shortcoming of traditional technology. During surgery, the position of the instrument tip may shift due to factors such as operator error and tissue traction. Traditional systems are unable to monitor the offset in real time and automatically correct it. Doctors need to frequently adjust the instrument position through visual observation, which leads to interruptions in the operation, prolongs the operation, and increases the probability of human error. For example, during a delicate submucosal dissection, even a slight positional deviation can lead to an incorrect dissection layer, affecting the accuracy of the surgery.

[0006] In terms of operational integrity assessment, traditional technologies lack a systematic evaluation system. The judgment of the operation effect after surgery mainly relies on the doctor's subjective examination and pathological results, and it is impossible to monitor the progress and quality of the operation in real time during the operation. For example, it is impossible to accurately calculate the remaining time of the operation phase, making it difficult to reasonably allocate surgical time; there is a lack of quantitative indicators for the precision verification of the completed area, and problems such as boundary offset cannot be discovered in time, resulting in a high incidence of postoperative complications. In addition, traditional systems are unable to integrate multi-dimensional data such as instrument movement rate and operation sequence, making it difficult to form a comprehensive operation evaluation report, which is not conducive to the optimization of surgical plans and the accumulation of doctor experience.

[0007] With the development of medical imaging technology, computer vision and intelligent control technology, the introduction of technologies such as multimodal image fusion, real-time positioning calibration and intelligent path planning into combined laparoscopic and digestive endoscopy surgery has become a key direction for solving the above problems. Although there are some studies on image-guided surgery in the existing technology, most of them remain at the application stage of single-modality images and lack the integration and collaborative processing of multi-source data (such as optical reflection signals, tissue morphological characteristics, and instrument coordinate data). At the same time, the existing system still has significant deficiencies in terms of real-time dynamic calibration and comprehensiveness of operation evaluation. There is an urgent need for a laparoscopic and digestive endoscopy combined positioning measurement system and method that can achieve precise positioning, intelligent planning, dynamic calibration and comprehensive evaluation to improve the safety and effectiveness of minimally invasive surgery. Summary of the Invention

[0008] The purpose of the present invention is to provide a laparoscopic-digestive-endoscopic combined positioning measurement system and method to solve the problems raised in the above background technology.

[0009] To achieve the above-mentioned object, the present invention provides the following technical solution: a laparoscopic-digestive-endoscopic combined positioning and measurement method, the method comprising:

[0010] Positioning mark initialization step: setting a first positioning mark in the surgical area, obtaining the initial position of the target organ, setting a second positioning mark of the target organ, and spatially matching the first positioning mark and the second positioning mark through optical reflection signals;

[0011] Image feature analysis step: collecting endoscopic imaging data, performing noise reduction processing on the imaging data to obtain a first image, extracting morphological features of the first image, determining tissue status based on the morphological features, and generating a measurement area boundary based on the tissue status;

[0012] Instrument path planning step: setting a third positioning marker of the endoscopic device, associating the third positioning marker with the second positioning marker through spatial coordinates, mapping the measurement area boundary to the endoscope field of view, obtaining anatomical path information, and planning the operation path according to the anatomical path information;

[0013] Dynamic positioning calibration step: acquiring the end position of the instrument in real time, comparing the end position with the operation path to obtain a first deviation value, and triggering a positioning correction instruction according to the first deviation value;

[0014] Operation integrity assessment steps: Integrate the instrument movement rate and the operation sequence number, calculate the remaining time of the operation phase based on the movement rate, transmit the remaining time and the operation sequence number to the main control terminal, perform accuracy verification on the completed area according to the preset ratio, generate a verification report and evaluate the operation integrity.

[0015] Preferably, the positioning mark initialization step further includes:

[0016] Acquire the surgical area in the patient's three-dimensional reconstructed model and set a first positioning mark, wherein the mark is composed of a random sequence and a first feature code, wherein the random sequence is created by a coordinate generator, and the feature code is a superposition value of the horizontal and vertical coordinates of the surgical area;

[0017] Identify the anatomical structure of the target organ and set a second positioning marker, wherein the marker consists of a random sequence and a second feature code, the random sequence is created by a coordinate generator, and the feature code consists of an organ type code and a spatial orientation code. The first positioning marker and the second positioning marker are spatially matched using infrared light waves.

[0018] Preferably, the image feature analysis step further includes:

[0019] Acquiring endoscopic imaging data, and performing frame sequence reassembly processing on the data to obtain a first image, wherein the processing includes motion compensation and edge sharpening;

[0020] Extracting morphological features of the first image, the morphological features including surface texture features and contour curvature features, the texture features including smoothness, wrinkle density, and blood vessel distribution density, and the curvature features including convex height and concave depth;

[0021] The tissue status levels are divided according to morphological characteristics, and the topological boundaries of the measurement area are generated based on the status levels.

[0022] Preferably, the judgment logic of the tissue status includes:

[0023] Call the tissue feature database, input the organ type code, and obtain the corresponding standard texture feature and curvature feature range;

[0024] Extract the texture features and curvature features of the current image and quantify them into a feature data set;

[0025] Calculate the coverage of the feature data set and the standard feature range, select the feature combination with the highest coverage, and use the tissue state corresponding to the feature combination as the current judgment result;

[0026] The area determined to be in a normal state is marked as an operable area.

[0027] Preferably, the instrument path planning step further includes:

[0028] Setting a third positioning mark of the endoscope device, the mark is composed of a random sequence and a third characteristic code, the random sequence is created by a coordinate generator, and the third characteristic code is a combination of the device model and the channel number;

[0029] Project the measurement area boundary onto the endoscopic imaging plane to obtain the coordinate set of key points of the anatomical path;

[0030] Generate the optimal operation path for the instrument based on the set of key point coordinates.

[0031] Preferably, the planning logic of the operation path includes:

[0032] Establish a three-dimensional space coordinate system and calculate the path curvature and device freedom parameters between key points;

[0033] Evaluate the path safety factor, which is determined by the path curvature, instrument extension angle, and tissue elastic modulus;

[0034] Path segments with safety factors higher than a threshold are selected and connected to form a continuous operation path.

[0035] Preferably, the dynamic positioning calibration step further comprises:

[0036] Collect the three-dimensional coordinate data of the instrument end in real time and compare it with the node coordinates of the operation path point by point;

[0037] When the coordinate offset exceeds the preset tolerance range, a positioning instruction including the correction direction and correction angle is generated;

[0038] The correction command is continuously updated until the end coordinates return to within the tolerance range.

[0039] Preferably, the operation integrity assessment step further comprises:

[0040] Obtain the movement trajectory data of the instrument end and calculate the displacement change per unit time;

[0041] The remaining time of the stage is calculated by combining the remaining path length and the displacement change;

[0042] Perform boundary coincidence check on completed areas and generate a verification report including offset statistics.

[0043] Preferably, the present invention further comprises a laparoscopic-digestive-endoscopic combined positioning and measurement system for implementing the above-mentioned laparoscopic-digestive-endoscopic combined positioning and measurement method, wherein the system comprises a positioning module, an analysis module, a correction module and an integration module;

[0044] The positioning module is used to set a first positioning mark, a second positioning mark and a third positioning mark, and collect target organ position data and endoscopic imaging data;

[0045] The analysis module is used to process imaging data and extract morphological features, generate measurement area boundaries and operation paths, and calculate the instrument end offset in real time;

[0046] The correction module is used to generate a spatial coordinate correction instruction according to the offset and control the execution mechanism;

[0047] The integration module is used to calculate the remaining time of the operation phase, perform accuracy verification and generate an operation integrity assessment report.

[0048] Preferably, the positioning module includes a multispectral imaging unit and a coordinate calibration unit, the analysis module includes a feature quantization processor and a path planning engine, the correction module includes a motion trajectory analyzer and an instruction generator, and the integration module includes a timing prediction model and a topology verification algorithm.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] In terms of positioning accuracy and spatial matching, by setting the first positioning marker, the second positioning marker, and the third positioning marker composed of a random sequence and a feature code, and using infrared light waves to achieve spatial matching, a three-dimensional coordinate reference system covering the surgical area, target organ, and endoscopic equipment was established. This positioning method breaks through the limitations of traditional visual positioning and can accurately determine the spatial position and mutual relationship of each component in real time. For example, before the operation begins, the initial position of the target organ can be accurately mapped to the surgical area coordinate system through spatial matching of the first positioning marker and the second positioning marker, avoiding positioning deviations caused by individual anatomical differences. During the operation, the coordinates of the third positioning marker are associated with the second positioning marker, so that the endoscopic field of view can match the organ's anatomical structure in real time. The doctor can intuitively see the projection of the measurement area boundary on the endoscopic imaging plane, which significantly improves the accuracy of lesion positioning and provides a basis for precise resection.

[0051] In terms of image feature analysis and tissue status judgment, the image quality is effectively improved and noise interference is reduced by performing frame sequence reorganization, dynamic compensation and edge sharpening on the endoscopic imaging data. Surface texture features (such as smoothness, wrinkle density, and blood vessel distribution density) and contour curvature features (such as protrusion height and depression depth) are further extracted and compared with the standard feature range in the tissue feature database to achieve a quantitative assessment of the tissue status. This judgment logic based on multi-dimensional morphological features can accurately distinguish normal tissue from diseased tissue. For example, in gastrointestinal polyp resection surgery, the boundaries of polyps can be accurately defined based on abnormalities in texture features (such as increased blood vessel distribution density) and changes in curvature features (such as protrusion height exceeding the normal range) to avoid missed or miscutting. At the same time, marking the normal state area as an operable area provides a safe boundary for operation path planning and reduces the risk of damaging healthy tissue.

[0052] In terms of the intelligence and safety of operation path planning, by establishing a three-dimensional spatial coordinate system, the path curvature between key points, the instrument's degree of freedom parameters, the tissue elastic modulus and other factors are comprehensively considered to quantitatively evaluate the path safety factor, and the path segments with a safety factor higher than the threshold are selected to form a continuous operation path. This planning logic fully combines the anatomical structure characteristics with the instrument operation characteristics. For example, when treating lesions close to large blood vessels, the system can automatically avoid high-curvature paths and select paths where the instrument extension angle is compatible with the tissue elasticity, reducing the risk of vascular damage caused by improper path selection. At the same time, the boundary of the measurement area is projected onto the endoscopic imaging plane to generate an optimal operation path containing a set of key point coordinates, allowing doctors to operate according to the preset safe path, thereby improving the standardization and repeatability of the operation.

[0053] In terms of real-time and accuracy of dynamic positioning calibration, the three-dimensional coordinate data of the instrument end is collected in real time and compared point by point with the node coordinates of the operation path. When the offset exceeds the preset tolerance range, a positioning instruction containing the correction direction and correction angle is immediately generated and continuously updated until the end coordinates return to the tolerance range. This closed-loop feedback mechanism can promptly correct positional deviations caused by the doctor's operation errors or tissue traction. For example, when performing submucosal injections or peeling operations, the system can monitor the position of the instrument end in real time. Once it deviates from the preset path, the correction instruction is immediately triggered to ensure that the operation always follows the planned path, improving the accuracy of the operation, shortening the operation time, and reducing tissue damage caused by repeated adjustments of the instrument.

[0054] In terms of the comprehensiveness and systematic nature of the operation integrity assessment, by integrating multi-dimensional data such as instrument movement rate, operation sequence number, and remaining path length, the remaining time of the operation phase is calculated in real time, helping doctors to reasonably allocate surgical time and avoid hasty or delayed operations due to improper time control. At the same time, the boundary overlap of the completed area is detected, and a verification report containing offset statistics is generated to achieve a quantitative assessment of the operation quality. For example, before the end of the operation, the system can automatically verify the boundaries of the resection area. If the offset is found to exceed the allowable range, the doctor can be reminded to make additional cuts or corrections in time to ensure the integrity of the operation. This full-process monitoring and postoperative evaluation system not only improves the quality of surgery, but also provides data support for the optimization of surgical plans and the improvement of doctors' skills.

[0055] Furthermore, the system architecture constructed by this invention (including a positioning module, an analysis module, a correction module, and an integration module) realizes full automation of the entire process, from positioning marker initialization and image feature analysis to dynamic calibration and operation evaluation. Through data exchange and collaborative work between the modules, a complete intelligent surgical assistance system is formed. The application of components such as a multispectral imaging unit, a coordinate calibration unit, and a feature quantification processor further enhances the system's adaptability to complex surgical scenarios, providing strong support for the application of laparoscopic and digestive endoscopy technologies in a wider range of clinical scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 This is a diagram showing the working principle of the laparoscopic-digestive-endoscopic combined positioning and measurement method of the present invention;

[0057] Figure 2 Design drawings for image feature analysis;

[0058] Figure 3 This is a design diagram for the organization status judgment logic. DETAILED DESCRIPTION

[0059] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0060] See also Figure 1-Figure 3 The present invention relates to a laparoscopic-digestive-endoscopic combined positioning and measurement method, and the specific implementation steps are as follows:

[0061] Obtain a three-dimensional model of the patient reconstructed through medical imaging (such as CT, MRI) before surgery, determine the surgical area in the model and set the first positioning marker. The marker consists of a random sequence and a first feature code, wherein the random sequence is randomly created by a coordinate generator, and the first feature code is the numerical superposition of the horizontal coordinate and the vertical coordinate of the surgical area, which is used to uniquely identify the spatial position of the surgical area. Then, the anatomical structure of the target organ is analyzed through image recognition technology, and a second positioning marker is set at a key position on the surface or inside the organ. The marker also contains a random sequence and a second feature code. The second feature code is composed of an organ type code (such as G-01 for the stomach, I-02 for the intestine, etc.) and a spatial orientation code (such as the head side, tail side, anterior wall, posterior wall, etc.). The first positioning marker and the second positioning marker are spatially matched by infrared light waves. Specifically, the infrared transmitting device and the receiving device capture the reflected signals of the two in real time, establish a spatial coordinate mapping relationship between the surgical area and the target organ, and ensure the uniformity of the positioning reference.

[0062] A high-definition camera at the front end of the endoscope captures real-time imaging data of the surgical area. The raw data is first subjected to noise reduction, using a median filter algorithm to remove salt-and-pepper noise. A motion compensation algorithm corrects image blur caused by physiological movements such as breathing and heartbeat. Finally, an edge sharpening algorithm enhances tissue boundary clarity, resulting in the first image. Subsequently, computer vision algorithms are used to extract morphological features from the first image: surface texture features include tissue smoothness (texture energy calculated using the gray-level co-occurrence matrix), wrinkle density (number of wrinkles per unit area), and vascularity density (percentage of vascular pixels). Contour curvature features include protrusion height (the difference between the highest point and the surrounding tissue) and concavity depth (the difference between the lowest point and the surrounding tissue). Based on the extracted morphological features, pre-set tissue status judgment rules are applied. For example, if smoothness falls below a threshold and vascularity density is abnormally elevated, an inflammatory state is identified; if the contour curvature exhibits abnormally sharp protrusions, a suspected pathological state is identified. This generates a boundary contour for the measurement area, clearly defining the tissue area requiring manipulation or assessment.

[0063] A third positioning marker is set on the endoscopic device (such as a laparoscopic instrument). The marker consists of a random sequence and a third characteristic code. The third characteristic code is a combination of the device model (such as LAP-2025) and the channel number (such as operating channel 1), which is used to identify the unique position of the instrument in space. The boundary of the measurement area is mapped to the real-time field of view of the endoscope through a coordinate transformation algorithm, and the key path points in the anatomical structure (such as vascular branch points, organ edge turning points, etc.) are identified to obtain the three-dimensional coordinate set of these key points. Based on these coordinate sets, a path optimization algorithm (such as the Dijkstra algorithm) is used to generate the optimal operation path for the instrument. This path needs to avoid dangerous structures such as important blood vessels and nerves, and take into account the operational freedom and tissue compliance of the instrument.

[0064] A micro-sensor (such as an electromagnetic positioning sensor) installed at the end of the instrument collects the end's three-dimensional coordinate data in real time, compares it point by point with the node coordinates on the operation path, and calculates the coordinate offset (i.e., the Euclidean distance between the current position and the target position). When the offset exceeds the preset tolerance range (such as 0.5mm), the positioning correction mechanism is triggered: a correction instruction is generated based on the offset direction and angle, and the instrument position is adjusted through the robotic arm control system. The correction instruction is continuously updated until the end coordinates return to the tolerance range, ensuring the accuracy of the operation path.

[0065] The movement trajectory data of the instrument end is collected in real time. The displacement change per unit time (i.e., movement rate) is calculated using a differential algorithm. The remaining time of the current operation phase is calculated based on the remaining path length. At the same time, the boundary coincidence of the area where the operation has been completed is detected. The postoperative area image is compared with the measurement area boundary planned before the operation through image registration technology. The offset statistics (such as average offset and maximum offset) are calculated, and a verification report containing the accuracy verification results is generated. Based on the operation sequence number (such as S001, S002, etc. generated in the order of steps) and the remaining time data, the integrity and standardization of the operation are comprehensively evaluated.

[0066] The present invention will be further described below in conjunction with Examples 1 to 5:

[0067] Example 1:

[0068] In the positioning mark initialization step, the specific implementation method for the setting and spatial matching process of the first positioning mark and the second positioning mark is as follows:

[0069] A 3D model of the patient, reconstructed from preoperative medical imaging data such as CT and MRI, is obtained, and the surgical area is precisely delineated within the 3D model. The determination of the surgical area should be based on preoperative diagnostic results. For example, for patients with gastric tumors, the surgical area can be defined as the gastric wall region encompassing the tumor and its surrounding safe resection margins. A first localization marker is placed within the surgical area, consisting of a random sequence and a first signature. The random sequence is created by a coordinate generator, which generates a random number combination based on the coordinate range of the surgical area in the 3D model. For example, if the surgical area's X-axis range is [10mm, 30mm], the Y-axis range is [15mm, 25mm], and the Z-axis range is [5mm, 18mm] in the 3D coordinate system, the coordinate generator can generate a 12-digit random number as the random sequence, such as "204587136921." The first signature is the sum of the horizontal and vertical coordinates of the surgical area, calculated by adding the horizontal and vertical coordinates of the surgical area's center point. Assuming that the horizontal coordinate of the center point of the surgical area is 20 mm and the vertical coordinate is 20 mm, the first characteristic code is 40. The first positioning mark is made of optical reflective material and has a high reflectivity infrared coating on the surface to facilitate the capture and recognition of infrared light waves.

[0070] Identify the anatomical structure of the target organ. Taking the small intestine as an example, the 3D model analyzes features such as the intestinal wall structure, mesenteric attachment location, and peristalsis direction, and sets a second positioning marker at a key location on the surface or inside the target organ. The second positioning marker also consists of a random sequence and a second feature code. The random sequence is generated in the same way as the first positioning marker, and is randomly generated by the coordinate generator within the coordinate range of the target organ. For example, for the small intestine segment located in the middle of the abdominal cavity, the random sequence "385274916023" is generated. The second feature code consists of an organ type code and a spatial orientation code. The organ type code uses a preset standard coding system, such as G-01 for the stomach, I-02 for the intestine, L-01 for the liver, and I-03 for the small intestine. The spatial orientation code is set according to the specific location of the target organ in the body, such as the head, tail, front wall, back wall, left wall, right wall, etc. If the positioning marker is set on the front wall of the upper small intestine, the spatial orientation code is A-01 (where "A" represents the front wall and "01" represents the upper section). The combined second feature code is I-03A-01. The physical form of the second positioning marker is consistent with the first positioning marker. Both are optical reflective markers with infrared coatings to ensure signal interaction compatibility between the two.

[0071] The first positioning mark and the second positioning mark are spatially matched through infrared light waves, which is specifically achieved through an infrared emitting device and a receiving device. The infrared emitting device is installed on a fixed bracket near the surgical area, and continuously emits infrared light waves with a wavelength of 850nm, which contains a modulated marker ID signal. The infrared coating on the surface of the first positioning mark and the second positioning mark has a high reflectivity for light waves of this wavelength, and the reflected signal is captured by multiple infrared receiving devices distributed around the surgical field. The receiving device has a built-in photoelectric sensor that can convert the optical signal into an electrical signal, and convert it into a digital signal through an analog-to-digital converter (ADC). The system analyzes the phase difference of the reflected signal through a phase detection algorithm to calculate the spatial distance and angle between the first positioning mark and the second positioning mark. Specifically, the relationship between the phase difference and the distance satisfies the formula: (where Δφ is the phase difference, d is the distance, and λ is the infrared wavelength). Solving this formula yields the linear distance between the two markers. Angle calculation is based on the spatial coordinates of multiple receiving devices, using triangulation to determine the relative azimuth and elevation angles of the two markers.

[0072] The real-time nature of spatial matching is achieved through high-frequency signal interaction. The infrared emission module sends 100 groups of pulse signals containing the marker ID per second to ensure that even if the patient moves slightly or the organ undergoes physiological displacement during the operation, the system can still update the spatial coordinate mapping relationship between the two markers in real time. To improve the matching accuracy, the system uses the Kalman filter algorithm to filter the collected distance and angle data to remove noise interference and achieve millimeter-level precision in spatial positioning. For example, when the coordinates of the first positioning marker are (x1, y1, z1) and the coordinates of the second positioning marker are (x2, y2, z2), the coordinate transformation matrix of the two is established through spatial matching, so that the spatial position of the surgical area and the target organ are uniformly represented in the same coordinate system, providing a benchmark for subsequent image analysis, path planning and other steps.

[0073] In practice, the placement of positioning markers must adhere to aseptic principles. Markers are made of medical-grade materials with a smooth, non-angular surface to avoid mechanical damage to tissue. The first positioning marker can be affixed to the abdominal wall near the surgical area using a biodegradable adhesive. The second positioning marker is fixed using a method tailored to the characteristics of the target organ: for thicker tissues like the stomach wall, the marker can be sutured and secured via fine needle puncture; for thinner tissues like the intestinal wall, micro-titanium clips are used to secure the marker. This ensures the marker remains stable during surgery and prevents displacement or dislodgement.

[0074] In addition, the system supports the setting of multiple groups of positioning markers to adapt to complex surgical scenarios. For example, when the operation involves multiple organs or requires staged positioning, multiple first positioning markers and second positioning markers can be set separately. Each group of markers is distinguished by a unique random sequence to avoid signal confusion. The management of markers is achieved through the system software interface. The operator can view the spatial coordinates, signal strength and matching status of each marker in real time on the interface. When the signal strength of a marker is detected to be lower than the threshold (such as lower than 50% of the initial intensity), the system automatically issues an alarm to prompt the operator to check the fixation of the marker or replace it with a new marker.

[0075] Example 2:

[0076] In the image feature analysis step, the endoscopic imaging data is first processed. The collected endoscopic imaging data contains continuous image frames. Due to the influence of physiological movements (such as breathing, digestive organ peristalsis) or instrument operation, the original data may have frame sequence confusion or motion blur. Therefore, it is necessary to first perform frame sequence reorganization processing. Specifically, the image frames are sorted by timestamps to ensure the consistency of their time sequence; at the same time, the optical flow method is used to analyze the pixel motion vectors between adjacent frames, and the displacement caused by tissue movement is dynamically compensated so that the dynamic tissue maintains positional consistency in continuous frames, thereby improving the temporal consistency of the image. After completing the frame sequence reorganization, the data is subjected to edge sharpening processing, and the grayscale gradient of the image edge is enhanced by the convolution kernel (such as the Sobel operator) to highlight the details of the tissue boundary and obtain a first image with higher clarity.

[0077] The morphological features of the first image are extracted, including surface texture and contour curvature. The surface texture feature extraction process is as follows: For smoothness, the standard deviation of the image's grayscale values ​​is calculated. A more uniform grayscale distribution and a smaller standard deviation indicate a smoother tissue surface. Wrinkle density is extracted using morphological opening and closing operations. Closing operations are first used to fill the pores within the wrinkles, followed by opening operations to remove noise. A contour detection algorithm is then used to identify and count the wrinkles, thereby determining the number of wrinkles per unit area. For vascular density extraction, the image is first converted from RGB to HSV color space. Threshold segmentation is performed using the characteristics of the blood vessels in the red channel (e.g., H component within the range of 0-10°, S component within the range of 30%-70%, and V component within the range of 20%-80%). Pixels in the vascular region are extracted, and the ratio of vascular pixels to total image pixels is calculated to obtain the vascular density.

[0078] The extraction of contour curvature features includes the calculation of convex height and concave depth. For each pixel in the image, the local surface is fitted by the grayscale values ​​of the neighboring pixels, and the curvature value of the point is calculated. The positive or negative curvature determines whether it is convex or concave. For the convex area, the point with the largest curvature value is selected as the convex vertex, and the grayscale difference between the vertex and the surrounding flat area (the area with a curvature value close to 0) is calculated. The grayscale difference is converted into the actual convex height through the calibration relationship between grayscale and actual physical distance (such as the grayscale-distance mapping table established by the calibration plate before surgery); for the concave area, the point with the smallest curvature value is selected as the bottom of the concave, and a similar method is used to calculate the actual distance between the bottom of the concave and the surrounding flat area, that is, the depth of the concave is obtained.

[0079] When determining tissue status, the tissue feature database is first called. This database pre-stores standard texture features and curvature feature ranges for different organ types. Specifically, each organ type in the database corresponds to an entry. For example, the entry corresponding to the stomach contains the standard deviation range of smoothness, wrinkle density range, vascular distribution density range, and normal range of contour curvature of the normal gastric mucosa. When determining the tissue status of the current image, the type code of the target organ is first entered (such as G-01 for the stomach) and the corresponding standard feature range is obtained from the database.

[0080] Extract the texture features and curvature features of the current image and quantify them into a feature data set. For example, for gastric images, the extracted data such as smoothness standard deviation, wrinkle density, vascular distribution density, protrusion height, and depression depth constitute a feature data set. Next, calculate the coverage of the feature data set and the standard feature range. The coverage is calculated as follows: for each feature, determine whether it falls within the corresponding standard feature range. If it falls within the range, it is counted as 1, otherwise it is counted as 0. The sum of the count values ​​of all features is divided by the total number of features, and the resulting ratio is the coverage. For example, if the feature data set contains 5 features, 4 of which fall within the standard range, the coverage is 80%.

[0081] The feature combination with the highest coverage is screened out, and the tissue state corresponding to the feature combination is used as the current judgment result. For example, if the feature combination with the highest coverage corresponds to a normal state, the current tissue is judged to be normal; if it corresponds to an inflammatory state, it is judged to be inflammatory. At the same time, the area judged to be in a normal state is marked as an operable area. Specifically, through image mask technology, a mask with the same shape as the operable area is generated on the first image. This mask can be used as a reference range for safe operation in subsequent operations and is highlighted in a specific color (such as green) on the endoscopic display screen for intuitive identification by the surgeon.

[0082] Throughout the entire image feature analysis process, data processing and feature extraction are performed by the feature quantization processor in the analysis module. The feature quantization processor uses an FPGA chip for parallel computing, enabling rapid processing of large amounts of image data to meet real-time requirements. For example, for endoscopic imaging data at 30 frames per second, the FPGA chip can complete a series of operations, including noise reduction, frame sequence reconstruction, edge sharpening, feature extraction, and tissue status determination, within milliseconds of each frame acquisition. This ensures timely provision of tissue status information and measurement area boundaries during surgery, supporting the accuracy of surgical operations.

[0083] In addition, the parsing module needs to transmit the generated measurement area boundary information to the instrument path planning step for subsequent path planning. The measurement area boundary is represented by a contour line composed of a series of coordinate points. These coordinate points correspond to the actual position of the target tissue in three-dimensional space. Through the coordinate transformation algorithm, they can be mapped to the two-dimensional plane of the endoscope's real-time field of view, providing the surgeon with clear guidance on the operating range. At the same time, the parsing module will also store the extracted morphological feature data and tissue status determination results to form record data of the surgical process, which can be used for postoperative review and analysis, providing a basis for surgical effect evaluation and experience summary.

[0084] Example 3:

[0085] In the instrument path planning step, the third positioning mark of the endoscopic device must first be set. This mark consists of a random sequence and a third feature code. The random sequence is randomly created by a coordinate generator according to preset rules. For example, a unique 16-bit string is generated based on parameters such as the device startup timestamp and the surgery ID to ensure the uniqueness of the mark. The third feature code is a combination of the device model and the channel number. The device model uses the standard encoding defined by the manufacturer (such as the OLYMPUS brand laparoscopic device model is LAP-2025, the Fujinon brand endoscope model is EC-600WM, etc.), and the channel number corresponds to the operating channel on the device (such as the main operating channel is No. 1, the auxiliary channel is No. 2, etc.). For example, if the device model is LAP-2025 and the No. 1 operating channel is used, the third feature code is LAP-2025-1. The third positioning mark is fixed to the physical carrier by an optical mark (such as a reflective patch, LED coding light) or an electromagnetic mark (such as a radio frequency identification tag) installed on the instrument handle or rod body to ensure that it can be accurately identified by the positioning module's sensor (such as an infrared camera, electromagnetic locator) in the surgical space.

[0086] The measurement area boundary is mapped to the endoscopic field of view, which is achieved by coordinate projection transformation. The measurement area boundary is represented by a set of three-dimensional coordinate points in the preoperative three-dimensional reconstruction model (such as {P i (x i ,y i ,zi )}, i=1,2,…,n), which need to be converted into the two-dimensional coordinates of the endoscope real-time imaging plane. First, establish the perspective projection model of endoscope imaging, which includes the endoscope's internal parameter matrix K (describing the lens's focal length, principal point coordinates and other internal parameters) and the external parameter matrices R and T (describing the endoscope's rotation matrix and translation vector in the world coordinate system). For each three-dimensional point P i (x i ,y i ,z i ), its two-dimensional coordinates p in the imaging plane i (u i ,v i ) is calculated using the following formula:

[0087] p i =K·(R·P i +T)

[0088] Among them, u i 、v i is the pixel coordinate, K is the endoscope internal parameter matrix, including the focal length f x 、f y and the principal point coordinates (c x ,c y ),Right now:

[0089]

[0090] R is a 3×3 rotation matrix that describes the rotation angle of the endoscope relative to the world coordinate system (such as the Euler angles α, β, and γ around the X, Y, and Z axes); T is a 3×1 translation vector that describes the position of the endoscope's optical center in the world coordinate system (t x ,t y ,t z Through the above-mentioned projection transformation, the boundaries of the three-dimensional measurement area can be converted into two-dimensional contour lines in the endoscopic field of view, which are superimposed on the real-time image, allowing the surgeon to intuitively identify the anatomical structure and operating range.

[0091] When obtaining anatomical path information, it is necessary to identify key anatomical structures in the endoscopic field of view and extract key points of the path. Key points include but are not limited to: lesion edges (such as tumor boundaries), vascular branch points (such as branches of the superior mesenteric artery), organ junctions (such as the gastroduodenal junction), safe operation boundaries (such as 5mm away from important blood vessels), etc. The coordinates of these key points are obtained by marking them on the image through manual interaction (such as the surgeon specifying points through a touch screen or foot pedal), or automatically detecting them through computer vision algorithms (such as target detection models based on deep learning to identify structures such as blood vessels and organ edges and locate key points). During automatic detection, the endoscopic image is first preprocessed (such as normalization and contrast enhancement), and then the trained neural network model (such as YOLOv5) is input to output the bounding box and key point coordinates of each anatomical structure to form a key point coordinate set {Q j (u j ,v j )},j=1,2,…,m, where u j 、v j is the pixel coordinate in the image, which needs to be converted into three-dimensional world coordinates through inverse projection transformation {Q j (x j ,y j ,z j )} to plan instrument paths in three-dimensional space.

[0092] When generating the optimal operation path for the instrument based on the key point coordinate set, a three-dimensional space coordinate system is first established. Taking the patient's operating table as the reference, the X axis is defined as the left-right direction (the patient's left side is negative and the right side is positive), the Y axis is defined as the head-tail direction (the patient's head is positive and the feet are negative), and the Z axis is the depth direction (the direction away from the operating table is positive). Each key point Q j (x j ,y j ,z j ) has a unique position in the coordinate system. Then calculate the path curvature and instrument freedom parameters between each key point. The path curvature κ represents the curvature of the path. For adjacent key points Q j With Q j+1 , assuming that the path between two points is a space curve, its curvature can be calculated by differential geometry method, the formula is:

[0093]

[0094] Where T(s) is the tangent vector of the curve at the arc length parameter s, T ′(s) is the derivative of the tangent vector with respect to the arc length, and v(s) is the velocity vector of the curve at that point. In practical applications, it can be approximated by discretization methods, such as fitting an arc with three adjacent points, where the curvature is the inverse of the arc radius. Instrument degree of freedom parameters include the rotatable angle of the instrument end and the range of motion of the joints. For example, a laparoscopic instrument has five degrees of freedom, and its end can rotate within a range of ±90°. The maximum bending angle of each joint is 120°. These parameters limit the feasibility of the path (for example, excessive curvature of the path may prevent the instrument from reaching the target).

[0095] When evaluating the path safety factor, the safety factor S is determined by the path curvature κ, the instrument extension angle θ, and the tissue elastic modulus E. The instrument extension angle θ is the angle between the instrument's current position and the target position, measured in real time by the angle sensor on the instrument. The tissue elastic modulus E is obtained through preoperative elastography (such as ultrasound elastography) and reflects the tissue's ability to resist deformation. A larger value indicates a stiffer tissue. The safety factor is calculated as follows:

[0096]

[0097] In the formula, the unit of κ is mm -1 , θ is in radians (rad), and E is in kilopascals (kPa). This formula indicates that the smaller the path curvature, the smaller the instrument extension angle, and the smaller the tissue elastic modulus, the higher the safety factor and the safer the path. For example, when the path is a straight line, κ = 0, and the safety factor is theoretically infinite (the maximum value is taken in actual calculations). As the instrument extension angle approaches its maximum range of motion, θ increases, and the safety factor decreases.

[0098] The specific logic of path planning is as follows: first, all key points are sorted in anatomical order (such as from the proximal end to the distal end), and then candidate path segments (such as straight line segments and Bezier curve segments) are generated between any two points. The curvature κ, instrument extension angle θ (assuming the maximum extension angle when the instrument moves along the path segment from the current position) and tissue elastic modulus E (the average elastic modulus of the tissue passed by the path segment) of each candidate path segment are calculated, and then substituted into the safety factor formula to calculate S. The safety factor S is higher than the preset threshold S. th Path segments (e.g., 0.5) are connected to form a continuous operation path. If multiple paths meet the requirements, the optimal path can be selected based on factors such as path length and operation time (e.g., shortest path first). The resulting operation path is displayed as a highlighted line on the navigation system interface, with the locations of key points and safety factors annotated for the surgeon's reference.

[0099] During the path planning process, the path planning engine of the parsing module achieves efficient calculations based on the GPU-accelerated algorithm library. The parallel computing capability of the GPU can handle the calculation tasks of multiple candidate path segments at the same time, significantly shortening the planning time and meeting the real-time requirements of surgery. For example, for a path planning task involving 10 key points, the GPU can complete the safety factor calculation and screening of all candidate path segments within tens of milliseconds, while traditional CPUs may take hundreds of milliseconds or even longer. In addition, the path planning engine also supports dynamic adjustment. When new anatomical structures are discovered during surgery (such as small blood vessels that were not identified before surgery), new key points can be manually added or the positions of existing key points can be adjusted to regenerate the operation path and ensure the safety and feasibility of the path.

[0100] The generated operation path must be consistent with the coordinate system of the positioning module, that is, all coordinates are based on the same world coordinate system. Therefore, before path planning, the positioning module's coordinate calibration unit must be used to uniformly calibrate the coordinates of the endoscope, instruments, and patient position to ensure spatial consistency among the 3D reconstructed model, endoscopic image, instrument position, and other data. The coordinate calibration process is achieved using devices such as laser rangefinders and infrared positioning markers. For example, multiple infrared reflective balls are attached to the patient's body surface, their positions are captured by an infrared camera, and the conversion relationship between the patient's coordinate system and the world coordinate system is established.

[0101] In summary, the instrument path planning step generates an optimal path that takes into account safety, feasibility, and operational efficiency through unique identification of positioning markers, coordinate projection transformation, key point extraction, curvature and safety factor calculation, etc., providing precise navigation guidance for laparoscopic digestive endoscopic surgery, ensuring that instrument operations are performed along the preset anatomical path, and reducing the risk of damaging surrounding tissues and structures.

[0102] Example 4:

[0103] During the dynamic positioning calibration step, the system uses a micro-sensor integrated into the end of the instrument to obtain its spatial position data in real time, enabling precise monitoring and adjustment of the instrument's operating path. For example, the monopolar electrocoagulation forceps commonly used in laparoscopic surgery have an electromagnetic positioning sensor installed at their end (such as the micro-probe of the Polhemus Fastrack system). This sensor can receive external electromagnetic field signals and calculate the end's three-dimensional coordinate data in real time, with a sampling rate of up to 100 times per second, ensuring the high timeliness of position information.

[0104] In specific implementation, the coordinate system is first calibrated uniformly through the positioning module before the operation. The three-dimensional reconstructed model of the patient's surgical area, the endoscopic field of view, and the position of the instrument are all mapped to the same world coordinate system. The coordinate system takes the center of the operating table as the origin, the X-axis is the left and right direction of the patient, the Y-axis is the head and tail direction, and the Z-axis is the depth direction perpendicular to the bed surface. For example, when the patient is in the supine position, the corresponding position range of the three-dimensional model of his liver in the coordinate system is X∈[150,250]mm, Y∈[300,400]mm, and Z∈[50,100]mm. The initial coordinates of the end of the instrument are (200,350,80)mm, and the coordinates of the first node on the operation path are (210,360,85)mm.

[0105] During surgery, the sensor continuously collects real-time coordinate data from the end of the instrument, such as outputting a set of coordinate values ​​(x, y, z) every 10 milliseconds. The system compares the real-time coordinates with the node coordinates on the operation path point by point and calculates the spatial offset between the two. The offset is calculated based on the Euclidean distance principle, that is, comparing the sum of the absolute values ​​of the difference between the current position and the target node on the three coordinate axes. For example, when the real-time coordinates are (205, 355, 82) mm and the target node coordinates are (210, 360, 85) mm, the X-axis offset is 5 mm, the Y-axis offset is 5 mm, and the Z-axis offset is 3 mm, with a total offset of 13 mm (in actual applications, different offset calculation methods can be selected according to the surgical accuracy requirements, such as the square root of the Euclidean distance).

[0106] When the offset exceeds the preset tolerance range, the system triggers the positioning correction mechanism. The tolerance range is pre-set according to the type of surgery and the accuracy of the operation. For example, in gastrointestinal perforation repair, the tolerance range can be set to 0.5mm; in tumor resection, the tolerance range can be reduced to 0.3mm due to the need to more accurately avoid the tissue around the lesion. Taking the tolerance range of 0.5mm as an example, if the offset obtained by a certain comparison calculation is 0.6mm, which exceeds the upper limit of the tolerance, the system immediately generates a positioning correction instruction. The correction instruction includes a correction direction and a correction angle: the correction direction is the vector direction from the current position to the target node. For example, if the current position is at the lower left of the target node, the correction direction is to the upper right; the correction angle is the angle between the vector and the instrument axis, which is determined by geometric calculation to guide the robotic arm to adjust the posture of the instrument.

[0107] The correction instructions are transmitted to the robotic arm actuator through the control system. The robotic arm adopts a multi-joint design (such as a 6-degree-of-freedom robotic arm), and each joint is equipped with a servo motor and an encoder to accurately control the translation and rotation of the instrument. For example, when the correction direction is to the upper right and the angle is 30°, the end effector of the robotic arm first moves 0.3mm in the positive direction of the X-axis and 0.4mm in the positive direction of the Y-axis (decomposition of the displacement according to trigonometric functions), and at the same time rotates 30° around the Z-axis to adjust the angle of the instrument so that the end faces the target node. During the correction process, the system continuously updates the correction instructions, performing coordinate comparison and instruction adjustment every 50 milliseconds until the end coordinates return to within the tolerance range.

[0108] The entire dynamic positioning calibration process is characterized by real-time and continuity. Taking the operation of removing intestinal polyps as an example, the instrument needs to move along the preset path from the base edge of the polyp to the distal end, with a path node set every 2mm. During the movement, if the end of the instrument deviates from the path node by 0.4mm (not exceeding the 0.5mm tolerance) due to intestinal peristalsis, the system will not trigger the correction temporarily; when the peristalsis amplitude increases and the offset reaches 0.6mm, the system immediately starts the correction, and fine-tunes the position of the instrument through the robotic arm so that the end returns to the node coordinate within ±0.5mm within 200 milliseconds. During this process, the surgeon can observe the real-time position of the end of the instrument and the path deviation prompt (such as the red arrow indicating the offset direction, and the numerical value showing the offset) through the endoscopic display screen, and can choose to manually intervene in the correction process or pause automatic correction.

[0109] The system also features dynamic calibration logging, storing in real time information such as the timestamp, offset, correction direction, correction angle, and actuator response parameters of each correction. For example, a log entry might include: "2025-05-28 14:30:15.234, offset 0.62mm, direction X+Y+Z+, angle 28.5°, motor speed 50rpm, correction time 180ms." This data can be used for postoperative operation review, analyzing the causes of instrument deviation (such as changes in tissue tension and physician operating habits), and providing a basis for optimizing path planning algorithms or improving instrument design.

[0110] In addition, the dynamic positioning calibration step forms a closed-loop feedback with the positioning module and the analysis module. The positioning module monitors tissue displacement (such as organ displacement caused by instrument contact) in real time through the multispectral imaging unit and transmits the data to the analysis module; the analysis module recalculates the path node coordinates and sends them to the dynamic positioning calibration module to achieve dynamic update of the operation path. For example, when the electrocoagulation forceps touches the stomach wall and causes the gastric body to shift by 3mm, the infrared camera of the positioning module captures the relative position change between the first positioning marker and the second positioning marker, and the analysis module then adjusts the subsequent path node coordinates. The calibration module makes corrections based on the new coordinates to ensure that the instrument always operates along the updated path.

[0111] In terms of hardware implementation, the correction module consists of a motion trajectory analyzer and a command generator. The motion trajectory analyzer utilizes a digital signal processor (DSP) to rapidly process high-frequency coordinate data and perform offset calculations and logical judgments. The command generator communicates with the robotic arm controller via the CAN bus, ensuring real-time and reliable command transmission. The entire calibration process maintains a latency of less than 100 milliseconds, meeting the stringent real-time requirements of minimally invasive surgery.

[0112] Example 5:

[0113] In the operation integrity assessment step, the system quantitatively evaluates the standardization and completion quality of the surgical operation by integrating the instrument motion data and the spatial information of the operation area. Taking gastric stromal tumor resection as an example, the system obtains the movement trajectory data of the instrument end in real time. The data is provided by the electromagnetic positioning sensor or optical tracking marker installed on the laparoscopic instrument, and contains the three-dimensional coordinates (x, y, z) of 100 sampling points per second. After removing the noise through the trajectory smoothing algorithm (such as Gaussian filtering), the displacement change per unit time, that is, the movement rate, is calculated. For example, if a certain trajectory moves from the coordinates (180,220,90) mm to (200,235,85) mm in 5 seconds, the total displacement is √[(200-180) 2 +(235-220) 2 +(85-90)2]=25mm, and the average moving speed is 5mm / s.

[0114] The system calculates the remaining duration of the procedure by combining the remaining path length with the movement rate. The remaining path length is the sum of the lengths of all unfinished path segments from the current position to the end of the path. For example, if the total operation path length is 50mm, 25mm has been completed, and the remaining path length is 25mm, and the current average movement rate is 5mm / s, the estimated remaining duration is 5 seconds. This function provides a time reference for the surgeon, facilitating optimal operation rhythm and avoiding the increased risk of tissue damage due to excessive time consumption.

[0115] The accuracy check of the completed area is achieved through boundary coincidence detection. Taking the wound surface after tumor resection as an example, the system aligns the boundary of the measurement area planned before surgery (such as the three-dimensional contour containing the 5mm safety margin of the tumor edge) with the wound edge in the real-time image after surgery. The registration process uses an image matching algorithm based on feature points. First, the feature points of the preoperative boundary (such as the contour inflection point and the curvature extreme point) are extracted, and then the corresponding feature points are identified in the postoperative image through the scale-invariant feature transform (SIFT) algorithm. The spatial transformation matrix (translation, rotation, scaling) of the two sets of feature points is calculated, and the preoperative boundary is mapped to the postoperative coordinate system. Next, the area of ​​the overlapping area between the mapped boundary and the actual wound edge, as well as the total area of ​​the preoperative boundary, are calculated. The ratio of the two is the boundary coincidence. For example, the total area of ​​the preoperative boundary is 200mm 2 , the overlapping area is 185mm 2 , the overlap is 92.5%. The system also calculates the offset. For example, the maximum offset occurs in the upper left corner of the wound, with an offset distance of 0.4mm and an average offset of 0.15mm. This data is integrated into the verification report, presenting the offset value distribution of each area in tabular form.

[0116] The hardware basis for operational integrity assessment is an integrated module, which includes a timing prediction model and a topological verification algorithm. The timing prediction model is trained based on historical surgical data and adopts a recursive neural network (RNN) structure. The input parameters include instrument type, operation stage, tissue type, current movement rate, etc., and the output is the predicted value of the remaining time. For example, for the same type of gastric stromal tumor resection operation, the model can predict the remaining time of the current operation based on the average rate of similar stages in the historical data, with an error range of ±10%. The topological verification algorithm uses the shortest path algorithm in graph theory to calculate the shortest distance between each point on the edge of the postoperative wound and the preoperative boundary, and generates an offset matrix for visually displaying the operation accuracy (such as marking the degree of offset with different colors on the three-dimensional model, red indicates an offset >0.3mm, and green indicates an offset ≤0.3mm).

[0117] Taking intestinal adhesion lysis as an example, the operation sequence number is automatically generated according to the steps, such as S001 for locating the adhesion area, S002 for separating the proximal adhesion, S003 for separating the distal adhesion, and S004 for stopping bleeding on the wound. The system tracks the completion status of each sequence in real time. When S002 is completed, it automatically triggers the detection of the boundary overlap of the separated area. If the overlap of a certain area is detected to be lower than the preset threshold (such as 85%), the system will issue an alarm prompt on the main control terminal and suggest that the doctor perform a supplementary operation on the area. At the same time, the integration module transmits the operation sequence number and the remaining time data to the main control terminal. The doctor can view the progress bar and time estimate of each stage through the touch screen, such as "S003 remaining time: 3 minutes, 60% completed".

[0118] The system's assessment of operational integrity also includes analysis of instrument movement patterns. For example, if an abnormal increase in the instrument movement rate is detected during the hemostasis stage (such as a sudden increase from the normal 2mm / s to 8mm / s), it may indicate that the doctor is operating too hastily and there is a risk of incomplete hemostasis. The system will issue a warning through a sound prompt or a flashing screen. In addition, the integration module automatically generates an operational integrity assessment report, which includes: division of operation stages, time spent in each stage, average movement rate, predicted value of remaining time, boundary overlap statistics, maximum and average offset values, alarm records, etc. The report is stored in the surgical record system in PDF format and can be used for postoperative medical record analysis or medical quality control.

[0119] In practice, the integration module communicates data with the positioning module, analysis module, and correction module. The positioning module provides instrument location and tissue coordinate data, the analysis module outputs the measurement area boundaries and path planning results, and the correction module provides feedback on the instrument's corrected trajectory. These data serve as the basis for evaluation. For example, when evaluating the accuracy of a particular operation, the integration module retrieves the preoperative boundary data generated by the analysis module, the actual instrument movement trajectory recorded by the positioning module, and the correction instruction records from the correction module to comprehensively determine whether the operation strictly adhered to the preset path and the effectiveness of the corrective measures.

[0120] Taking cholecystectomy as an example, after the cystic duct division operation is completed, the system performs a boundary coincidence test on the edge of the divided cystic duct. The preoperatively planned division boundary was 0.8mm from the common bile duct. Postoperative testing revealed an average offset of 0.2mm and a maximum offset of 0.5mm, both within the allowable tolerance of 0.6mm, confirming that the operation was complete. If an area is detected with an offset of 0.7mm, exceeding the tolerance, the system generates a verification report containing the specific location, prompting the physician to check whether there is a risk of accidental damage to the area.

[0121] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "includes," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus.

[0122] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A laparoscopic-digestive-endoscopic combined positioning and measurement method, characterized in that: The following steps are involved: Positioning mark initialization step: setting a first positioning mark in the surgical area, obtaining the initial position of the target organ, setting a second positioning mark of the target organ, and spatially matching the first positioning mark and the second positioning mark through optical reflection signals; Image feature analysis step: collecting endoscopic imaging data, performing noise reduction processing on the imaging data to obtain a first image, extracting morphological features of the first image, determining tissue status based on the morphological features, and generating a measurement area boundary based on the tissue status; Instrument path planning step: setting a third positioning marker of the endoscopic device, associating the third positioning marker with the second positioning marker through spatial coordinates, mapping the measurement area boundary to the endoscope field of view, obtaining anatomical path information, and planning the operation path according to the anatomical path information; Dynamic positioning calibration step: acquiring the end position of the instrument in real time, comparing the end position with the operation path to obtain a first deviation value, and triggering a positioning correction instruction according to the first deviation value; Operation integrity assessment steps: Integrate the instrument movement rate and the operation sequence number, calculate the remaining time of the operation phase based on the movement rate, transmit the remaining time and the operation sequence number to the main control terminal, perform accuracy verification on the completed area according to the preset ratio, generate a verification report and evaluate the operation integrity.

2. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 1, characterized in that: The positioning mark initialization step further includes: Acquire the surgical area in the patient's three-dimensional reconstructed model and set a first positioning mark, wherein the mark is composed of a random sequence and a first feature code, wherein the random sequence is created by a coordinate generator, and the feature code is a superposition value of the horizontal and vertical coordinates of the surgical area; Identify the anatomical structure of the target organ and set a second positioning marker, wherein the marker consists of a random sequence and a second feature code, the random sequence is created by a coordinate generator, and the feature code consists of an organ type code and a spatial orientation code. The first positioning marker and the second positioning marker are spatially matched using infrared light waves.

3. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 1, characterized in that: The image feature analysis step further includes: Acquiring endoscopic imaging data, and performing frame sequence reassembly processing on the data to obtain a first image, wherein the processing includes motion compensation and edge sharpening; Extracting morphological features of the first image, the morphological features including surface texture features and contour curvature features, the texture features including smoothness, wrinkle density, and blood vessel distribution density, and the curvature features including convex height and concave depth; The tissue status levels are divided according to morphological characteristics, and the topological boundaries of the measurement area are generated based on the status levels.

4. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 3, characterized in that: The judgment logic of the organizational status includes: Call the tissue feature database, input the organ type code, and obtain the corresponding standard texture feature and curvature feature range; Extract the texture features and curvature features of the current image and quantify them into a feature data set; Calculate the coverage of the feature data set and the standard feature range, select the feature combination with the highest coverage, and use the tissue state corresponding to the feature combination as the current judgment result; The area determined to be in a normal state is marked as an operable area.

5. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 1, characterized in that: The instrument path planning step further includes: Setting a third positioning mark of the endoscope device, the mark is composed of a random sequence and a third characteristic code, the random sequence is created by a coordinate generator, and the third characteristic code is a combination of the device model and the channel number; Project the measurement area boundary onto the endoscopic imaging plane to obtain the coordinate set of key points of the anatomical path; Generate the optimal operation path for the instrument based on the set of key point coordinates.

6. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 5, characterized in that: The planning logic of the operation path includes: Establish a three-dimensional space coordinate system and calculate the path curvature and device freedom parameters between key points; Evaluate the path safety factor, which is determined by the path curvature, instrument extension angle, and tissue elastic modulus; Path segments with safety factors higher than a threshold are selected and connected to form a continuous operation path.

7. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 1, characterized in that: The dynamic positioning calibration step further comprises: Collect the three-dimensional coordinate data of the instrument end in real time and compare it with the node coordinates of the operation path point by point; When the coordinate offset exceeds the preset tolerance range, a positioning instruction including the correction direction and correction angle is generated; Continue to update the correction command until the end coordinate returns to within the tolerance range.

8. The laparoscopic-digestive-endoscopic combined positioning and measurement method according to claim 1, characterized in that: The operational integrity assessment step further comprises: Obtain the movement trajectory data of the instrument end and calculate the displacement change per unit time; The remaining time of the stage is calculated by combining the remaining path length and the displacement change; Perform boundary coincidence check on completed areas and generate a verification report including offset statistics.

9. A laparoscopic-digestive-endoscopic combined positioning and measurement system, used to implement a laparoscopic-digestive-endoscopic combined positioning and measurement method according to any one of claims 1 to 8, characterized in that: Including positioning module, analysis module, correction module and integration module; The positioning module is used to set a first positioning mark, a second positioning mark and a third positioning mark, and collect target organ position data and endoscopic imaging data; The analysis module is used to process imaging data and extract morphological features, generate measurement area boundaries and operation paths, and calculate the instrument end offset in real time; The correction module is used to generate a spatial coordinate correction instruction according to the offset and control the execution mechanism; The integration module is used to calculate the remaining time of the operation phase, perform accuracy verification and generate an operation integrity assessment report.

10. The laparoscopic-digestive-endoscopic combined positioning and measurement system according to claim 9, characterized in that: The positioning module includes a multispectral imaging unit and a coordinate calibration unit, the analysis module includes a feature quantization processor and a path planning engine, the correction module includes a motion trajectory analyzer and an instruction generator, and the integration module includes a timing prediction model and a topology verification algorithm.

Images