Ablation needle pose monitoring methods, devices, media, and products

Abstract

The application discloses an ablation needle posture monitoring method, device, medium and product, and the method comprises the steps of establishing a three-dimensional anatomical model according to preoperative images; obtaining inertia parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data and physical model data of an ablation needle; inputting the inertia parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data and physical model data into a pre-trained posture prediction model to obtain predicted posture information of the ablation needle in a positioning coordinate system, wherein the predicted posture information comprises the position, direction and needle body shape of the ablation needle; determining a spatial registration relationship between the positioning coordinate system and an image coordinate system corresponding to the three-dimensional anatomical model; converting the predicted posture information to the image coordinate system according to the spatial registration relationship and fusing the predicted posture information with the three-dimensional anatomical model to determine the positional relationship of the ablation needle relative to a target point and a dangerous area, and to visually display the posture of the ablation needle in the tissue.

Description

Technical Field

[0001] This application relates to the field of medical device monitoring technology, and in particular to a method, device, medium and product for monitoring the posture of an ablation needle. Background Technology

[0002] Ablation therapy is a treatment method that involves inserting an ablation needle into the target tissue and applying ablation energy to the target tissue, causing coagulative necrosis. During the needle insertion and ablation process, it is necessary to determine the position and orientation of the ablation needle within the patient's body, and to assess the positional relationship between the ablation needle and the ablation target, as well as surrounding dangerous areas such as blood vessels and nerves.

[0003] Currently, the puncture process of ablation needles is usually observed and guided by preoperative imaging, intraoperative imaging, or positioning equipment. However, preoperative imaging is mainly used to display the patient's anatomical structure and determine the ablation target, and it is difficult to directly reflect the actual posture of the ablation needle after it enters the tissue; the images obtained during the operation can usually only show the position of the ablation needle at a specific moment or in a specific section, and it is difficult to continuously and completely represent the position, orientation, and morphology of the ablation needle in the tissue.

[0004] After the ablation needle enters the tissue, it may change direction or bend due to tissue resistance, the force applied to the needle tip, and the different characteristics of the tissue. It is difficult to accurately reflect the morphology of the needle inside the tissue based solely on the external positioning results of the ablation needle; it is also difficult to determine the overall position, orientation, and morphological changes of the ablation needle simultaneously using only a single type of sensor data.

[0005] Therefore, existing methods for monitoring the posture of ablation needles are insufficient to comprehensively utilize the motion state, stress state, tissue state, spatial location, and physical characteristics of the ablation needle to continuously determine the position, orientation, and shape of the ablation needle within the tissue. They are also insufficient to accurately integrate the actual posture of the ablation needle with the three-dimensional anatomical model established from preoperative images. Summary of the Invention

[0006] One objective of this application is to provide an ablation needle posture monitoring method, device, medium, and product, at least to solve the problem that existing ablation needle posture monitoring methods are unable to comprehensively reflect the position, orientation, and morphology of the ablation needle within the tissue.

[0007] To achieve the above objectives, some embodiments of this application provide the following aspects: This application provides a method for monitoring the posture of an ablation needle, the method comprising: A three-dimensional anatomical model containing target points and danger zones is created based on preoperative images; The inertial parameters, tip pressure data, impedance data, positioning data, insertion depth data, and physical model data of the ablation needle are obtained. The inertial parameters include the angular velocity data and acceleration data of the ablation needle. The inertial parameters, the needle tip pressure data, the impedance data, the positioning data, the needle insertion depth data, and the physical model data are input into a pre-trained attitude prediction model to obtain the predicted attitude information of the ablation needle in the positioning coordinate system. The predicted attitude information includes the position, orientation, and shape of the ablation needle. Determine the spatial registration relationship between the positioning coordinate system and the image coordinate system corresponding to the three-dimensional anatomical model; Based on the spatial registration relationship, the predicted pose information is transformed into the image coordinate system and fused with the three-dimensional anatomical model to determine the positional relationship of the ablation needle relative to the target point and the danger zone, and the pose of the ablation needle in the tissue is visualized.

[0008] Secondly, some embodiments of this application also provide an electronic device, the electronic device comprising: one or more processors; and a memory storing computer program instructions, which, when executed, cause the processor to perform the steps of the method described above.

[0009] Thirdly, some embodiments of this application also provide a computer-readable medium having computer program instructions stored thereon, which can be executed by a processor to implement the steps of the method described above.

[0010] Fourthly, some embodiments of this application also provide a computer program product, including a computer program / instructions that, when executed by a processor, implement the steps of the method described above.

[0011] Compared with related technologies, the solution provided in this application integrates the inertial parameters, tip pressure data, impedance data, positioning data, insertion depth data, and physical model data of the ablation needle into a pre-trained attitude prediction model. This allows the attitude prediction model to comprehensively consider the motion state of the ablation needle, the force state of the tip, the state of the tissue, its spatial location, the depth of insertion into the tissue, and the physical characteristics of the needle body to predict the position, orientation, and morphology of the ablation needle within the tissue. Compared to determining the attitude of the ablation needle based on data from a single sensor, this method improves the accuracy of monitoring the overall attitude and morphology of the ablation needle. Furthermore, by determining the spatial registration relationship between the positioning coordinate system and the image coordinate system, the predicted attitude information is transformed into the image coordinate system and fused with the preoperative three-dimensional anatomical model. This allows the current attitude, historical trajectory, target point, and danger zone of the ablation needle to be displayed together in the three-dimensional anatomical model, facilitating the determination of the positional relationship between the ablation needle and the target point and danger zone. Attached Figure Description

[0012] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0013] Figure 1 A flowchart illustrating an ablation needle posture monitoring method provided as an exemplary embodiment of this application; Figure 2 An exemplary structural diagram of the electronic device provided for some embodiments of this application. Detailed Implementation

[0014] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0015] The ablation needle described in this application can be a radiofrequency ablation surgical electrode used in nerve radiofrequency ablation. The ablation needle includes a handle, a needle body, and a needle tip located at the distal end of the needle body. The needle tip may include an effective working section for delivering radiofrequency energy to the tissue. The ablation needle is equipped with an inertial sensing unit for acquiring inertial data and a pressure sensor for acquiring needle tip pressure data. The inertial sensing unit may be located inside or outside the handle, and the pressure sensor may be located at the needle tip.

[0016] Figure 1 An exemplary flowchart of an ablation needle posture monitoring method provided for an exemplary embodiment of this application is provided, the method including the following steps.

[0017] S101. Establish a three-dimensional anatomical model containing target points and danger zones based on preoperative images.

[0018] Specifically, before performing the ablation procedure, the surgical area is scanned using a computed tomography (CT) scanner or magnetic resonance imaging (MRI) device to obtain corresponding preoperative tomographic images. These preoperative images may include bones, muscles, blood vessels, nerves, and other anatomical structures relevant to the ablation procedure.

[0019] Deep learning-based image segmentation networks can be used to automatically or semi-automatically segment different anatomical structures in preoperative images. For example, U-Net networks or variants of U-Net networks can be used to identify bones, muscles, blood vessels, and nerves, and the corresponding segmentation results can be generated after confirmation or adjustment by the doctor. The segmented two-dimensional image slices are then reconstructed into three-dimensional voxels to obtain a three-dimensional anatomical model corresponding to the patient's surgical area.

[0020] Based on the treatment plan, the target points or target areas for ablation are determined in the three-dimensional anatomical model, and non-target nerves, blood vessels, and other tissue areas that should be avoided during puncture are identified as danger zones. Surface markers or bony landmarks that can be used for spatial registration can also be marked in the three-dimensional anatomical model.

[0021] S102. Obtain the inertial parameters, tip pressure data, impedance data, positioning data, insertion depth data, and physical model data of the ablation needle.

[0022] Specifically, a miniature inertial measurement unit (MMU) is integrated inside or outside the handle of the radiofrequency surgical electrode. The MMU includes at least a three-axis gyroscope and a three-axis accelerometer. The gyroscope detects the angular velocity data when the radiofrequency surgical electrode rotates around three coordinate axes, and the accelerometer detects the acceleration data when the radiofrequency surgical electrode moves along the three coordinate axes. The angular velocity data and the acceleration data together constitute the inertial parameters of the ablation needle.

[0023] The miniature inertial measurement unit may also include a triaxial magnetometer. The triaxial magnetometer is used to detect geomagnetic direction data and utilizes this data to assist in correcting the heading angle of the ablation needle, thereby enhancing the stability of the heading angle calculation. In one specific embodiment, the data sampling frequency of the miniature inertial measurement unit is no less than 200Hz, enabling it to acquire dynamic changes during rapid puncture or rapid orientation adjustment of the radiofrequency surgical electrode.

[0024] A miniature pressure sensor can be integrated at the distal end of the tip of a radiofrequency surgical electrode. This miniature pressure sensor can be a microelectromechanical system (MEMS) piezoresistive sensor or a capacitive sensor, used to detect the axial and lateral forces acting on the tip in real time when the tip contacts the tissue. The axial force is the force acting on the tip along the direction of needle extension, while the lateral force is the force acting on the tip perpendicular to the direction of needle extension. The data output by the miniature pressure sensor forms the tip pressure data.

[0025] When a needle tip contacts and passes through different tissue layers such as fascia, muscle, or bone, the magnitude and variation of axial and lateral forces may differ. Therefore, continuous changes in needle tip pressure data can be used to determine whether the needle tip is in contact with tissue, whether it has passed through the corresponding tissue layer, and whether it is experiencing significant lateral tissue resistance. Needle tip pressure data can also serve as an auxiliary constraint for attitude prediction, correcting the cumulative drift that occurs when relying solely on continuous integration of inertial parameters.

[0026] Impedance data can be obtained through the radiofrequency host or an impedance detection circuit connected to the radiofrequency surgical electrode, and is used to characterize the electrical properties of the tissue in contact with the needle tip. During puncture, impedance data can be combined with needle tip pressure data to help determine the state of the tissue in contact with the needle tip. When both needle tip pressure data and impedance data change simultaneously, the synchronous change of both can be used as an auxiliary basis for judging tissue contact or tissue layer penetration.

[0027] Positioning data can be obtained using optical or magnetic positioning devices. Optical positioning devices can detect optical marks set on the handle of the ablation needle, while magnetic positioning devices can detect magnetic positioning components set on the ablation needle. Positioning data is used to represent the spatial position of the reference point at the handle end of the ablation needle, the needle tip, or other positioning objects in the positioning coordinate system.

[0028] In one embodiment, the moment when the needle tip begins to contact the tissue is determined based on needle tip pressure data, and the position of the ablation needle reference point obtained at that moment is determined as the initial insertion position. During subsequent punctures, the displacement of the ablation needle reference point relative to the initial insertion position is determined based on positioning data, the insertion direction of the ablation needle is determined based on inertial parameters, and the insertion depth data is determined by combining the displacement, the insertion direction, and the structural parameters of the ablation needle. The insertion depth data can be used to represent the depth to which the ablation needle enters the tissue and to provide auxiliary constraints on the position of the ablation needle along the insertion direction.

[0029] In another embodiment, a depth detection component can be provided on the handle or puncture guide of the radiofrequency surgical electrode to obtain needle insertion depth data. In practical applications, the method of obtaining needle insertion depth data can be selected according to the structure of the radiofrequency surgical electrode and the surgical procedure.

[0030] Before collecting the aforementioned real-time data, a digital model of the currently used radiofrequency surgical electrode can be created. Using computer-aided design modeling, the geometric dimensional parameters of the electrode's handle, needle body, and exposed tip are extracted. These parameters may include needle body length, needle body diameter, exposed tip length, and the positional relationship between the handle reference point and the needle tip. Further, the material properties and bending stiffness of the needle body are obtained to form physical model data that matches the current ablation needle.

[0031] A rigid body kinematic model can be established based on the geometric dimensions of the ablation needle, and a reference point at the handle end, the needle tip, and key points of the needle body can be set in the digital needle body model. The rigid body kinematic model provides the geometric basis for subsequently determining the spatial coordinates of the needle tip and key points of the needle body based on the position and orientation of the reference point.

[0032] S103. Input the inertial parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data into the pre-trained attitude prediction model to obtain the predicted attitude information of the ablation needle in the positioning coordinate system.

[0033] Before inputting the attitude prediction model, data from different sources are synchronized in time and preprocessed. Preprocessing of inertial parameters may include filtering and denoising, zero-bias compensation, and outlier removal; preprocessing of needle tip pressure data, impedance data, positioning data, and needle insertion depth data may include filtering and denoising, continuity checks, and outlier removal. After preprocessing, the data are correlated according to their acquisition time.

[0034] The attitude prediction model can be a time series regression model based on a long short-term memory network or a converter architecture. The model determines the motion state and directional changes of the ablation needle based on inertial parameters, the force state of the needle tip based on needle tip pressure data, information related to the tissue in contact with the needle tip based on impedance data, the spatial position constraints of the ablation needle based on positioning data, and the insertion distance of the ablation needle along the insertion direction based on insertion depth data. Combined with the needle structure and material properties characterized by the physical model data, the model obtains the predicted attitude information of the ablation needle in the positioning coordinate system.

[0035] Predicted attitude information includes the position, orientation, and shape of the ablation needle. Position can include the position of the needle tip, the position of the reference point at the handle end, or the positions of multiple key points on the needle body; orientation can be represented by attitude quaternions, rotation matrices, or corresponding attitude angles; needle shape can be represented by the needle centerline, the coordinate sequence of key points on the needle body, or morphological parameters used to adjust the digital needle model.

[0036] S104. Determine the spatial registration relationship between the positioning coordinate system and the image coordinate system corresponding to the three-dimensional anatomical model.

[0037] The positioning data is represented in an optical or magnetic positioning coordinate system, while the target points and danger zones in the 3D anatomical model are represented in an image coordinate system. Therefore, it is necessary to establish a spatial registration relationship between the positioning coordinate system and the image coordinate system using multiple positioning reference points.

[0038] The positioning reference points can be surface markers set on the patient's body surface before surgery and identifiable in preoperative images, or bony landmarks whose positions can be determined in both preoperative images and the surgical site. The positions of each positioning reference point in the positioning coordinate system and the image coordinate system are obtained separately. The rigid transformation relationship between the two coordinate systems is calculated based on the corresponding positioning reference points, and this rigid transformation relationship is used as the spatial registration relationship.

[0039] S105. Based on the spatial registration relationship, the predicted pose information is converted to the image coordinate system and fused with the three-dimensional anatomical model to determine the positional relationship of the ablation needle relative to the target point and the danger zone, and then visualized.

[0040] Based on the spatial registration relationship, the position, orientation, and shape of the ablation needle are transformed from the positioning coordinate system to the image coordinate system. Based on the transformed predicted attitude information, the position and orientation of the digital needle model in the image coordinate system are updated, and the spatial geometry of the digital needle model is adjusted according to the predicted needle shape.

[0041] The updated digital needle model is overlaid on the 3D anatomical model to present the real-time position, insertion depth, deflection angle, needle tip coordinates, and needle morphology of the ablation needle within the tissue. The distance and orientation of the needle tip relative to the target point are determined based on the position of the needle tip and the target point. The distance and orientation of the ablation needle relative to the danger zone are determined based on the spatial relationship between the key points of the needle body and the danger zone.

[0042] The fused results can be displayed on an operating room monitor or on a head-mounted augmented reality device for real-time observation by the surgeon. As new data is collected, the predicted posture information and the display status of the digital needle model are continuously updated, thus forming a real-time three-dimensional visualization navigation result of the ablation needle in the body.

[0043] This embodiment integrates inertial and pressure detection components onto the ablation needle, and combines impedance, positioning, needle depth, and needle physical characteristics for attitude prediction. This reduces the bias caused by relying solely on the doctor's touch and experience to judge the position of the ablation needle. By fusing the predicted attitude information with the preoperative three-dimensional anatomical model, the number of times the position is verified by repeated image scanning during the operation can be reduced, and spatial position reference can be provided to determine whether the needle tip is close to the target point or danger zone.

[0044] In one embodiment, the training process of the pose prediction model includes: While the simulated ablation needle is in different motion and stress states, the inertial parameters, needle tip pressure data, impedance data, positioning data and needle insertion depth data of the simulated ablation needle are collected simultaneously, and physical model data matching the simulated ablation needle and actual attitude information obtained by the micro electromagnetic sensor integrated into the simulated ablation needle are acquired. Using the inertial parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data as training samples, and the actual posture information as training labels, the model to be trained is trained to obtain the posture prediction model.

[0045] Specifically, the simulated ablation needle has the same length, diameter, material properties, and bending stiffness as the actual ablation needle. The simulated ablation needle has a miniature inertial measurement unit inside or outside the handle, a miniature pressure sensor at the tip, and impedance detection components, positioning components, and corresponding detection structures for obtaining needle insertion depth data.

[0046] Training samples can be obtained through in vitro tissue simulation experiments and in vivo experiments. In vitro tissue simulation experiments are used to generate training data under different tissue resistances, different needle insertion depths, and different needle bending states, while in vivo experiments are used to generate training data that closely approximates the actual tissue environment.

[0047] During the training data acquisition process, the simulated ablation needle is controlled to perform different operations such as forward movement, backward movement, rotation, tilting, and bending under lateral force. Simultaneously, angular velocity data, acceleration data, needle tip pressure data, impedance data, positioning data, and insertion depth data of the simulated ablation needle are acquired, and physical model data matching the simulated ablation needle is obtained.

[0048] One or more miniature electromagnetic sensors can be integrated at the tip or key locations of the simulated ablation needle. These sensors are used to obtain the actual position and orientation of the simulated ablation needle. With multiple miniature electromagnetic sensors, a sequence of actual coordinates for key points on the needle body can be formed based on the positions of each sensor, thus representing the actual shape of the simulated ablation needle. The actual position, orientation, and shape obtained from the miniature electromagnetic sensors together constitute the actual attitude information.

[0049] Before generating training samples, inertial parameters, needle tip pressure data, impedance data, positioning data, and needle insertion depth data are synchronized in time and preprocessed. Preprocessing may include filtering and noise reduction, inertial bias compensation, outlier removal, and data amplitude normalization. Based on the acquisition time of each data point, data corresponding to the same motion and force process are combined to form training samples.

[0050] Inertial parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data are used as inputs to the attitude prediction model, while the actual attitude information obtained from the miniature electromagnetic sensor is used as training labels. The data format of the training labels is consistent with the output data format of the attitude prediction model, so that the predicted position, predicted direction, and predicted needle shape output during the training phase can be compared with the actual position, actual direction, and actual needle shape, respectively.

[0051] Attitude prediction models can employ either long short-term memory networks or time-series regression models with a converter architecture. Long short-term memory networks can learn the relationship between ablation needle attitude and time based on data from multiple consecutive acquisition times; converter architectures can extract time-series features for attitude prediction based on the connections between multi-source data at different acquisition times.

[0052] During training, training samples are input into the model to be trained to obtain the predicted posture information of the simulated ablation needle. The model parameters are adjusted based on the difference between the predicted posture information and the actual posture information. Through training samples with multiple needle insertion depths, multiple needle insertion directions, and multiple stress states, the model learns the correspondence between habitual parameters, pressure changes, tissue impedance, positioning results, needle insertion depth, and needle physical properties and the actual posture.

[0053] After training, the posture prediction model is deployed in the ablation needle posture monitoring device. During actual surgery, the miniature electromagnetic sensors used to generate training labels are no longer necessary detection components; the pre-trained posture prediction model can be used to process the multi-source data collected in practice.

[0054] This embodiment uses actual attitude information obtained from a miniature electromagnetic sensor as training labels, enabling the attitude prediction model to learn the correspondence between multi-source data and the actual position, orientation, and shape of the ablation needle. Needle insertion depth data is input during both the training and actual use phases, avoiding inconsistencies between the model training input and the actual prediction input.

[0055] In one embodiment, before inputting the inertial parameters into the attitude prediction model, the angular velocity data and the acceleration data are fused using an extended Kalman filter algorithm or an unscented Kalman filter algorithm. The attitude direction of the ablation needle is recursively updated based on the angular velocity data, and the attitude direction is corrected based on the gravity component in the acceleration data to obtain inertial attitude features. The inertial attitude characteristics, the needle tip pressure data, the impedance data, the positioning data, the needle insertion depth data, and the physical model data are input into the attitude prediction model.

[0056] Specifically, a miniature inertial measurement unit (IMU) is integrated inside or outside the handle of the radiofrequency surgical electrode. Placing the miniature inertial measurement unit at the end of the handle reduces the impact of the sensor on the fine-diameter needle and the effective working electrode segment, and allows the physician to hold and move the radiofrequency surgical electrode in the same way as before.

[0057] The miniature inertial measurement unit (MMU) includes at least a three-axis gyroscope and a three-axis accelerometer. The three-axis gyroscope detects the angular velocity data of the ablation needle as it rotates around the three coordinate axes, and the three-axis accelerometer detects the acceleration data of the ablation needle as it moves along the three coordinate axes. The MMU may also include a three-axis magnetometer, which is used to detect geomagnetic direction data to help correct the heading angle of the ablation needle.

[0058] In one specific embodiment, the data sampling frequency of the miniature inertial measurement unit is no less than 200Hz to collect the dynamic changes of the ablation needle during rapid puncture, rapid rotation, or short-term adjustment of the needle insertion direction. The acquisition time for each set of angular velocity and acceleration data is recorded for subsequent time correlation with needle tip pressure data, impedance data, positioning data, and needle depth data.

[0059] Before starting attitude monitoring, the ablation needle is placed in a preset reference state, and angular velocity and acceleration data are collected under this state. The zero bias of the three-axis gyroscope and accelerometer is determined based on the collected results. During actual data acquisition, the angular velocity and acceleration data are filtered for noise reduction and zero bias compensation is performed. Abnormal data is identified based on the continuous changes in data between adjacent acquisition times. Data that significantly exceeds the normal range or is inconsistent with the current motion state of the ablation needle is removed from the current fusion process.

[0060] In one embodiment, attitude quaternions are used to represent the attitude orientation of the ablation needle. The attitude quaternion corresponding to the previous acquisition moment is used as the basis for attitude recursion at the current acquisition moment. The rotational change of the ablation needle relative to the previous acquisition moment is determined based on the angular velocity data at the current acquisition moment, and the attitude quaternion is updated recursively.

[0061] Since continuous integration of angular velocity data can easily lead to cumulative drift, the component corresponding to the direction of gravity is determined from the acceleration data at the current acquisition moment. This gravity direction is then used to correct the recursively derived attitude quaternions to suppress pitch and roll drift. In the case where the miniature inertial measurement unit includes a triaxial magnetometer, the yaw direction can also be further corrected based on the geomagnetic direction data detected by the triaxial magnetometer.

[0062] When using the Extended Kalman Filter (EKF) algorithm, the attitude state recursively derived from angular velocity data is used as the state prediction, and the direction obtained from the gravitational component in the acceleration data is used as the observation information. The state prediction is then corrected based on the observation information. When using the Unscented Kalman Filter (UKF) algorithm, the nonlinear relationship in the attitude state change process is handled through sampling point propagation, and the propagated attitude state is corrected based on the acceleration data. In practical applications, either the EKF algorithm or the UKF algorithm can be selected based on the attitude change characteristics and data processing capabilities.

[0063] The attitude quaternion, after recursive updating and gravity orientation correction, is used as the inertial attitude feature. The inertial attitude feature is used to characterize the attitude orientation of the ablation needle after fusion processing, and is not the final predicted attitude information output by the attitude prediction model.

[0064] The inertial attitude characteristics, needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data are all input into the attitude prediction model. The attitude prediction model obtains the basic attitude orientation of the ablation needle based on the inertial attitude characteristics, the force state of the needle tip based on the needle tip pressure data, information related to the tissue contacted by the needle tip based on the impedance data, the spatial position constraints of the ablation needle based on the positioning data, the depth constraints of the ablation needle entering the tissue based on the needle insertion depth data, and the geometric structure and stress deformation characteristics of the ablation needle based on the physical model data. Thus, the position, orientation, and morphology of the ablation needle are obtained.

[0065] In this embodiment, the attitude orientation of the ablation needle is recursively updated using angular velocity data, and the recursive result is corrected using the gravity component in the acceleration data. This suppresses pitch and roll drift caused by continuous integration of only angular velocity data. The fused inertial attitude features are then input into the attitude prediction model along with needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data. This allows for further constraints on the inertial attitude features using data from different sources, thereby improving the stability of the predicted position, orientation, and morphology of the ablation needle.

[0066] In one embodiment, the inertial parameters, the needle tip pressure data, the impedance data, the positioning data, and the needle insertion depth data are time-correlated according to the acquisition time to form a multi-source time series; The multi-source time series and physical model data matching the current ablation needle are input into the attitude prediction model, and the attitude prediction model predicts the attitude of the ablation needle at the current moment based on the changes of various data over time.

[0067] Specifically, miniature inertial measurement units, miniature pressure sensors, impedance detection circuits, and positioning devices may have different data sampling frequencies and data transmission delays, and the update time of needle insertion depth data may also differ from the acquisition time of inertial parameters. Therefore, the corresponding acquisition time should be recorded separately when acquiring various data.

[0068] A unified time reference can be provided by the ablation needle attitude monitoring device, enabling the miniature inertial measurement unit, miniature pressure sensor, impedance detection circuit, and positioning device to record the acquisition time according to a unified clock. For detection components that cannot directly use a unified clock, the reception time can be recorded when the corresponding data enters the ablation needle attitude monitoring device, and the reception time can be corrected according to a predetermined data transmission delay.

[0069] Before time mapping, data from different sources were preprocessed. Inertial parameters underwent filtering, noise reduction, zero-bias compensation, and outlier removal; needle tip pressure data underwent zero-point calibration, noise filtering, and abrupt data identification; impedance data underwent interference filtering and outlier impedance value removal; positioning data underwent positioning loss identification and position jump removal; and needle insertion depth data underwent zero-point calibration and continuity checks. The acquisition time corresponding to each data point was retained during preprocessing.

[0070] Based on the target acquisition time predicted from the current attitude, data corresponding to the target acquisition time is determined from various data sources. For data with the same acquisition time, a direct correspondence is established; for data with slightly different acquisition times, the data with the smallest time difference from the target acquisition time can be selected, or the data corresponding to the target acquisition time can be determined based on data located before and after the target acquisition time.

[0071] For example, the sampling frequency of inertial parameters can be no less than 200Hz, while the sampling frequencies of needle tip pressure data, impedance data, and positioning data can be lower than the sampling frequency of inertial parameters. In this case, inertial parameters that are acquired at the same or similar times as the acquisition times of the needle tip pressure data, impedance data, and positioning data can be determined separately to avoid combining data from different puncture states into the same set of model inputs.

[0072] When the inertial parameters have been fused according to Example 3, the acquisition time corresponding to the inertial attitude features is retained, and a correspondence is established between the inertial attitude features and the needle tip pressure data, impedance data, positioning data, and needle insertion depth data based on the acquisition time. At this time, the inertial parameters in the multi-source time series can be replaced by the inertial attitude features.

[0073] The data corresponding to the completed time are arranged in chronological order of acquisition time to form a multi-source time series. The multi-source time series can include data from the current acquisition time and multiple consecutive acquisition times preceding the current acquisition time, used to represent the changes in the ablation needle movement state, needle tip force state, tissue impedance, spatial position, and needle insertion depth over time.

[0074] For example, as the needle tip contacts the tissue and continues to move into the tissue, the needle tip pressure data may change first, followed by a continuous increase in the needle depth data. The impedance data changes with the tissue the needle tip contacts, while the positioning data reflects the spatial movement of the ablation needle reference point. The posture prediction model can distinguish different puncture states, such as the needle tip contacting the tissue, passing through the tissue layer, and continuing to move into the tissue, based on the sequential relationship of the above data changes over time.

[0075] Physical model data, used to characterize the length, diameter, material properties, and bending stiffness of the current ablation needle, typically does not change with acquisition time during the same puncture. Therefore, matching physical model data is determined based on the currently used ablation needle, and this physical model data is input into the attitude prediction model along with the multi-source time series data, without requiring the physical model data to be re-acquired at each acquisition time.

[0076] Attitude prediction models can employ long short-term memory networks or time-series regression models with converter architectures. The model determines the motion and rotation of the ablation needle based on changes in inertial parameters over time, the force applied to the needle tip and the tissue penetration process based on changes in needle tip pressure data over time, the changes in tissue electrical properties based on changes in impedance data over time, the spatial movement of the ablation needle based on changes in positioning data over time, and the process of the ablation needle entering the tissue based on changes in needle depth data over time. From these, the model outputs the predicted attitude information at the current acquisition moment.

[0077] This embodiment, by mapping data from different sources and sampling frequencies to a specific time based on the acquisition time, ensures that the inertial parameters, needle tip pressure data, impedance data, positioning data, and needle insertion depth data input to the attitude prediction model correspond to the same puncture stage, avoiding incorrect combinations of data from different attitude states. By forming a multi-source time series from data collected at multiple consecutive times, the attitude prediction model can combine the dynamic changes of various data to predict the current attitude, thereby reducing the errors caused by predictions based solely on data from a single acquisition time.

[0078] In one embodiment, the physical model data is determined based on the length, diameter, material properties, and bending stiffness of the current ablation needle; The attitude prediction model determines the current needle tip stress state of the ablation needle based on the needle tip pressure data, and predicts the needle body morphology of the current ablation needle under the needle tip stress state by combining the needle body physical characteristics characterized by the physical model data.

[0079] Specifically, before using the ablation needle, a computer-aided design modeling method is used to digitally model the radiofrequency surgical electrode. The established digital model can include the electrode handle, the needle body, and the exposed end located at the distal end of the needle body. The exposed end is the effective working electrode segment for applying radiofrequency energy to the target tissue.

[0080] During the digital modeling process, the needle length, needle diameter, exposed end length, positional relationship between the handle reference point and the needle tip, needle material properties, and bending stiffness of the current ablation needle are obtained, and physical model data is formed based on the above parameters. The handle reference point, needle tip, and multiple key needle points distributed along the needle extension direction can also be set in the digital needle model.

[0081] The length and diameter of the ablation needle are used to characterize the needle's geometry, while material properties and bending stiffness are used to characterize the physical properties of the needle's shape changes under external forces. Different models or specifications of ablation needles have corresponding physical model data. When starting attitude monitoring, the physical model data matching the current ablation needle can be selected based on its model, specifications, or identification information.

[0082] A miniature pressure sensor is integrated at the distal end of the tip of a radiofrequency surgical electrode. This miniature pressure sensor can be a microelectromechanical system (MEMS) piezoresistive sensor or a capacitive sensor, used to detect the axial and lateral forces acting on the needle tip in real time when it contacts the tissue. The axial force is the force acting on the needle tip along the needle's extension direction, and the lateral force is the force acting on the needle tip perpendicular to the needle's extension direction.

[0083] During puncture, as the needle tip contacts and passes through different tissue layers such as fascia, muscle, or bone, the magnitude and changes of axial and lateral forces change accordingly. The posture prediction model determines the current force state of the needle tip based on the magnitude of the needle tip pressure data and its changes over time.

[0084] An increase in axial force indicates that the needle tip experiences tissue resistance along the insertion direction. When the axial force changes accordingly after the increase, this change can be used as an auxiliary criterion for judging whether the needle tip has penetrated the tissue layer. An increase in lateral force indicates that the needle tip or needle body experiences lateral tissue resistance, which may cause deflection or bending. Impedance data can be used to help determine the state of the tissue in contact with the needle tip, thus providing further reference for judging the force state of the needle tip.

[0085] The attitude prediction model combines the stress state of the needle tip with the physical model data of the current ablation needle. For ablation needles with high bending stiffness, the predicted bending degree is relatively small under the same lateral force; for ablation needles with low bending stiffness, the predicted bending degree is relatively large under the same lateral force. For ablation needles with different lengths, diameters, or material properties, the attitude prediction model determines the needle shape based on the matching physical model data.

[0086] The predicted needle morphology can be represented by the needle centerline, a sequence of spatial coordinates of multiple key needle points, or morphological parameters used to adjust the digital needle model. Based on the predicted needle morphology, the position of the centerline or key points of the digital needle model is adjusted to reflect the bending state of the ablation needle under tissue resistance.

[0087] The attitude prediction model can also combine positioning data to determine the spatial position of the reference point at the end of the ablation needle handle or other positioning objects, combine needle insertion depth data to determine the needle segment entering the tissue, and combine inertial parameters to determine the attitude orientation of the ablation needle. Based on this, the predicted needle shape is combined with the position and orientation of the ablation needle to form complete predicted attitude information.

[0088] When replacing ablation needles with different models or structures, the physical model data is redefined based on the replaced ablation needle, and a digital needle body model matching the replaced ablation needle is established or selected. The redefined physical model data is then used for subsequent needle body morphology prediction.

[0089] This embodiment determines the physical model data based on the length, diameter, material properties, and bending stiffness of the current ablation needle, and determines the needle tip stress state based on the needle tip pressure data. This allows the attitude prediction model to predict the needle body morphology based on the actual structure and stress conditions of the current ablation needle. Therefore, it reduces the deviation in needle body morphology and needle tip position caused by always treating small-diameter ablation needles as rigid straight rods, and improves the consistency between the digital needle body model and the actual physical state of the ablation needle.

[0090] In one embodiment, the positioning data is obtained through an optical positioning device or a magnetic positioning device; The positions of multiple positioning reference points in the positioning coordinate system and the image coordinate system are determined respectively. Based on the positions of the corresponding positioning reference points in the two coordinate systems, the rigid transformation relationship between the positioning coordinate system and the image coordinate system is determined, and the rigid transformation relationship is used as the spatial registration relationship.

[0091] Specifically, when using an optical positioning device, an optical mark that can be recognized by the optical positioning device can be set on the handle of the ablation needle. The optical positioning device determines the position of the reference point at the end of the ablation needle handle in the optical positioning coordinate system based on the position of the optical mark, and the optical positioning coordinate system serves as the positioning coordinate system.

[0092] When using a magnetic positioning device, a magnetic positioning component can be placed in the ablation needle handle or other locations that do not affect the transmission of radio frequency energy. The magnetic positioning device determines the position of the corresponding reference point in the magnetic positioning coordinate system based on the detection result of the magnetic positioning component in the magnetic field, and the magnetic positioning coordinate system serves as the positioning coordinate system.

[0093] Preoperative images can be obtained using computed tomography (CT) or magnetic resonance imaging (MRI). Before the preoperative scan, multiple surface markers that can be identified in the preoperative images can be placed on the patient's body surface. Alternatively, bony landmarks whose locations can be determined in both preoperative images and at the surgical site can be selected. These surface markers or bony landmarks serve as positioning reference points.

[0094] The positions of each positioning reference point in the image coordinate system are determined based on preoperative images. Before the operation begins, the positions of the same positioning reference points in the positioning coordinate system are determined using optical or magnetic positioning equipment, and a one-to-one correspondence between each positioning reference point in the two coordinate systems is established.

[0095] Multiple positioning reference points are dispersed in space to determine the directional and positional differences between the two coordinate systems through the spatial relationships between these reference points. Based on the corresponding positions of each reference point in the two coordinate systems, the rotation and translation relationships of the positioning coordinate system relative to the image coordinate system are determined, forming a rigid transformation relationship.

[0096] In one embodiment, the rigid transformation relationship includes rotation parameters and translation parameters. The rotation parameters are used to convert the spatial orientation in the positioning coordinate system to the spatial orientation in the image coordinate system, and the translation parameters are used to translate the rotated position so that the corresponding positioning reference points in the two coordinate systems coincide or meet a preset registration accuracy.

[0097] After obtaining the predicted attitude information of the ablation needle, the positions of the needle tip, the reference point at the handle end, and the key points of each needle body are transformed to the image coordinate system according to the rigid transformation relationship. For the attitude direction of the ablation needle, the transformation is performed according to the rotation relationship in the rigid transformation relationship; for the needle body shape represented by the centerline of the needle body or the coordinate sequence of key points, the corresponding spatial coordinates are transformed respectively.

[0098] The converted position, orientation, and shape of the ablation needle are aligned with the target and danger zones in the 3D anatomical model within the same image coordinate system. Based on the positions of the needle tip, key points of the needle body, the target, and the danger zones in the image coordinate system, the distance and orientation of the ablation needle relative to the target, blood vessels, nerves, and other danger zones can be determined.

[0099] When the patient's position changes or the positioning reference point moves, the position of each positioning reference point in the positioning coordinate system can be re-detected, and the spatial registration relationship can be re-determined or updated so that the predicted posture of the ablation needle maintains the corresponding spatial correspondence with the three-dimensional anatomical model.

[0100] In this embodiment, by using the corresponding positions of multiple positioning reference points in the positioning coordinate system and the image coordinate system to determine the rigid transformation relationship, the predicted posture of the ablation needle obtained by the positioning device can be transformed to the image coordinate system used by the three-dimensional anatomical model, so that the ablation needle, target point and danger area have a unified spatial coordinate basis, thereby providing conditions for determining the positional relationship between the ablation needle and the target point and danger area.

[0101] In one embodiment, the inertial parameters, needle tip pressure data, impedance data, positioning data, and needle insertion depth data are continuously acquired at different acquisition times. The data obtained at each acquisition time and the physical model data are respectively input into the attitude prediction model to obtain the predicted attitude information at each acquisition time. Based on the spatial registration relationship, the predicted pose information at each acquisition time is converted to the image coordinate system to form the historical motion trajectory of the ablation needle; The current predicted posture information of the ablation needle, the historical motion trajectory, the target point, and the danger zone are all displayed in the three-dimensional anatomical model.

[0102] Specifically, during the process of the ablation needle entering and moving within the tissue, angular velocity and acceleration data of the ablation needle are continuously acquired according to the acquisition time, and needle tip pressure, impedance, positioning, and insertion depth data are acquired simultaneously.

[0103] For each acquisition moment, the corresponding data are determined based on the acquisition time, and the corresponding data, along with the physical model data matching the current ablation needle, are input into the attitude prediction model. The attitude prediction model outputs the position, orientation, and shape of the ablation needle at each acquisition moment, forming the predicted attitude information for each acquisition moment.

[0104] Physical model data typically does not change during a single puncture. Therefore, physical model data can be pre-determined based on the current length, diameter, material properties, and bending stiffness of the ablation needle, and the same physical model data can be input into the attitude prediction model along with data obtained at each acquisition moment.

[0105] The acquisition time corresponding to each predicted attitude information is retained, and the predicted attitude information at each acquisition time is transformed from the positioning coordinate system to the image coordinate system according to the spatial registration relationship between the positioning coordinate system and the image coordinate system. The transformed predicted attitude information is arranged in chronological order of acquisition time.

[0106] In one embodiment, the needle tip position is extracted from the predicted posture information at each acquisition time, and the needle tip positions corresponding to different acquisition times are sequentially connected according to the acquisition time to form the historical motion trajectory of the needle tip. The ablation needle direction and needle body shape corresponding to each acquisition time can also be retained to represent the changes in the ablation needle's posture, orientation, and bending state during puncture.

[0107] The predicted posture information corresponding to the latest acquisition moment is determined as the current predicted posture information. Based on the current predicted posture information, the current position, orientation, and shape of the ablation needle are displayed in the 3D anatomical model, along with the current insertion depth, deflection angle, and needle tip coordinates.

[0108] The position of the needle tip in the image coordinate system is determined based on the current predicted attitude information, and the distance and orientation of the needle tip relative to the target point are determined based on the position of the needle tip and the position of the target point. The distance and orientation of the ablation needle relative to the danger zone are determined based on the spatial relationship between the needle centerline or key points of the needle body and the danger zone.

[0109] The current predicted posture information, historical movement trajectory, target point and danger zone of the ablation needle are displayed together in the three-dimensional anatomical model, allowing doctors to simultaneously observe the current position, direction and shape of the ablation needle, the path previously taken by the ablation needle, and the positional relationship between the ablation needle and the surrounding anatomical structures.

[0110] Historical movement trajectories can also be used to replay the movement of the ablation needle from its initial insertion position to its current position, assisting doctors in comparing the actual insertion path with the pre-planned path. After the ablation procedure is completed, the final position of the ablation needle, the position of the effective working electrode segment, and the historical movement trajectory can be retained, providing spatial location references for determining whether the ablation area covers the target point.

[0111] The visualization results can be displayed on an operating room monitor or on a head-mounted augmented reality device. As new multi-source data is acquired, the posture prediction model outputs new predicted posture information and uses this information to update the current display status, historical trajectory, insertion depth, deflection angle, tip coordinates, and distance and orientation of the ablation needle relative to the target and danger zone.

[0112] This embodiment continuously acquires predicted posture information at different acquisition times and transforms each predicted posture information into an image coordinate system, thus forming the historical movement trajectory of the ablation needle within the tissue. Displaying the current predicted posture information, historical movement trajectory, target point, and danger zone together allows physicians to simultaneously observe the current state and previous movement of the ablation needle. This facilitates judgment on whether the actual needle insertion path deviates from the pre-planned path and provides an intuitive spatial reference for adjusting subsequent needle insertion directions, avoiding danger zones, and determining whether the effective working electrode segment has reached the target point.

[0113] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this application. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this application.

[0114] Furthermore, some embodiments of this application also provide an electronic device. The electronic device can be various forms of digital computer, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, etc. The electronic device can also be various forms of mobile devices, such as cellular phones, smartphones, wearable devices, and other similar computing devices.

[0115] The electronic device includes: one or more processors; and a memory storing computer program instructions that, when executed, cause the processor to perform the steps of the methods provided in any one or more of the above embodiments. Figure 2 An exemplary structural diagram of the electronic device is disclosed. The electronic device includes one or more processors 1101, a memory 1102, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components are interconnected via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the electronic device, including instructions stored in or on memory to display graphical information of a GUI on an external input / output device (such as a display device coupled to the interface). In some other embodiments, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple electronic devices can be connected, each providing some of the necessary operations. The components, their connections and relationships, and their functions shown herein are merely examples and are not intended to limit the implementation of the present application described and / or claimed herein.

[0116] The electronic device may further include an input device 1103 and an output device 1104. The processor 1101, memory 1102, input device 1103 and output device 1104 may be connected by a bus or other means, as shown in the figure, which is connected by a bus 1105.

[0117] Input device 1103 can receive input numerical or character information, and generate key signal inputs related to user settings and function control of the electronic device, such as a touch screen, keypad, mouse, trackpad, touchpad, joystick, one or more mouse buttons, trackball, joystick, etc. Output device 1104 may include a display device, auxiliary lighting device (e.g., LED), and haptic feedback device (e.g., vibration motor). The display device may include, but is not limited to, a liquid crystal display, a light-emitting diode display, and a plasma display. In some embodiments, the display device may be a touch screen.

[0118] To provide interaction with the user, the electronic device can be a computer. The computer has: a display device (e.g., a cathode ray tube or LCD monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse) through which the user provides input to the computer. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback); and input from the user can be received in any form (e.g., voice input or tactile input).

[0119] In this embodiment, a computer-readable medium stores a computer program / instructions that, when executed by a processor, implement the steps of the methods provided in any one or more of the above embodiments. This computer-readable medium may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into that device. The aforementioned computer-readable medium carries one or more computer-readable instructions.

[0120] The memory 1102 can serve as a non-transitory computer-readable storage medium, used to store non-transitory software programs, non-transitory computer-executable programs, and modules. The processor 1101 executes various functional applications and data processing of the server by running the non-transitory software programs, instructions, and modules stored in the memory 1102, thereby implementing the program instructions / modules corresponding to the methods provided in any one or more of the embodiments described above in this application.

[0121] The memory 1102 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the electronic device. Furthermore, the memory 1102 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory 1102 may optionally include memory remotely located relative to the processor 1101, and these remote memories can be connected to the electronic device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0122] It should be noted that the computer-readable medium described in this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable media can be, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatuses, or devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0123] Computer-readable media include permanent and non-permanent, removable and non-removable media, which can store information by any method or technology. Information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase-change memory, static random access memory, dynamic random access memory, other types of random access memory, read-only memory, electrically erasable programmable read-only memory, flash memory or other memory technologies, read-only optical discs, digital versatile optical discs or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transfer medium that can be used to store information accessible by a computing device.

[0124] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including local area networks (LANs) or wide area networks (WANs), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0125] In the above embodiments, all or part of the implementation can be achieved through software, hardware, firmware, or any combination thereof. For example, it can be implemented using an application-specific integrated circuit (ASIC), a general-purpose computer, or any other similar hardware device. In some embodiments, the software program of this application can be executed by a processor to implement the above steps or functions. Similarly, the software program of this application (including related data structures) can be stored in a computer-readable recording medium, such as RAM memory, magnetic or optical drives, floppy disks, and similar devices. In addition, some steps or functions of this application can be implemented in hardware, for example, as circuitry that cooperates with a processor to perform the various steps or functions.

[0126] The computer program product provided in this application includes one or more computer programs / instructions. When executed by a processor, these computer programs / instructions generate, in whole or in part, the processes or functions described in this application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions may be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0127] The flowcharts or block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of devices, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-specific system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0128] The scope of this application is defined by the appended claims rather than the foregoing description, and is therefore intended to encompass all variations falling within the meaning and scope of equivalents of the claims. No reference numerals in the claims should be construed as limiting the scope of the claims. Furthermore, it is clear that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural.

[0129] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily made by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims, and the above embodiments should be regarded as exemplary and non-limiting.

Claims

1. A method for monitoring the posture of an ablation needle, characterized in that, The method includes: A three-dimensional anatomical model containing target points and danger zones is created based on preoperative images; The inertial parameters, tip pressure data, impedance data, positioning data, insertion depth data, and physical model data of the ablation needle are obtained. The inertial parameters include the angular velocity data and acceleration data of the ablation needle. The inertial parameters, the needle tip pressure data, the impedance data, the positioning data, the needle insertion depth data, and the physical model data are input into a pre-trained attitude prediction model to obtain the predicted attitude information of the ablation needle in the positioning coordinate system. The predicted attitude information includes the position, orientation, and shape of the ablation needle. Determine the spatial registration relationship between the positioning coordinate system and the image coordinate system corresponding to the three-dimensional anatomical model; Based on the spatial registration relationship, the predicted pose information is transformed into the image coordinate system and fused with the three-dimensional anatomical model to determine the positional relationship of the ablation needle relative to the target point and the danger zone, and the pose of the ablation needle in the tissue is visualized.

2. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, The training process of the pose prediction model includes: While the simulated ablation needle is in different motion and stress states, the inertial parameters, needle tip pressure data, impedance data, positioning data and needle insertion depth data of the simulated ablation needle are collected simultaneously, and physical model data matching the simulated ablation needle and actual attitude information obtained by the micro electromagnetic sensor integrated into the simulated ablation needle are acquired. Using the inertial parameters, needle tip pressure data, impedance data, positioning data, needle insertion depth data, and physical model data as training samples, and the actual posture information as training labels, the model to be trained is trained to obtain the posture prediction model.

3. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, Before inputting the inertial parameters into the attitude prediction model, the angular velocity data and the acceleration data are fused using an extended Kalman filter algorithm or an unscented Kalman filter algorithm. The attitude direction of the ablation needle is updated recursively based on the angular velocity data, and the attitude direction is corrected based on the gravity component in the acceleration data to obtain the inertial attitude features. The inertial attitude characteristics, the needle tip pressure data, the impedance data, the positioning data, the needle insertion depth data, and the physical model data are input into the attitude prediction model.

4. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, Based on the acquisition time, the inertial parameters, the needle tip pressure data, the impedance data, the positioning data, and the needle insertion depth data are correlated over time to form a multi-source time series; The multi-source time series and physical model data matching the current ablation needle are input into the attitude prediction model, and the attitude prediction model predicts the attitude of the ablation needle at the current moment based on the changes of various data over time.

5. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, The physical model data is determined based on the length, diameter, material properties, and bending stiffness of the current ablation needle; The attitude prediction model determines the current needle tip stress state of the ablation needle based on the needle tip pressure data, and predicts the needle body morphology of the current ablation needle under the needle tip stress state by combining the needle body physical characteristics characterized by the physical model data.

6. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, The positioning data is obtained through optical positioning equipment or magnetic positioning equipment; The positions of multiple positioning reference points in the positioning coordinate system and the image coordinate system are determined respectively. Based on the positions of the corresponding positioning reference points in the two coordinate systems, the rigid transformation relationship between the positioning coordinate system and the image coordinate system is determined, and the rigid transformation relationship is used as the spatial registration relationship.

7. The method for monitoring the posture of an ablation needle according to claim 1, characterized in that, The inertial parameters, needle tip pressure data, impedance data, positioning data, and needle insertion depth data are continuously acquired at different acquisition times. The data obtained at each acquisition time and the physical model data are respectively input into the attitude prediction model to obtain the predicted attitude information at each acquisition time. Based on the spatial registration relationship, the predicted pose information at each acquisition time is converted to the image coordinate system to form the historical motion trajectory of the ablation needle; The current predicted posture information of the ablation needle, the historical motion trajectory, the target point, and the danger zone are all displayed in the three-dimensional anatomical model.

8. An electronic device, characterized in that, It includes a processor and a memory, wherein the memory stores a computer program, which, when executed by the processor, implements the ablation needle posture monitoring method according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the ablation needle posture monitoring method according to any one of claims 1 to 7.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the ablation needle posture monitoring method according to any one of claims 1 to 7.

Images