A lung puncture line control method and device, a puncture robot and a medium

Abstract

The present application relates to the technical field of computer control, and especially relates to a lung puncture line control method and device, a puncture robot and a medium, the method is applied to the puncture robot, and the method comprises the following steps: analyzing a CT image to obtain a lung target position; building a deep learning model to obtain real-time pose information of a biopsy needle in a magnetic field; controlling the biopsy needle of the puncture robot to move according to a target puncture path; judging whether the real-time pose information meets preset real-time pose information; if not, correcting the real-time pose information of the biopsy needle, so that the biopsy needle reaches the lung target position according to the target puncture path, and then the puncture needle reaches the lung target position according to an ideal state line, without causing damage to other tissues, and the positioning accuracy of the biopsy needle is improved.

Description

Technical Field

[0001] This invention relates to the field of computer control technology, and in particular to a method, device, puncture robot and medium for controlling the circuit of lung puncture. Background Technology

[0002] Lung cancer has become the leading cause of cancer death. A biopsy involves removing a portion of tissue from a suspected lung cancer tumor for pathological examination. Obtaining a sufficient biopsy sample is crucial for high-quality subsequent pathological analysis. Lung biopsy is the most commonly used method for obtaining biopsy samples.

[0003] Currently, lung biopsy requires multiple CT scans, which exposes the patient to a large amount of X-ray radiation in a short period, increasing the risk of cancer. Furthermore, lung biopsy is dependent on the surgeon's experience and carries the risk of complications.

[0004] Therefore, how to accurately deliver the biopsy needle to the target location is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] In view of the above problems, the present invention provides a method, apparatus, puncture robot and medium for controlling the lung puncture path to overcome or at least partially solve the above problems.

[0006] In a first aspect, the present invention provides a method for controlling the path of lung puncture, applied in a puncture robot, comprising: Obtain the target location of the lungs; Based on the target location of the lung, determine the target puncture path; The biopsy needle of the puncture robot is controlled to move along the target puncture path; Monitor the real-time pose information of the biopsy needle; Determine whether the real-time pose information meets the real-time preset pose information; If not, the real-time position information of the biopsy needle is corrected so that the biopsy needle reaches the target location in the lung according to the target puncture path.

[0007] Preferably, obtaining the target location of the lung includes: Acquire lung CT images; Based on the lung CT images, the target location of the lung is determined, which is the location of the lesion tissue in the lung.

[0008] Preferably, determining the target puncture path based on the target location of the lung includes: A deep learning algorithm is used to simulate the segmentation of the organs in the chest, and a volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface of the simulated chest's overall structure to generate sampling points; The sampling points are respectively connected to the target location in the lung to form a pre-selected puncture path; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

[0009] Preferably, the biopsy needle is provided with a Hall element, and the biopsy needle is located in the magnetic field generated by the three-dimensional rectangular circulating coil model, so that the biopsy needle generates an induced voltage. The induced voltage is different when the pose information of the biopsy needle changes.

[0010] Preferably, before determining whether the real-time pose information meets the real-time preset pose information, the method further includes: Acquire the real-time sensing voltage of the biopsy needle; Acquire pose information prediction model; The real-time induced voltage is input into the pose information prediction model to obtain real-time preset pose information.

[0011] Preferably, the pose information prediction model includes: Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and angle data of the biopsy needle as actually measured. The historical induced voltage data is used as input data, and the historical coordinate position and historical angle data are used as output data. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model.

[0012] Preferably, correcting the real-time pose information of the biopsy needle so that the biopsy needle reaches the target location in the lung according to the lung puncture path includes: Determine the pose difference information between the real-time pose information and the real-time preset pose information; Based on the positional difference information, the biopsy needle is adjusted to the real-time preset positional information so that the biopsy needle reaches the target position of the lung according to the lung puncture path.

[0013] Secondly, the present invention also provides a lung puncture circuit control device, applied in a puncture robot, comprising: The acquisition module is used to acquire the target location of the lungs; The determination module is used to determine the target puncture path based on the target location in the lung; The control module is used to control the biopsy needle of the puncture robot to move along the target puncture path; A monitoring module is used to monitor the real-time pose information of the biopsy needle; The judgment module is used to determine whether the real-time pose information meets the real-time preset pose information; The correction module is used to correct the real-time pose information of the biopsy needle, if not, so that the biopsy needle reaches the target position in the lung according to the target puncture path.

[0014] Thirdly, the present invention also provides a puncture robot, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the method described in the first aspect.

[0015] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described in the first aspect.

[0016] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages: This invention provides a lung puncture path control method applied in a puncture robot, comprising: acquiring the target location of the lung; determining the target puncture path based on the target location of the lung; controlling the biopsy needle of the puncture robot to move along the target puncture path; monitoring the real-time pose information of the biopsy needle; determining whether the real-time pose information meets the real-time preset pose information; if not, correcting the real-time pose information of the biopsy needle so that the biopsy needle reaches the target location of the lung along the target puncture path, thereby ensuring that the puncture needle reaches the target location of the lung along an ideal path without causing damage to other tissues, and improving the positioning accuracy of the biopsy needle. Attached Figure Description

[0017] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings: Figure 1 A schematic flowchart of the lung puncture line control method in an embodiment of the present invention is shown; Figure 2A schematic diagram of the target puncture path determined in an embodiment of the present invention is shown; Figure 3 The illustration shows a scene of a puncture robot creating an external magnetic field environment in an embodiment of the present invention; Figure 4 This diagram illustrates the neural network model architecture used during training in an embodiment of the present invention. Figure 5 A schematic diagram of the lung puncture circuit control device in an embodiment of the present invention is shown; Figure 6 A schematic diagram of the puncture robot in an embodiment of the present invention is shown. Detailed Implementation

[0018] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0019] Example 1

[0020] Embodiments of the present invention provide a method for controlling the lung puncture pathway, such as... Figure 1 As shown, it includes: S101, Obtain the target location of the lungs; S102, Determine the target puncture path based on the target location of the lung; S103, controls the biopsy needle of the puncture robot to move along the target puncture path; S104, monitors the real-time pose information of the biopsy needle; S105, determine whether the real-time pose information meets the real-time preset pose information; S106, if not, correct the real-time position information of the biopsy needle so that the biopsy needle reaches the target location in the lung along the target puncture path.

[0021] First, in S101, the specific process of obtaining the target location of the lung is as follows: Acquire lung CT images, and based on these images, determine the target location of the lung, which is the location of the lesion tissue in the lung.

[0022] Using the above method, only one lung CT image is required, thus minimizing harm to the patient. Based on this CT image, the suspected location of the lesion and the size of the tumor can be determined.

[0023] Next, S102 is performed to determine the target puncture path based on the target location of the lung.

[0024] Specifically, deep learning algorithms are used to simulate the segmentation of the organs in the chest, and volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface simulating the overall structure of the chest to generate sampling points; Pre-selected puncture paths are formed between the sampling points and the target locations in the lungs; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

[0025] First, deep learning algorithms are used to simulate and segment various organs in the chest, including skin, bones, heart, lungs, trachea, blood vessels, and masses. Then, volumetric tracheal radiographs (VRT) are used to reconstruct the 3D structure of each trachea in the chest, obtaining a simulated overall chest structure. Next, power line reconstruction is performed on the skin surface simulating the overall chest structure, and sampling points are obtained by cutting into the puncture site. Then, pre-selected puncture paths are formed between the sampling points and the target location in the abdomen, specifically by connecting the two locations.

[0026] Since these pre-selected puncture paths are not optimal, it is necessary to filter the pre-selected puncture paths based on path constraints in order to determine the target puncture path.

[0027] Specific path constraints include hard constraints and soft constraints. Hard constraints are the clinical guidelines that must be followed in puncture planning; soft constraints are further screening of paths that meet the conditions to obtain the optimal puncture path.

[0028] Among them, the hard constraint is to calculate the distance of each sampling point from the target location in the lung, and if it is less than the length of the puncture needle, then the puncture path is discarded.

[0029] Based on the simulated segmentation of organs, if the puncture path passes through the grid of organs, bones, or important blood vessels, the puncture path will be discarded.

[0030] If the angle between the puncture direction and the generated skin fitting plane is less than 20°, the puncture path will be discarded.

[0031] Soft constraints include: length constraints: constraints on the total length of the puncture path and the length of the puncture within the lung; distance constraints: constraints on the distance between the puncture path and its extension and important tissues and organs; and angle constraints: constraints on the puncture angle between the puncture path and the skin surface and the lung surface.

[0032] By using the above constraints, a large number of puncture paths that do not meet the constraints can be eliminated, thereby obtaining the target puncture path.

[0033] Next, execute S103 to control the biopsy needle of the puncture robot to move along the template puncture path.

[0034] like Figure 3 The image shows a scenario where the puncture robot operates within an external magnetic field environment. In practical applications, the biopsy needle 301 does not perfectly match the preset posture. Therefore, in order to monitor the posture of the biopsy needle in real time, an external magnetic field environment is created during puncture using the puncture robot to monitor the posture of the biopsy needle.

[0035] Specifically, a Hall element 302 is provided on the biopsy needle 301 to sense changes in the magnetic field. A three-dimensional rectangular circulating coil model 303 is set externally to generate a magnetic field, which causes the biopsy needle 301 to generate an induced voltage. The induced voltage generated is different when the position information of the biopsy needle 301 changes.

[0036] The magnetic field generating section uses an FY2300 signal generator to produce eight sets of sinusoidal signals with different frequencies. These signals are then amplified by an AD623 signal amplifier and connected to eight coils on the PCB electromagnetic emission board. The changing current passing through the coils generates a rectangular circulating magnetic field. Based on Pissa's law, a descriptive model of the magnetic field generated by the magnetic field emission source is established, representing the relationship between the spatial position of a point relative to the magnetic field emission source and the magnetic induction intensity at that point. Simultaneously, according to Faraday's law of electromagnetic induction, two induction coils in the puncture needle generate corresponding induced voltages in the alternating magnetic field, used to determine the position and angle of the biopsy needle. According to the law of electromagnetic induction, the magnetic induction intensity can be calculated from the induced voltage generated by the sensor in the magnetic field. A coupling relationship between the sensor pose and the induced voltage is established. The induced voltage is an analog signal, converted to a digital signal by a USB-6009 signal acquisition card and input to the computer. The algorithm processes this signal based on the above coupling relationship to determine the spatial positions of the two coils, thereby determining the specific position and angle of the coils.

[0037] Therefore, after S103, the real-time sensing voltage of the biopsy needle under real-time monitoring is first obtained; Acquire pose information prediction model; The real-time induced voltage is input into the pose information prediction model to obtain the actual predicted pose information.

[0038] Specifically, the pose information prediction model includes: Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and posture data of the biopsy needle measured in real time. Historical induced voltage data is used as input data, and historical coordinate position and historical attitude data are used as output data. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any one of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model.

[0039] By training with a large amount of historical measurement data to gradually improve model performance, and by using a suitable neural network model architecture, complex relationships in the sensed voltage signal can be effectively captured, thereby achieving high-precision pose calculation. The model's predictive ability will be continuously improved through iteration and optimization to adapt to pose calculation needs under different environments and conditions. Figure 4 The diagram shows the neural network model architecture used during training. The training and test sets are divided in a 6:4 ratio. Self-attention is used to capture multivariate correlations, and layer normalization and FFN are used to learn the global representation of the sequence.

[0040] Of course, in addition to these, image pose detection algorithms can also be used; variants of the U-net convolutional neural network: commonly used for robot localization and pose estimation; rotational positioning network structure, specifically a rotational positioning network structure diagram after feature engineering based on the input; pose estimation using Hough random forest, and pose calculation methods based on magnetic field knowledge transfer.

[0041] Among them, the pose calculation method based on magnetic field knowledge transfer utilizes transfer learning, an important approach for applying acquired knowledge from one domain to another. This method trains an intelligent calculation model using laboratory data samples, giving the model basic pose calculation capabilities. Then, the network parameters of the preceding and following layers of this model are used as initial network parameters for complex medical scenarios. By continuously narrowing the distribution differences between laboratory samples and complex medical scenario samples in the new space, the network parameters of the extracted complex medical scenario model are updated. This makes the model clinically applicable in different organisms.

[0042] The plan is to freeze the structure of the first few layers of the network, as these layers already have experience in extracting target features. Their weights and parameters are then locked, and the network is loaded into the upcoming clinical environment. This frozen network layer is then combined with the training of the later layers to ultimately form a network model suitable for the target domain environment. The advantage of transfer learning is that a new network model can handle new localization tasks with minimal training, requires less data, and effectively saves costs.

[0043] In vivo experiments were conducted under different environments to verify the accuracy and calculation speed of the electromagnetic generation device, and to resolve electromagnetic interference issues from other devices, thereby verifying the reliability of the pose information prediction model for transfer and use.

[0044] Finally, the ideal pose information of the biopsy needle (real-time preset pose information) is obtained using this pose information prediction model. At the same time, S104 is executed to monitor the real-time pose information of the biopsy needle.

[0045] Specifically, this method relies on receiving a rectangular circulating current signal through an induction coil, and then analyzing it to obtain the angle and position information of the biopsy needle. However, because this method introduces a deviation in the angle information, adding a Hall element helps to reduce the angle error.

[0046] Next, step S105 is executed to determine whether the real-time pose information meets the real-time preset pose information. Specifically, the pose information includes coordinate position information and angle information, and the corresponding coordinate position information and angle information are compared respectively.

[0047] Next, S106, if not, correct the real-time position information of the biopsy needle so that the biopsy needle reaches the target location in the lung along the lung puncture path.

[0048] Specifically, the pose difference information between real-time pose information and real-time preset pose information is determined; Based on the positional difference information, the biopsy needle is adjusted to the real-time preset positional information so that the biopsy needle reaches the target position in the lung according to the lung puncture path.

[0049] Correcting the real-time pose information of the biopsy needle involves adjusting its pose information to match the ideal, preset pose information. Adjusting the biopsy needle to this ideal pose improves its accuracy in positioning and execution.

[0050] One or more technical solutions in the embodiments of the present invention have at least the following technical effects or advantages: This invention provides a lung puncture path control method applied in a puncture robot, comprising: acquiring the target location of the lung; determining the target puncture path based on the target location of the lung; controlling the biopsy needle of the puncture robot to move along the target puncture path; monitoring the real-time pose information of the biopsy needle; determining whether the real-time pose information meets the real-time preset pose information; if not, correcting the real-time pose information of the biopsy needle so that the biopsy needle reaches the target location of the lung along the target puncture path, thereby ensuring that the puncture needle reaches the target location of the lung along an ideal path without causing damage to other tissues, and improving the positioning accuracy of the biopsy needle.

[0051] Example 2

[0052] Based on the same inventive concept, this invention also provides a lung puncture circuit control device, which is applied in a puncture robot, such as... Figure 5 As shown, it includes: Acquisition module 501 is used to acquire the target location of the lungs; The determining module 502 is used to determine the target puncture path based on the lung target location; Control module 503 is used to control the biopsy needle of the puncture robot to move according to the target puncture path; The monitoring module 504 is used to monitor the real-time pose information of the biopsy needle; The judgment module 505 is used to determine whether the real-time pose information meets the real-time preset pose information; The correction module 506 is used to correct the real-time pose information of the biopsy needle if not to ensure that the biopsy needle reaches the target position in the lung according to the target puncture path.

[0053] In one alternative implementation, the acquisition module 501 is used for: Acquire lung CT images; Based on the lung CT images, the target location of the lung is determined, which is the location of the lesion tissue in the lung.

[0054] In one alternative implementation, the determining module 502 is configured to: A deep learning algorithm is used to simulate the segmentation of the organs in the chest, and a volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface of the simulated chest's overall structure to generate sampling points; The sampling points are respectively connected to the target location in the lung to form a pre-selected puncture path; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

[0055] In one optional embodiment, a Hall element is provided on the biopsy needle, and the biopsy needle is located in the magnetic field generated by the three-dimensional rectangular circulating coil model, so that the biopsy needle generates an induced voltage. The induced voltage is different when the pose information of the biopsy needle changes.

[0056] In one optional implementation, it further includes: a real-time preset pose information acquisition module, used for: Before determining whether the real-time pose information meets the real-time preset pose information, the real-time sensing voltage of the biopsy needle is acquired. Acquire pose information prediction model; The real-time induced voltage is input into the pose information prediction model to obtain real-time preset pose information.

[0057] In one optional implementation, the module for obtaining real-time preset pose information is further used for: Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and angle data of the biopsy needle as actually measured. The historical induced voltage data is used as input data, and the historical coordinate position and historical angle data are used as output data. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model.

[0058] In one alternative implementation, the correction module 506 is configured to: Determine the pose difference information between the real-time pose information and the real-time preset pose information; Based on the positional difference information, the biopsy needle is adjusted to the real-time preset positional information so that the biopsy needle reaches the target position of the lung according to the lung puncture path.

[0059] Example 3

[0060] Based on the same inventive concept, embodiments of the present invention provide a puncture robot, such as... Figure 6 As shown, it includes a memory 604, a processor 602, and a computer program stored in the memory 604 and executable on the processor 602. When the processor 602 executes the program, it implements the steps of the above-described lung puncture circuit control method.

[0061] Among them, Figure 6In this document, a bus architecture (represented by bus 600) is used. Bus 600 may include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 602 and memory represented by memory 604. Bus 600 may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 606 provides an interface between bus 600 and receiver 601 and transmitter 603. Receiver 601 and transmitter 603 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 602 is responsible for managing bus 600 and general processing, while memory 604 can be used to store data used by processor 602 during operation.

[0062] Example 4

[0063] Based on the same inventive concept, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described lung puncture circuit control method.

[0064] The algorithms and displays provided herein are not inherently related to any particular computer, virtual system, or other device. Various general-purpose systems can also be used in conjunction with the teachings herein. The required structure for constructing such systems is apparent from the above description. Furthermore, this invention is not directed to any particular programming language. It should be understood that the contents of the invention described herein can be implemented using various programming languages, and the above description of specific languages ​​is for the purpose of disclosing the best mode of implementation of the invention.

[0065] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. In some instances, well-known methods, structures, and techniques have not been shown in detail so as not to obscure the understanding of this specification.

[0066] Similarly, it should be understood that, in order to streamline this disclosure and aid in understanding one or more of the various inventive aspects, in the above description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof. However, this method of disclosure should not be construed as reflecting an intention that the claimed invention requires more features than are explicitly recited in each embodiment. Rather, as reflected in each embodiment, inventive aspects lie in fewer than all features of the single embodiment disclosed above. Therefore, the claims, following the detailed description, are hereby expressly incorporated into this detailed description, wherein each claim itself is a separate embodiment of the invention.

[0067] Those skilled in the art will understand that modules in the device of the embodiments can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components. Except where at least some of such features and / or processes or units are mutually exclusive, any combination can be used to combine all features disclosed in this specification (including the accompanying claims, abstract, and drawings) and all processes or units of any method or device so disclosed. Unless expressly stated otherwise, each feature disclosed in this specification (including the accompanying claims, abstract, and drawings) may be replaced by an alternative feature that serves the same, equivalent, or similar purpose.

[0068] Furthermore, those skilled in the art will understand that although some embodiments herein include certain features included in other embodiments but not others, combinations of features from different embodiments are meant to be within the scope of the invention and form different embodiments. For example, in the specific implementation, any of the claimed embodiments can be used in any combination.

[0069] The various component embodiments of the present invention can be implemented in hardware, or as software modules running on one or more processors, or a combination thereof. Those skilled in the art will understand that microprocessors or digital signal processors (DSPs) can be used in practice to implement some or all of the functions of some or all of the components in the lung puncture circuit control device or puncture robot according to embodiments of the present invention. The present invention can also be implemented as a device or apparatus program (e.g., a computer program and computer program product) for performing part or all of the methods described herein. Such programs implementing the present invention can be stored on a computer-readable medium or can be in the form of one or more signals. Such signals can be downloaded from an Internet website, provided on a carrier signal, or provided in any other form.

[0070] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

Claims

1. A lung puncture circuit control device, applied in a puncture robot, characterized in that, include: The acquisition module is used to acquire the target location of the lungs; The determination module is used to determine the target puncture path based on the target location in the lung; The control module is used to control the biopsy needle of the puncture robot to move along the target puncture path. The biopsy needle is equipped with a Hall element and is located in the magnetic field generated by the three-dimensional rectangular circulating coil model, so that the biopsy needle generates an induced voltage. The induced voltage is different when the pose information of the biopsy needle changes. A monitoring module is used to monitor the real-time pose information of the biopsy needle; The real-time pose prediction module is used to: acquire the real-time induced voltage of the monitored biopsy needle; Also used for: Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and angle data of the biopsy needle as actually measured. The historical induced voltage data is used as input samples, and the historical coordinate position and historical angle data are used as output samples. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any one of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model; Pose information prediction model: used to obtain real-time predicted pose information based on the real-time sensed voltage; The judgment module is used to determine whether the real-time pose information meets the real-time predicted pose information; The correction module is used to correct the real-time pose information of the biopsy needle, if not, so that the biopsy needle reaches the target position in the lung according to the target puncture path.

2. The apparatus as claimed in claim 1, characterized in that, The acquisition module is used for: Acquire lung CT images; Based on the lung CT images, the target location of the lung is determined, which is the location of the lesion tissue in the lung.

3. The apparatus as described in claim 1, characterized in that, The determination module is used for: A deep learning algorithm is used to simulate the segmentation of the organs in the chest, and a volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface of the simulated chest's overall structure to generate sampling points; The sampling points are respectively connected to the target location in the lung to form a pre-selected puncture path; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

4. The apparatus as claimed in claim 1, characterized in that, The correction module is used for: Determine the pose difference information between the real-time pose information and the real-time predicted pose information; Based on the pose difference information, the biopsy needle is adjusted to the real-time predicted pose information so that the biopsy needle reaches the target position of the lung according to the target puncture path.

5. A puncture robot, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the following method: Obtain the target location of the lungs; Based on the target location of the lung, determine the target puncture path; The biopsy needle of the puncture robot is controlled to move along the target puncture path. The biopsy needle is equipped with a Hall element and is located in the magnetic field generated by the three-dimensional rectangular circulating coil model, which causes the biopsy needle to generate an induced voltage. The induced voltage is different when the pose information of the biopsy needle changes. Monitor the real-time pose information of the biopsy needle; Acquire the real-time sensing voltage of the biopsy needle being monitored; Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and angle data of the biopsy needle as actually measured. The historical induced voltage data is used as input samples, and the historical coordinate position and historical angle data are used as output samples. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any one of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model; Based on the real-time sensed voltage, real-time predicted pose information is obtained; Determine whether the real-time pose information satisfies the real-time predicted pose information; If not, the real-time position information of the biopsy needle is corrected so that the biopsy needle reaches the target location in the lung according to the target puncture path.

6. The puncture robot as described in claim 5, characterized in that, When the processor executes the program, it specifically implements the following method: A deep learning algorithm is used to simulate the segmentation of the organs in the chest, and a volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface of the simulated chest's overall structure to generate sampling points; The sampling points are respectively connected to the target location in the lung to form a pre-selected puncture path; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

7. The puncture robot as described in claim 5, characterized in that, When the processor executes the program, it specifically implements the following method: Determine the pose difference information between the real-time pose information and the real-time predicted pose information; Based on the pose difference information, the biopsy needle is adjusted to the real-time predicted pose information so that the biopsy needle reaches the target position of the lung according to the target puncture path.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When this program is executed by the processor, it implements the following method: Obtain the target location of the lungs; Based on the target location of the lung, determine the target puncture path; The biopsy needle of the puncture robot is controlled to move along the target puncture path. The biopsy needle is equipped with a Hall element and is located in the magnetic field generated by the three-dimensional rectangular circulating coil model, which causes the biopsy needle to generate an induced voltage. The induced voltage is different when the pose information of the biopsy needle changes. Monitor the real-time pose information of the biopsy needle; Acquire the real-time sensing voltage of the biopsy needle being monitored; Acquire historical measurement data, which includes historical induced voltage data sensed by the biopsy needle and historical coordinate position and angle data of the biopsy needle as actually measured. The historical induced voltage data is used as input samples, and the historical coordinate position and historical angle data are used as output samples. Both are input into a deep learning neural network model for training to obtain a pose information prediction model. The deep learning neural network model can be any one of the following: Attention mechanism network model, convolutional neural recurrent network model, ConvLSTM, deep graph neural network model, and bidirectional long short-term memory network model; Based on the real-time sensed voltage, real-time predicted pose information is obtained; Determine whether the real-time pose information satisfies the real-time predicted pose information; If not, the real-time position information of the biopsy needle is corrected so that the biopsy needle reaches the target location in the lung according to the target puncture path.

9. The computer-readable storage medium as claimed in claim 8, characterized in that, When this program is executed by the processor, it implements the following method: A deep learning algorithm is used to simulate the segmentation of the organs in the chest, and a volumetric roaming method is used to reconstruct the three-dimensional structure of the organs in the chest to obtain the simulated overall structure of the chest. Point cloud reconstruction is performed on the skin surface of the simulated chest's overall structure to generate sampling points; The sampling points are respectively connected to the target location in the lung to form a pre-selected puncture path; Based on path constraints, the pre-selected puncture paths are screened to determine the target puncture path.

10. The computer-readable storage medium as claimed in claim 8, characterized in that, When this program is executed by the processor, it implements the following method: Determine the pose difference information between the real-time pose information and the real-time predicted pose information; Based on the pose difference information, the biopsy needle is adjusted to the real-time predicted pose information so that the biopsy needle reaches the target position of the lung according to the target puncture path.

Images