Precise prostate puncture system

Abstract

The application provides a prostate precise puncture system, comprising: an image acquisition module configured to acquire anus image data of a patient; an anus detection module configured to receive the anus image data to detect the anus position of the patient in real time based on the anus image data; the posture adjustment module is configured to control the mechanical arm to perform posture adjustment, so that the mechanical arm pushes the ultrasonic probe to pass through the anus position to obtain prostate ultrasonic image data; and the needle insertion correction module is configured to obtain a needle insertion puncture position based on the needle insertion parameters and transmit the needle insertion puncture position to the posture adjustment module, so that the posture adjustment module performs needle insertion puncture on the focus planning point of the prostate based on the needle insertion puncture position and the prostate ultrasonic image data. The difficulty of manually placing the probe and the posture of the mechanical arm in the prostate puncture operation is improved, the needle inserting position is corrected, the sampling precision is improved, anatomical structures such as the urethra and the pubis are avoided, and the occurrence rate of puncture related complications and the puncture failure rate are reduced.

Description

Technical Field

[0001] The present application belongs to the field of prostate technology, and in particular relates to a prostate precision puncture system. Background Art

[0002] Prostate cancer is a common malignant tumor of the male reproductive system, occurring primarily in middle-aged and older men. Early symptoms are subtle and may include dysuria, frequent urination, hematuria, or bone pain (if metastasized). Its development is related to age, genetics, hormone levels, and lifestyle. While some types progress slowly, high-risk types can spread rapidly.

[0003] Currently, the diagnosis of prostate cancer still relies on prostate biopsy to clarify its pathology. This method typically involves ultrasound-guided transrectal or perineal puncture to obtain a prostate tissue sample. However, the accuracy of prostate puncture is largely dependent on the surgeon's experience and proficiency, and in clinical practice, it is often necessary to increase the number of puncture needles to improve puncture accuracy, which can easily increase the incidence of prostate puncture-related complications. Therefore, how to achieve accurate prostate puncture has become a technical problem that needs to be solved by those skilled in the art. Summary of the Invention

[0004] The present application provides a prostate precision puncture system for reducing the incidence of prostate puncture-related complications and puncture failure rate, and achieving precise prostate puncture.

[0005] According to one aspect of the present application, a prostate precision puncture system is provided, comprising:

[0006] An image acquisition module is configured to acquire anal image data of a patient;

[0007] an anus detection module, configured to receive the anus image data and detect the anus position of the patient in real time based on the anus image data;

[0008] a posture adjustment module, configured to control the robotic arm to perform posture adjustment so that the robotic arm pushes the ultrasound probe through the anus to obtain prostate ultrasound image data;

[0009] The needle insertion correction module is configured to obtain the needle insertion puncture position based on the needle insertion parameters and transmit the needle insertion puncture position to the posture adjustment module, so that the posture adjustment module performs needle puncture on the prostate lesion planning point based on the needle insertion puncture position and the prostate ultrasound image data.

[0010] In some implementations, the needle insertion correction module includes a needle insertion correction model, and the needle insertion parameters serve as input to the needle insertion correction model;

[0011] The needle insertion correction model is configured to obtain the needle insertion deviation angle and position deviation distance through a fully connected layer and a feature cross layer, and obtain the needle insertion puncture position based on the needle insertion deviation angle and the position deviation distance.

[0012] In some implementations, the image acquisition module is further configured to: divide the anus image data into equal areas and determine a set area, so that the set area is fixedly corresponding to the ultrasound probe.

[0013] In some implementations, the anus detection module is further configured to: determine the partition position of the anus position in the anus image data, and determine whether the partition position is consistent with the set area; if not, provide feedback to the posture adjustment module so that the posture adjustment module controls the robotic arm to adjust its posture until the partition position is consistent with the set area.

[0014] In some implementations, the anus detection module includes an anus detection model, and the anus image data serves as input to the anus detection model;

[0015] The anus detection model is configured to perform feature extraction on the anus image data to obtain a feature map, downsample the feature map and splice the feature map; and perform non-maximum suppression on the spliced ​​feature map and return a probability matrix to obtain the anus position.

[0016] In some implementations, the needle insertion correction module is further configured to obtain a positional deviation between the needle insertion puncture position and the lesion planning point, and to correct subsequent needle insertion puncture positions of the lesion planning point based on the positional deviation.

[0017] In some implementations, the posture adjustment module is further configured to determine whether the infrared distance between the robotic arm and the patient's perineum meets a distance threshold; when the infrared distance meets the threshold condition, the posture adjustment module controls the robotic arm to push the ultrasound probe through the anus position and acquire the prostate ultrasound image data in real time.

[0018] In some implementations, the needle advancement correction module is further configured to plan a needle track based on the prostate ultrasound image data so that the needle track avoids a non-punctureable area, wherein the non-punctureable area includes a urethra and a pubic bone.

[0019] In some implementations, the posture adjustment module determines a base portion of the prostate through the ultrasound probe; and the posture adjustment module uses the base portion as an entry depth of the ultrasound probe.

[0020] In some implementations, the needle insertion parameters include: patient age, prostate-specific antigen, weight, prostate tissue shape, prostate tissue volume, prostate ultrasound image visualization, and skin hardness.

[0021] First of all, this application uses the anus detection model to realize target tracking detection of anal image data, realizes automatic search for the patient's anus, and allows the probe to smoothly enter the body and the robotic arm needle insertion point to fit the perineum, improving the difficulty of manually placing the probe and robotic arm posture, and making the operation more convenient.

[0022] Secondly, the application uses a needle insertion correction model to automatically adjust the puncture position based on the needle insertion parameters, improving prostate sampling accuracy, maximizing sampling of suspicious lesions, increasing the sensitivity and specificity of prostate puncture, and avoiding multiple unnecessary punctures. Furthermore, the application can avoid anatomical structures such as the urethra and pubic bone, reducing the incidence of prostate puncture-related complications and puncture failure rates.

[0023] Finally, this application can realize the automation and intelligence of prostate puncture surgery, reduce the dependence on the surgeon's experience, reduce the doctor's workload during surgery and possible preoperative misjudgment problems, and greatly improve the efficiency and accuracy of puncture. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Shown is a schematic diagram of the application of a prostate precision puncture system in one embodiment of the present application.

[0025] Figure 2 Shown is a schematic diagram of the architecture of a prostate precision puncture system according to one embodiment of the present application.

[0026] Figure 3 Shown is a schematic diagram of the relationship between the image acquisition module and the ultrasound probe in one embodiment of the present application.

[0027] Figure 4 Shown is a schematic diagram of the principle of an anus detection model in one embodiment of the present application.

[0028] Figure 5 Shown is a schematic diagram of the principle of a posture adjustment module in one embodiment of the present application.

[0029] Figure 6 Shown is a schematic diagram of the principle of a needle insertion correction model in one embodiment of the present application.

[0030] Figure 7 Shown is a schematic diagram of needle puncture deviation in one embodiment of the present application.

[0031] Figure 8 Shown is a schematic diagram of collecting training data for a needle insertion correction model in one embodiment of the present application.

[0032] Figure 9 Shown is a schematic diagram of the principle of characteristic cross layers in the needle insertion correction model in one embodiment of the present application.

[0033] Figure 10 Shown is a schematic diagram of the principle of the fully connected layer in the needle insertion correction model in one embodiment of the present application. DETAILED DESCRIPTION

[0034] The following describes the embodiments of the present application through specific examples. Those skilled in the art can easily understand the other advantages and effects of the present application from the content disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0035] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. Therefore, the illustrations only show components related to the present application and are not drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component can be changed at will, and the component layout type may also be more complicated.

[0036] In addition, the descriptions of "first", "second", etc. in this application are for descriptive purposes only and should not be understood as indicating or implying their relative importance or implicitly indicating the number of the technical features indicated. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of such features. In addition, the technical solutions between the various embodiments can be combined with each other, but this must be based on the fact that they can be implemented by ordinary technicians in this field. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by this application.

[0037] At present, a variety of transperineal prostate puncture techniques have gradually developed, mainly including transperineal prostate puncture biopsy guided by positioning templates, free-hand transperineal prostate puncture biopsy, and MRI-assisted targeted transperineal prostate puncture biopsy. Among them, transperineal prostate puncture biopsy guided by positioning templates appeared earlier. Its advantage is that it uses a fixed puncture template, which can evenly sample the entire prostate, better reflect the actual clinical risk grading, and reduce the risk of infection caused by the increase in the number of puncture needles. However, despite the above advantages, the operation is relatively complicated during the puncture practice, requiring the use of a special prostate puncture stand and prostate positioning template, and it needs to be completed under general anesthesia. At the same time, it is impossible to dynamically adjust the positioning template according to each patient's prostate size, urethra, pubic bone and other pelvic anatomical structure parameters. This undoubtedly increases the incidence of puncture-related complications and the puncture failure rate.

[0038] To further address the practical challenges of template-guided transperineal prostate biopsy, freehand transperineal prostate biopsy under local anesthesia using only ultrasound equipment has gradually gained attention. This technique involves designing a scalable puncture holder and cannula needle, connecting the ultrasound probe and needle on a flat surface. Sampling is then performed using a fan-shaped puncture technique throughout the prostate, while avoiding critical anatomical structures such as the urethra under intraoperative ultrasound guidance. The overall tumor detection rate can reach 88.6%, and the csPCa (clinically significant prostate cancer) detection rate can reach 82.9%. However, this approach also has certain drawbacks. For example, due to the lack of template guidance, this puncture method relies more on the surgeon's own technique, and the surgeon's experience and proficiency may affect the accuracy of the puncture. Furthermore, deformation of the needle during insertion and prostate deformation caused by changes in patient position can also reduce puncture accuracy. Therefore, in clinical practice, it is often necessary to increase the number of puncture needles to improve puncture accuracy, which increases the incidence of prostate puncture-related complications. Therefore, achieving precise prostate puncture is a technical problem that needs to be solved urgently.

[0039] At least to address the above problems, the present invention provides a prostate precision puncture system that can be applied to Figure 1 In the trolley shown, the robot arm is automatically pushed into the patient's body to perform puncture sampling. After each point puncture is completed, the robot arm moves outside the body and obtains the current sample. Figure 1As shown, the trolley 1 is provided with a camera 10, an infrared device 11, a robotic arm 12, and an ultrasonic probe 13. After the patient is in the lithotomy position, the trolley 1 is pushed closer to the patient, and the patient's MRI image data and the patient's needle insertion parameters are imported. At this time, the camera 10 on the probe axis of the robotic arm 12 will collect the anus image data facing it, and the anus detection model will perform real-time detection of the patient's anus position, and the feedback will be sent to the positioning axis of the robotic arm 12 for sliding adjustment up and down and sliding back and forth. When the detected anus position is in the set area in the anus image data, the infrared device 11 on the needle insertion axis of the robotic arm 12 is called for distance measurement. When the infrared distance is less than the set 1-2 cm, the initial posture adjustment of the needle insertion axis and the probe axis of the robotic arm 12 is completed.

[0040] Afterwards, the ultrasonic probe is pushed into the anus by the probe shaft of the robotic arm 12 to obtain an ultrasonic image and complete the scanning of the organ, and the spatial position during the operation is determined in the ultrasonic image based on the overall range of the prostate and the position of the suspected lesions outlined on various levels of the MRI image data by the assistive assessment software before the operation. After the positioning is completed, the puncture robot locates the lesion planning point (puncture point) in vitro and calculates the needle track of the puncture needle to avoid key parts such as the urethra and pubic bone. When everything is ready, the needle insertion position (the point after the lesion planning point is corrected) is calculated based on the needle insertion parameters according to the needle insertion correction model during the puncture process, so as to control the needle insertion axis of the robotic arm 12 to achieve needle insertion puncture and avoid needle insertion deviation. If a position deviation is found during the first needle insertion, the puncture position of the subsequent points is corrected again based on this position deviation information. When all points are punctured or manually canceled, the robotic arm returns to the initial state. In addition, a left and right rotating shaft 110 , an up and down telescopic shaft 111 and a bottom sliding shaft 112 are provided below the trolley 1 , so as to enable the trolley 1 to move freely.

[0041] The technical solutions in the embodiments of the present application will be described in detail below with reference to the accompanying drawings in the embodiments of the present application.

[0042] Figure 2 Shown is a schematic diagram of the architecture of the prostate precision puncture system described in the embodiment of this application. Figure 2 As shown, the prostate precision puncture system 2 includes an image acquisition module 20, an anus detection module 21, a posture adjustment module 22 and a needle insertion correction module 23.

[0043] The image acquisition module 20 is configured to acquire anal image data of the patient.

[0044] The anus detection module 21 is configured to receive the anus image data to detect the anus position of the patient in real time based on the anus image data.

[0045] The posture adjustment module 22 is configured to control the robotic arm to perform posture adjustment so that the robotic arm pushes the ultrasound probe into the patient's anus to obtain prostate ultrasound image data.

[0046] Among them, the needle insertion correction module 23 is configured to obtain the needle insertion puncture position based on the needle insertion parameters, and transmit the needle insertion puncture position to the posture adjustment module, so that the posture adjustment module performs needle puncture on the prostate lesion planning point based on the needle insertion puncture position and the prostate ultrasound image data.

[0047] In some embodiments, the image acquisition module 20 divides the anus image data into equal areas and determines a set area, so that the set area is fixedly corresponding to the ultrasound probe. Figure 3 As shown, in some embodiments, the ultrasound probe is positioned above the image acquisition module 20 and, together with the image acquisition module 20, is secured to the probe shaft of the robotic arm. The image acquisition module 20 then divides the acquired image data into nine equal parts, aligning the ultrasound probe with a specific partition, or designated area. Specifically, the actual distance between the image acquisition module 20 and the ultrasound probe is determined, and the imaging parameters of the image acquisition module 20 are set so that the ultrasound probe fits within a designated area of ​​the ultrasound image data.

[0048] Furthermore, after collecting the anus image data, the image acquisition module 20 also pre-processes the image, converts the image into tensor form data and transmits it to the anus detection module 21, so that the anus detection module 21 detects the patient's anus position based on the received data.

[0049] In some embodiments, the anus detection module 21 determines the anus position in the partition position of the anus image data and determines whether the partition position is consistent with the set area; if not, the feedback is sent to the posture adjustment module 22 so that the posture adjustment module 22 controls the robot arm to adjust its posture until the partition position is consistent with the set area. Figure 3 Since the ultrasound probe always satisfies a set area on the ultrasound image, when the anus position detected by the anus detection module 21 is in the set area in the partition position on the anus image data, it means that the upper, lower, left and right positions of the initial posture of the robotic arm can be determined at this time, that is, only the probe axis of the robotic arm needs to move forward as a whole, and the ultrasound probe can smoothly pass through the anus into the human body, which is convenient for subsequent surgical positioning and errors in manual placement of the robotic arm.

[0050] In some embodiments, the anus detection module 21 includes an anus detection module, the anus image data is used as the input of the anus detection model, the anus detection model performs feature extraction on the anus image data to obtain a feature map; the anus detection model 21 downsamples the feature map and splices the feature map; the anus detection model 21 performs non-maximum suppression on the spliced ​​feature map and returns a probability matrix to obtain the anus position. Figure 4 As shown in the figure, the anus detection model adopts the classic SDD type structure. The anus image data first passes through the convolution input layer and then the convolutional neural network performs feature extraction, gradually downsamples the feature map size, splices the feature map, and performs non-maximum suppression on the spliced ​​feature vector to determine the optimality of the target frame and obtain the output detection frame, that is, the anus position.

[0051] In some embodiments, the anus detection model is trained by supervised learning. First, anus images of multiple patients in the lithotomy position are collected, and the target areas of the collected images are labeled to obtain a set of data pairs corresponding to the images and labels. Afterwards, in order to obtain better training results, the data pair set is grouped into a training set, a validation set, and a test set according to an 8:1:1 ratio. Among them, the data pair set input is processed and transformed into a tensor form, and input into the designed anus detection model for feature extraction and prediction of the anus position. Each iteration calculates the deviation between the output probability matrix of the anus position and the corresponding label, and returns the deviation value to the model for parameter update and the start of the next iteration process. After multiple iterative trainings, the best model among these results is used as the anus detection model in the application, thereby realizing automatic detection of the anus.

[0052] In fact, the image acquisition module 20 and the anus detection module 21 perform real-time data interaction. When the image acquisition module 20 moves, the current anus image data is collected in real time, and the anus detection module 21 automatically detects the anus position of the current image, and determines whether the anus position in the partition position of the anus image data is consistent with the set area, and continuously feeds back to the posture adjustment module 22, so that the posture adjustment module 22 continuously adjusts up, down, left and right to obtain different anus image data, until the detected anus position is consistent with the set area, it means that the ultrasound probe is facing the anus at this time, and the initial posture adjustment of the posture adjustment module 22 is completed.

[0053] In some embodiments, the posture adjustment module 22 determines whether the infrared distance between the robotic arm and the patient's perineum meets a distance threshold. When the infrared distance meets the threshold condition, the posture adjustment module 22 controls the robotic arm to push the ultrasound probe into the patient's anus and obtain the prostate ultrasound image data in real time. That is, when the posture adjustment module 22 has completed the posture adjustment, that is, the anus is in the set area on the anus image data, the posture adjustment module 22 uses the infrared device to measure the distance and control the ultrasound probe to enter the patient's body. Figure 5 As shown, after the initial posture adjustment of the posture adjustment module 22 is completed by the image acquisition module 20 and the anus detection module 21, the robotic arm is controlled to move toward the patient, and the infrared distance between the robotic arm and the patient's perineum is measured using infrared equipment. When the infrared distance is less than or equal to d, the posture adjustment module 22 controls the robotic arm to stop moving forward as a whole, and the probe shaft of the robotic arm can be used to push the ultrasound probe into the patient's body.

[0054] In some embodiments, the posture adjustment module 22 determines the base of the prostate through the ultrasound probe; the posture adjustment module 22 uses the base as the insertion depth of the ultrasound probe. Figure 5 As mentioned above, when the posture adjustment module 22 controls the robotic arm to push the ultrasound probe into the patient's body, the prostate ultrasound image can be obtained in real time based on the ultrasound probe to determine the base position of the prostate, and the robotic arm is controlled to push the ultrasound probe to the base of the prostate. At this time, the depth of this position is the initial entry depth of the ultrasound probe.

[0055] Once everything is ready, a needle puncture is performed to collect a prostate sample to determine whether it is positive or negative. In some embodiments, a real-time ultrasound image is acquired after the ultrasound probe enters the anus. Software is then used to fuse the preoperative MRI image with the real-time ultrasound image during the procedure. This allows the overall prostate area outlined at each level of the MRI image data and the location of the suspected lesion to be mapped to the real-time ultrasound image to determine the spatial location during the procedure.

[0056] Furthermore, since the malignancy levels of the core area and peripheral area of ​​the same suspected lesion may be inconsistent, in order to achieve more accurate puncture, the embodiment of the present application uses a film reading model to grade the interior of the lesion, and then plans the puncture needle track for the core area of ​​the lesion, thereby maximizing the sampling value of the core area of ​​the lesion and improving the puncture accuracy.

[0057] In some embodiments, the film reading model can adopt a pre-trained deep learning model, which can realize semantic segmentation of the lesion area based on the training process, and perform PI-RADS scoring on each segmented area to grade the interior of the lesion and obtain the core area of ​​the lesion.

[0058] When the positioning and calculation are completed, puncture of the planned lesion point can be performed. However, in actual clinical situations, due to the heterogeneity of tissue structure, tissue resistance and elastic retraction, needle design and other reasons, the puncture needle will have a slight deviation during actual puncture, and the small deviation of the needle position will cause a large deviation of the actual sampling point, thereby affecting the accuracy of puncture sampling and affecting clinical diagnosis and treatment. Therefore, the needle correction module 23 can obtain the needle puncture position based on the needle insertion parameters, and transmit the needle puncture position to the posture adjustment module 22, so that the posture adjustment module 22 performs needle puncture on the lesion planning point of the prostate based on the needle puncture position and the prostate ultrasound image data. That is, the needle correction module 23 can automatically adjust the needle puncture position to correct the deviation according to the deviation of the needle puncture position caused by the needle insertion parameters during automatic needle insertion, thereby achieving accurate sampling of the suspicious lesion position. Continue to refer to Figure 5 After the needle insertion correction module 23 feeds back the needle insertion position to the posture adjustment module 22, the posture adjustment module 23 can control the needle insertion axis of the robotic arm to perform needle insertion.

[0059] In some embodiments, the needle insertion parameters include: patient age, prostate-specific antigen, weight, prostate tissue shape, prostate tissue volume, prostate ultrasound image visualization, and skin hardness. In other words, these needle insertion parameters may lead to deviations in the needle insertion position.

[0060] In some embodiments, the needle insertion correction module 23 includes a needle insertion correction model, and the needle insertion parameters are used as input of the needle insertion correction model. The needle insertion correction model is configured to obtain the needle insertion deviation angle and position deviation distance through a fully connected layer and a feature cross layer, and obtain the needle insertion puncture position based on the needle insertion deviation angle and the position deviation distance.

[0061] Among them, the structure and principle of the needle correction model are as follows Figure 6 As shown in the figure. Using the needle insertion parameters of different patients as input, the needle insertion correction model can obtain different feature vectors. After the feature vectors pass through the fully connected layer and the feature cross layer and then the splicing layer, the bending angle and deviation distance (needle insertion deviation) can be obtained. Based on these two, the needle insertion position can be obtained. In other words, the needle insertion position is actually the puncture position after the lesion planning point is corrected based on the needle insertion deviation caused by the needle insertion parameters.

[0062] In some embodiments, as Figure 7 As shown in the figure, A is the actual needle insertion position and B is the lesion planning point. The deviation distance between the two points A and B, as well as the bending angle between the two needle insertion routes, can be calculated through the spatial coordinates of the two points. Figure 8As shown in the figure, the needle insertion parameters of past patients and the corresponding deviation distance and bending angle (needle insertion deviation) are collected and a corresponding relationship is established between them to form a data set for training the needle insertion correction model. After the parameters of the needle insertion correction model are iteratively converged and trained, the needle insertion correction model can predict the needle insertion deviation that occurs during actual biopsy needle insertion based on the needle insertion parameters of new patients to obtain the needle insertion position, thereby making targeted corrections to the lesion planning point.

[0063] In some embodiments, as Figure 9 As shown in the figure, the output X1 of each layer of the feature cross-product layer in the needle correction model is the output of the previous layer * the transposed output of the previous layer * the weight + the output of the previous layer, which can be expressed as: X1 = X0 * X0T * W1 + X0. In this way, the output vector of each layer undergoes a cross product, increasing its nonlinearity.

[0064] In some embodiments, as Figure 10 As shown, each neuron node in each layer of the fully connected layer of the needle correction model is connected to each neuron node in the next layer, that is, the output X1 of the lower layer can be expressed as the output X0*weight of the previous layer, that is, X1=X0*W1.

[0065] In some embodiments, the needle insertion correction module 23 obtains the position deviation between the needle insertion puncture position and the lesion planning point; the needle insertion correction module 23 corrects the subsequent lesion planning point based on the position deviation. That is, after the first needle puncture is performed based on the corrected needle insertion puncture position, it is found that there is still a certain deviation between the puncture point and the lesion planning point, then the needle insertion correction module 23 will obtain the deviation information and correct the subsequent needle insertion puncture position of the lesion planning point based on the deviation information. That is, after the needle insertion puncture position is obtained based on the needle insertion correction model, needle puncture can be performed. At this time, if it is found that there is still a certain deviation in the corrected needle insertion puncture position, the needle insertion puncture position of the subsequent lesion planning point is promptly corrected according to the current deviation information.

[0066] In some embodiments, the needle insertion correction module 23 plans the needle path based on the prostate ultrasound image data, avoiding non-punctureable areas, including the urethra and pubic bone. Specifically, during the automatic needle path calculation process, the needle insertion correction module 23 can identify and avoid key areas such as the urethra and pubic bone based on intraoperative prostate ultrasound image data, providing more personalized and flexible positioning capabilities, thereby reducing the incidence of puncture-related complications and puncture failure rates.

[0067] Therefore, the prostate precision puncture system provided by the present application can automatically obtain the patient's anal position through the anal detection module, and allow the ultrasonic probe to smoothly enter the body and the robotic arm needle insertion point to fit the perineum, which improves the difficulty of manually placing the probe and the robotic arm posture, and is more convenient to operate. Afterwards, the present application automatically corrects the puncture position based on the needle insertion parameters through the needle insertion correction module, improves the accuracy of prostate sampling, maximizes the sampling of suspicious lesions, improves the sensitivity and specificity of prostate puncture, and avoids multiple unnecessary punctures. At the same time, the present application can avoid anatomical structures such as the urethra and pubic bone, and reduce the incidence of complications related to prostate puncture and the puncture failure rate. In addition, the present application can realize the automation and intelligence of prostate puncture surgery, reduce the reliance on the surgeon's experience, reduce the doctor's workload during surgery and possible preoperative misjudgment problems, and greatly improve the efficiency and accuracy of puncture.

[0068] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices or methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of modules / units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules or units, which can be electrical, mechanical or other forms.

[0069] The modules / units described as separate components may or may not be physically separate, and the components displayed as modules / units may or may not be physical modules, that is, they may be located in one place or distributed across multiple network elements. Some or all of the modules / units may be selected according to actual needs to achieve the purpose of the embodiments of the present application. For example, the functional modules / units in the various embodiments of the present application may be integrated into a processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into a single module / unit.

[0070] Those skilled in the art should further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the composition and steps of each example according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0071] The descriptions of the processes or structures corresponding to the above figures have different emphases. For parts that are not described in detail in a certain process or structure, please refer to the relevant descriptions of other processes or structures.

[0072] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Anyone skilled in the art may modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by one of ordinary skill in the art without departing from the spirit and technical concepts disclosed in this application shall be covered by the claims of this application.

Claims

1. A prostate precision puncture system, characterized in that: The system comprises: An image acquisition module is configured to acquire anal image data of a patient; an anus detection module, configured to receive the anus image data and detect the anus position of the patient in real time based on the anus image data; a posture adjustment module, configured to control the robotic arm to perform posture adjustment so that the robotic arm pushes the ultrasound probe through the anus to obtain prostate ultrasound image data; The needle insertion correction module is configured to obtain the needle insertion puncture position based on the needle insertion parameters and transmit the needle insertion puncture position to the posture adjustment module, so that the posture adjustment module performs needle puncture on the prostate lesion planning point based on the needle insertion puncture position and the prostate ultrasound image data.

2. The prostate precision puncture system according to claim 1, characterized in that: The needle insertion correction module includes a needle insertion correction model, and the needle insertion parameters serve as inputs to the needle insertion correction model; The needle insertion correction model is configured to obtain the needle insertion deviation angle and position deviation distance through a fully connected layer and a feature cross layer, and obtain the needle insertion puncture position based on the needle insertion deviation angle and the position deviation distance.

3. The prostate precision puncture system according to claim 1, characterized in that: The image acquisition module is further configured to: divide the anus image data into equal areas and determine a set area, so that the set area is fixedly corresponding to the ultrasound probe.

4. The prostate precision puncture system according to claim 2, characterized in that: The anus detection module is further configured to: determine the anus position in the partition position of the anus image data, and judge whether the partition position is consistent with the set area; if not, feedback is sent to the posture adjustment module so that the posture adjustment module controls the robotic arm to adjust its posture until the partition position is consistent with the set area.

5. The prostate precision puncture system according to claim 1, characterized in that: The anus detection module includes an anus detection model, and the anus image data serves as input to the anus detection model; The anus detection model is configured to perform feature extraction on the anus image data to obtain a feature map, downsample the feature map and splice the feature map; and perform non-maximum suppression on the spliced ​​feature map and return a probability matrix to obtain the anus position.

6. The prostate precision puncture system according to claim 1, characterized in that: The needle insertion correction module is further configured to obtain a positional deviation between the needle insertion puncture position and the lesion planning point, and to correct a subsequent needle insertion puncture position of the lesion planning point based on the positional deviation.

7. The prostate precision puncture system according to claim 1, characterized in that: The posture adjustment module is also configured to determine whether the infrared distance between the robotic arm and the patient's perineum meets a distance threshold; when the infrared distance meets the threshold condition, the posture adjustment module controls the robotic arm to push the ultrasound probe through the anus position and obtain the prostate ultrasound image data in real time.

8. The prostate precision puncture system according to claim 1, characterized in that: The needle insertion correction module is further configured to plan a needle track based on the prostate ultrasound image data so that the needle track avoids a non-punctureable area, wherein the non-punctureable area includes the urethra and the pubic bone.

9. The prostate precision puncture system according to claim 1, characterized in that: The posture adjustment module determines the base of the prostate through the ultrasound probe; The posture adjustment module uses the base portion as the insertion depth of the ultrasound probe.

10. The prostate precision puncture system according to claim 1, characterized in that: The needle insertion parameters include: patient's age, prostate specific antigen, weight, prostate tissue shape, prostate tissue volume, prostate ultrasound image development level and skin hardness.

Images