Autonomous ultrasonic robot for full-automatic scanning of thyroid gland

Abstract

The utility model discloses an autonomous ultrasonic robot for full-automatic scanning of thyroid gland. The ultrasonic robot comprises a depth camera, a working chair, an ultrasonic probe, an ultrasonic probe socket, a force sensor, a mechanical arm, a mechanical arm control machine and an ultrasonic control panel. A body surface thyroid positioning module, an image quality optimization module, a robot control module and a real-time image analysis module are arranged in the ultrasonic control panel. The autonomous ultrasonic robot for full-automatic scanning of the thyroid gland can identify and estimate the position of the thyroid gland, simulates a scanning technique of a doctor to carry out transverse and longitudinal cutting ultrasonic scanning, and carries out discriminant analysis on an obtained ultrasonic image. According to the thyroid scanning device, high-quality images can be obtained in the collecting process, comfort and safety of a patient are guaranteed through force feedback, a thyroid target area is scanned under the condition that the safety of the patient is guaranteed finally, human assistance is not needed, and the ultrasonic image quality close to that of professional doctors is obtained.

Description

Technical Field

[0001] This utility model relates to the field of medical testing technology, specifically to an autonomous ultrasound robot for fully automated thyroid scans. Background Technology

[0002] Ultrasound diagnosis boasts significant advantages such as being non-invasive, real-time, portable, and low-cost, making it an indispensable part of modern medical imaging technology. However, the ultrasound examination process can place doctors in uncomfortable positions, and the repetitive nature of these procedures can lead to severe fatigue and injury. Joint and muscle damage in ultrasound physicians can be exacerbated by the constant pressure applied by the probe during the examination. Furthermore, the relatively flexible nature of ultrasound scanning methods can result in inconsistent results due to non-standardized procedures and operations. Simultaneously, ultrasound examinations demand high levels of medical image analysis skills and clinical experience from the operator; different doctors often employ different techniques, leading to significant variability in results and inconsistent diagnoses, thus lacking universal applicability. Therefore, to reduce the workload of doctors, minimize repetitive scanning operations, improve examination efficiency, standardize procedures, save labor costs, and allow doctors to focus more on ultrasound diagnosis and ultrasound-guided treatment during surgery, ensuring optimal surgical outcomes, it is essential to conduct research on the application of robots in ultrasound examinations.

[0003] Ultrasound robots can perform long-duration, high-intensity work, reducing the stress and muscle strain on doctors. For example, doctors need to apply greater pressure to obtain high-quality images when scanning patients with high body mass indexes, which can cause muscle strain. Compared to handheld ultrasound probes, robots offer higher positioning accuracy and more precise adjustment of contact force at the end effector, which is beneficial for acquiring high-quality ultrasound images. More importantly, ultrasound robots are highly adaptable to different environments and can replace doctors in environments with radiation or high risk of infection. Ultrasound robots often have sufficient degrees of freedom to ensure that the robotic arm can move the ultrasound probe to the desired position as required; the accuracy of force feedback and the flexibility of the robotic arm provide significant advantages. Autonomous medical ultrasound robots have great potential for applications in reducing the workload of ultrasound physicians, improving medical services, producing standardized diagnostic results, and avoiding direct contact with patients.

[0004] Current ultrasound robot technology has made significant progress in clinical applications. For example, Hefei Hebin Intelligent Robot Co., Ltd. has published a patent for "An Ultrasound Diagnostic Robot" (application number: CN202110154601.8). This robot plays a positive role in reducing the workload of medical workers, but it still has some key shortcomings, especially in the control of output force during ultrasound scanning. Traditional ultrasound robots fail to effectively monitor and adjust the pressure and contact force applied to the patient's body during scanning. This means that during ultrasound examination, the robot cannot sense in real time whether the applied force is too large or too small, thus failing to ensure patient comfort and safety. Excessive pressure may damage the patient's skin or soft tissue, or even affect the clarity of the ultrasound image; while insufficient pressure may lead to insufficient contact, affecting image quality, or even missing potential diagnostic information. Due to the lack of a suitable force feedback mechanism, this inability to precisely control the output force also leaves considerable room for improvement in the safety and operational accuracy of current ultrasound robot technology. Utility Model Content

[0005] In view of the needs of the aforementioned practical problems and the existing technology involved, the purpose of this utility model is to provide an autonomous ultrasound robot for fully automated thyroid scans. This ultrasound robot can detect the coordinate data of the patient's thyroid gland in real time, control the robotic arm to automatically approach the thyroid gland and perform transverse and longitudinal scans, and use a neural network model to detect abnormal lesions in the scanned images, identifying and locating thyroid nodules. During the ultrasound scan, force and torque data collected by force sensors are used to control the force output of the robotic arm, ensuring the safety of the scan process. This utility model reduces the workload of doctors and improves the recognition rate of thyroid nodules.

[0006] The objective of this utility model is achieved through at least one of the following technical solutions.

[0007] An autonomous ultrasound robot for fully automated thyroid scans includes a depth camera, a work chair, an ultrasound probe, an ultrasound probe socket, a force sensor, a robotic arm, and an ultrasound control board.

[0008] Force sensors are mounted at the end joints of the robotic arm. The force sensors are physically connected to the ultrasonic probe socket and connected to the ultrasonic control board via a data transmission line. The ultrasonic probe is mounted on the ultrasonic probe socket. The robotic arm is connected to the ultrasonic control board via wires. The depth camera is connected to the ultrasonic control board via a data cable.

[0009] A depth camera performs visual localization and acquires point cloud data of the neck, collecting the patient's neck coordinate data and sending it to the ultrasound control board. The ultrasound control board processes the received neck coordinate data to obtain control commands for the robotic arm, which is then sent to the robotic arm to move closer to the thyroid region of the patient's neck. After the ultrasound probe transmits preliminary image data back to the ultrasound control board, the board sends control commands to move the robotic arm, causing the ultrasound probe to perform transverse and longitudinal scans to obtain transverse and longitudinal images of the thyroid gland. After obtaining the transverse and longitudinal images of the thyroid gland, the ultrasound control board uses the acquired image data as input to a pre-trained neural network model for detection, identification, and localization of thyroid nodules.

[0010] Furthermore, the end of the robotic arm is equipped with a socket for mounting a force sensor, and the force sensor has a slot for connecting an ultrasonic probe.

[0011] Furthermore, the force sensor, ultrasonic probe, depth camera, and robotic arm are connected to the ultrasonic control board via data transmission lines.

[0012] Furthermore, the depth camera has stereo vision capabilities to provide three-dimensional coordinate data for precise localization of the thyroid region in the patient's neck.

[0013] Furthermore, the robotic arm has 6 degrees of freedom, enabling flexible movement in multiple directions to adapt to the anatomical structure of different patients' necks, ensuring that the ultrasound probe can fully cover the thyroid region.

[0014] Furthermore, the ultrasonic system consisting of the ultrasonic probe and the ultrasonic control board has a depth range of 18mm to 184mm and a frequency range of 6.5MHz to 10MHz, which meets the requirements of the scanning task.

[0015] Furthermore, the ultrasound control board has an image processing function for preprocessing the image data transmitted back by the ultrasound probe, including filtering, enhancement, and noise reduction, to improve image quality and facilitate subsequent nodule identification.

[0016] Furthermore, the neural network model is trained using deep learning to identify various types of thyroid nodules and classify them based on nodule characteristics, including size, shape, and echo characteristics.

[0017] Furthermore, during the ultrasonic scanning process, the force sensor returns the force and torque data of the robotic arm in real time, ensuring that the output force is within the set range and that the output force is stable, thereby achieving safe control during the scanning process.

[0018] Furthermore, it includes a robotic arm controller, which is connected to the robotic arm via wires to directly control the robotic arm. This controller is used for equipment debugging and ensures that all parts of the robotic arm are working properly.

[0019] The robot features a modular design, allowing for the rapid replacement and upgrading of components such as ultrasonic probes, force sensors, and depth cameras to adapt to technological advancements and maintenance needs.

[0020] Compared with existing technologies, the advantages of this utility model are:

[0021] This invention integrates advanced force sensing technology and a feedback control mechanism to monitor the contact force applied during ultrasound scanning in real time and automatically adjust its magnitude. The core advantage of this technology lies in its ability to precisely sense and regulate the pressure applied by the robot during scanning, ensuring that the pressure remains within a safe and comfortable range. Through the force feedback system, the robot can adaptively adjust the scanning force according to the patient's actual condition, avoiding unnecessary damage to the skin and soft tissues, thereby ensuring patient safety. Simultaneously, precise force control also helps improve the quality of ultrasound images, as stable and appropriate pressure ensures more uniform contact between the sensor and the skin, resulting in clearer and more accurate diagnostic images.

[0022] This invention can ensure high-quality images during the acquisition process and ensure patient comfort and safety through force feedback. Ultimately, it can scan the target area of ​​the thyroid gland under the condition of ensuring patient safety without human assistance, and obtain ultrasound image quality that is close to that of a professional doctor. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of the structure of an autonomous ultrasound robot for fully automated thyroid scan in an embodiment of the present invention;

[0024] Figure 2 This is a control principle diagram of an autonomous ultrasound robot for fully automated thyroid scan in an embodiment of the present invention;

[0025] Figure 3 This is a schematic diagram of the workflow in an embodiment of the present utility model. Detailed Implementation

[0026] To more clearly illustrate the technical solutions in the embodiments of this utility model or the prior art, the accompanying drawings and technical solutions used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this utility model. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. The specific implementation of this utility model will be further described below in conjunction with the accompanying drawings.

[0027] The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to be limiting of this specification. The singular forms “a,” “described,” and “the” as used in one or more embodiments of this specification and in the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in one or more embodiments of this specification refers to and includes any or all possible combinations of one or more associated listed items.

[0028] It should be understood that although terms such as "first," "second," etc., may be used to describe various information in one or more embodiments of this specification, such information should not be limited to these terms. These terms are used only to distinguish similar objects and are not necessarily used to indicate the order or sequence of features described in one or more embodiments of this specification. Furthermore, the terms "possessing," "comprising," and similar expressions are intended to indicate coverage of a non-exclusive scope. For example, a process, method, system, product, or device that includes a series of steps or modules is not necessarily limited to the detailed list but may include inherent content related to those steps or modules that is not listed.

[0029] Example:

[0030] An autonomous ultrasound robot for fully automated thyroid scans, such as Figure 1 As shown, it includes a depth camera 1, a work chair 2, an ultrasonic probe 3, an ultrasonic probe socket 4, a force sensor 5, a robotic arm 6, and an ultrasonic control board 8;

[0031] Force sensor 5 is mounted at the end joint of robotic arm 6. Force sensor 5 is physically connected to ultrasonic probe socket 4 and connected to ultrasonic control board 8 through data transmission line. Ultrasonic probe 3 is mounted on ultrasonic probe socket 4. Robotic arm 6 is connected to ultrasonic control board 8 through wires. Depth camera 1 is connected to ultrasonic control board 8 through data cable.

[0032] Depth camera 1 performs visual positioning and acquires neck point cloud data, collects the patient's neck coordinate data, and sends the neck coordinate data to ultrasound control board 8. Ultrasound control board 8 calculates the received patient's neck coordinate data to obtain control commands for robotic arm 6, and sends control commands to robotic arm 6 for execution, causing robotic arm 6 to approach the patient's thyroid region. After ultrasound probe 3 returns preliminary image data to ultrasound control board 8, ultrasound control board 8 sends control commands to move robotic arm 6, driving ultrasound probe 3 to perform transverse and longitudinal scans to obtain transverse and longitudinal images of the thyroid gland. After obtaining the transverse and longitudinal images of the thyroid gland, ultrasound control board 8 uses the acquired image data as input to a pre-trained neural network model for detection, identification, and localization of thyroid nodules.

[0033] In one embodiment, the robotic arm 6 is a Universal Robots UR3 robotic arm from Denmark, with a payload of 3 kg, capable of performing the scanning task in this scenario. The end of the robotic arm 6 is equipped with a socket for mounting a force sensor 5, and the force sensor 5 has a slot for connecting an ultrasonic probe 3.

[0034] Furthermore, the force sensor 5, ultrasonic probe 3, depth camera 1, and robotic arm 6 are connected to the ultrasonic control board 8 via data transmission lines.

[0035] In one embodiment, the depth camera 1 is a Microsoft Azure Kinect DK RGBD camera with stereo vision capabilities, used to provide three-dimensional coordinate data to achieve precise positioning of the thyroid region in the patient's neck.

[0036] Furthermore, the robotic arm 6 has 6 degrees of freedom, enabling flexible movement in multiple directions to adapt to the anatomical structure of different patients' necks, ensuring that the ultrasound probe 3 can cover the thyroid region in all directions.

[0037] In one embodiment, the ultrasonic probe 3 and the ultrasonic control board 8 are selected from the Sonostar C5 laptop color Doppler ultrasonic system. The ultrasonic system composed of the ultrasonic probe 3 and the ultrasonic control board 8 has a depth range of 18mm to 184mm and a frequency range of 6.5MHz to 10MHz, which meets the requirements of the scanning task.

[0038] Furthermore, the ultrasound control board 8 has an image processing function, which is used to preprocess the image data transmitted back by the ultrasound probe 3, including filtering, enhancement and noise reduction, to improve image quality and facilitate subsequent nodule identification.

[0039] In one embodiment, the neural network model uses UNet++ and is trained with deep learning to identify multiple types of thyroid nodules and classify them based on nodule characteristics, including size, shape, and echo characteristics.

[0040] In one embodiment, the force sensor 5 is a Robotiq 6D force / torque sensor, which returns the force and torque data of the robotic arm 6 in real time during ultrasonic scanning, ensuring that the output force is within the set range and that the output force is stable, thereby achieving safe control during the scanning process.

[0041] Furthermore, it includes a robotic arm controller 7, which is connected to the robotic arm 6 via wires to directly control the robotic arm 6 for equipment debugging and to ensure that all parts of the robotic arm 6 work normally.

[0042] The robot features a modular design, allowing for the rapid replacement and upgrading of components such as the ultrasonic probe 3, force sensor 5, and depth camera 1 to adapt to technological advancements and maintenance needs.

[0043] In one embodiment, such as Figure 2 and Figure 3 As shown, when the patient sits on the work chair 2, the depth camera 1 automatically collects the coordinate data of the patient's thyroid gland in the neck. The robotic arm 6 moves to the vicinity of the thyroid gland according to the coordinates to perform a scan, and the ultrasound control board 8 detects thyroid nodules based on the images. During the ultrasound scan, the ultrasound control board 8 controls the force output range of the robotic arm by analyzing the data from the force sensor 5 to ensure safety.

[0044] In one embodiment, when a patient has thyroid nodules, an ultrasound robot can automatically scan the thyroid region, obtaining and collecting transverse and longitudinal scan images. The collected ultrasound image data is input into a neural network for analysis to determine the location of the thyroid nodules, providing a basis for further diagnosis by the doctor.

[0045] This invention integrates advanced force sensing technology and a feedback control mechanism to monitor the contact force applied during ultrasound scanning in real time and automatically adjust its magnitude. The core advantage of this technology lies in its ability to precisely sense and regulate the pressure applied by the robot during scanning, ensuring that the pressure remains within a safe and comfortable range. Through the force feedback system, the robot can adaptively adjust the scanning force according to the patient's actual condition, avoiding unnecessary damage to the skin and soft tissues, thereby ensuring patient safety. Simultaneously, precise force control also helps improve the quality of ultrasound images, as stable and appropriate pressure ensures more uniform contact between the sensor and the skin, resulting in clearer and more accurate diagnostic images.

[0046] The above description is merely a specific embodiment of this utility model, but the protection scope of this utility model is not limited thereto. Any changes or substitutions conceived without inventive effort should be included within the protection scope of this utility model. Therefore, the protection scope of this utility model should be determined by the scope defined in the claims.

Claims

1. An autonomous ultrasound robot for full automatic scanning of the thyroid gland, characterized in that, The system comprises a depth camera (1), a work chair (2), an ultrasonic probe (3), an ultrasonic probe socket (4), a force sensor (5), a mechanical arm (6), and an ultrasonic control board (8). The force sensor (5) is installed at the end joint of the mechanical arm (6), and is physically connected to the ultrasonic probe socket (4) and connected to the ultrasonic control board (8) through a data transmission line. The ultrasonic probe (3) is installed on the ultrasonic probe socket (4). The mechanical arm (6) is connected to the ultrasonic control board (8) through a wire. The depth camera (1) is connected to the ultrasonic control board (8) through a data line. The depth camera (1) performs visual positioning and obtains point cloud data of the neck, collects the neck coordinate data of the patient, and sends the neck coordinate data to the ultrasonic control board (8). The ultrasonic control board (8) calculates the received neck coordinate data of the patient, obtains the control instruction of the mechanical arm (6), and sends the control instruction to the mechanical arm (6) for execution, so that the mechanical arm (6) approaches the thyroid region of the patient's neck. After the ultrasonic probe (3) returns the preliminary image data to the ultrasonic control board (8), the ultrasonic control board (8) sends a control instruction to move the mechanical arm (6) to drive the ultrasonic probe (3) to perform transverse and longitudinal scanning to obtain thyroid transverse and longitudinal images. After obtaining the thyroid transverse and longitudinal images, the ultrasonic control board (8) transmits the collected image data as input to the pre-trained neural network model for detection, identification, and positioning of thyroid nodules.

2. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, characterized in that, The mechanical arm (6) is provided with a jack at the end for installing the force sensor (5). The force sensor (5) is provided with a slot for connecting the ultrasonic probe (3).

3. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The force sensor (5), the ultrasonic probe (3), the depth camera (1), and the mechanical arm (6) are respectively connected to the ultrasonic control board (8) through data transmission lines.

4. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The depth camera (1) has a stereo vision function for providing three-dimensional coordinate data to achieve accurate positioning of the thyroid region of the patient's neck.

5. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The mechanical arm (6) has 6 degrees of freedom for flexible movement in multiple directions to adapt to the anatomical structure of different patients' necks, ensuring that the ultrasonic probe (3) can cover the thyroid region in all directions.

6. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The ultrasonic wave system composed of the ultrasonic probe (3) and the ultrasonic control board (8) has a depth range of 18mm to 184mm and a frequency range of 6.5MHz to 10MHz, meeting the requirements of the scanning task.

7. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The ultrasonic control board (8) has an image processing function for preprocessing the image data returned by the ultrasonic probe (3), including filtering, enhancement, and noise reduction to improve image quality and facilitate subsequent nodule identification.

8. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, characterized in that, The neural network model is trained through deep learning to identify various types of thyroid nodules and classify them based on their characteristics, including size, shape, and echo characteristics.

9. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, wherein, The force sensor (5) returns real-time force and torque data of the mechanical arm (6) during ultrasonic scanning, ensuring that the output force is within a set range and stable, achieving safety control during scanning.

10. An autonomous ultrasound robot for full automatic scanning of thyroid according to claim 1, characterized in that, The mechanical arm control machine (7) is connected with the mechanical arm (6) through a wire, directly controls the mechanical arm (6), is used for debugging work of the equipment, and ensures that each part of the mechanical arm (6) works normally.

Images