Cooperative control system of breast cancer diagnosis and treatment integrated robot

Abstract

The invention relates to the technical field of medical diagnosis auxiliary equipment, in particular to a cooperative control system of a breast cancer diagnosis and treatment integrated robot. According to the technical scheme, the cooperative control system of the breast cancer diagnosis and treatment integrated robot comprises an image acquisition and processing module, a focus positioning and navigation module, a diagnosis decision support module, a treatment planning and execution module, a real-time monitoring and feedback module and a cooperative control center module; according to the system, the cooperative control center is constructed, the diagnosis decision support module and the treatment planning and execution module which are originally independent from each other are combined, the system can automatically and directly convert multiple features output by the diagnosis module into specific instructions, the risk of misinformation transmission or delay caused by human factors is greatly reduced, and the accuracy of diagnosis is improved. Therefore, a data-driven closed loop without manual intervention is formed, and it is effectively ensured that the diagnosis scheme can be completely executed.

Description

Technical Field

[0001] This invention relates to the field of medical diagnostic auxiliary equipment technology, and in particular to an integrated robot collaborative control system for breast cancer diagnosis and treatment. Background Technology

[0002] The integrated robotic collaborative control system for breast cancer diagnosis and treatment is an intelligent medical platform that integrates artificial intelligence image recognition, multimodal robot collaboration, and real-time dynamic control. By integrating a breast ultrasound / mammography image automatic analysis module, a minimally invasive surgical robot execution unit, and a central control algorithm, it achieves full automation and seamless connection from early tumor screening, precise localization, pathological biopsy to targeted therapy. While reducing human error, it significantly improves the efficiency of breast cancer diagnosis and treatment and the accuracy of minimally invasive surgery through multi-robot collaborative operation and closed-loop feedback mechanism, ultimately forming an integrated closed-loop medical solution of "diagnosis-treatment-follow-up".

[0003] Existing integrated robotic systems for breast cancer diagnosis and treatment generally suffer from the problem that after the diagnostic module completes image acquisition and analysis, the data generated is usually sent to the treatment module in the form of a static report. This involves a step that requires manual interpretation and re-entry, which leads to the loss of a large amount of background information that could improve treatment accuracy. The overall response speed and decision consistency of the system are highly dependent on the operator's experience and real-time judgment, which can easily lead to delays and the risk of human error.

[0004] To address the aforementioned issues, this solution constructs a collaborative control center that integrates the previously independent diagnostic decision support module with the treatment planning and execution module. The system can automatically convert multiple features output by the diagnostic module into specific instructions, greatly reducing the risk of information mistransmission or delays caused by human factors. This forms a data-driven closed loop that requires no human intervention, effectively ensuring that the diagnostic plan can be executed completely. Summary of the Invention

[0005] To overcome the common problems in existing integrated robotic systems for breast cancer diagnosis and treatment, where the diagnostic module typically sends data to the treatment module in the form of static reports after completing image acquisition and analysis, which involves a step requiring manual interpretation and re-entry, resulting in the loss of a large amount of background information that could improve treatment accuracy, and where the overall system response speed and decision consistency are highly dependent on the operator's experience and real-time judgment, easily leading to delays and the risk of human error.

[0006] The technical solution of this invention is: an integrated robotic collaborative control system for breast cancer diagnosis and treatment, comprising the following modules: Image acquisition and processing module: used to acquire breast medical images and perform image processing and quality enhancement; Lesion localization and navigation module: used to accurately calculate the location of lesions based on image data and plan the robot's movement path; Diagnostic decision support module: used to analyze image features, assist doctors in classifying lesions as benign or malignant, and generate diagnostic reports; Treatment planning and execution module: used to plan treatment parameters based on diagnostic results and control the robot to perform treatment operations; Real-time monitoring and feedback module: used to monitor the patient's physiological parameters and treatment effects, and provide feedback to dynamically adjust the treatment; Collaborative Control Center Module: Used to integrate all modules.

[0007] Preferably, the image acquisition and processing module includes: A11: Image acquisition unit, including an ultrasonic probe, an optical camera, and an infrared sensor, used to acquire raw image data using multiple sensors; A12: Image processing unit, including image processing chip, GPU and memory module, used for denoising, enhancing and segmenting acquired images; A13: Data storage unit, including hard disk drives, solid-state drives and cloud storage interfaces, for storing raw and processed image data.

[0008] As a preferred option, the lesion localization and navigation module includes: A21: Positioning unit, including a laser rangefinder, inertial measurement unit, and GPS receiver, used to calculate the three-dimensional coordinates of the lesion through multi-sensor fusion; A22: Path planning unit, including path planning processor, obstacle avoidance sensors, and map building software, used to generate safe and efficient movement paths to avoid obstacles; A23: Navigation execution unit, including motor drivers, wheels or robotic arms and encoders, for controlling the robot to move along a planned path.

[0009] Preferably, the lesion localization and navigation module includes the following steps when it is in operation: S11: The laser rangefinder performs self-calibration, the inertial measurement unit performs zero-position offset compensation, and the GPS receiver receives satellite signals to determine the robot's rough initial position; S12: A laser rangefinder scans the contours of the breast tissue, an inertial measurement unit monitors the robot's posture changes in real time, and an optical camera captures visual markers. The data from these three sources are then synchronized to the positioning unit processor. S13: Perform precise matrix transformation and registration on the lesion image coordinates provided by the image acquisition module, the robot's own coordinate system, and the patient's world coordinate system; S14: Based on the registered coordinate system, the precise coordinates of the lesion center point in three-dimensional space are calculated by fusing laser ranging and image depth data and using a triangulation algorithm; S15: The path planning processor plans a collision-free initial movement path in the environment model generated by the map building software based on the 3D coordinates of the lesion and the robot's current position; S16: As the robot moves along the initial path, the obstacle avoidance sensor continuously detects obstacles ahead. Once an obstacle is detected, path replanning is immediately triggered to generate a local detour path. S17: The navigation execution unit converts the final path command into pulse signals, which drive the motor driver to control the rotation speed and direction of the wheels or robotic arm, enabling the robot to move towards the target position; S18: The encoder provides real-time feedback on the actual rotation angle and displacement of the motor or robotic arm joint, compares it with the commanded position, and forms a closed-loop control to eliminate accumulated errors.

[0010] Preferably, the diagnostic decision support module includes: A31: Feature extraction unit, including a feature extraction algorithm processor, a neural network accelerator, and cache memory, for automatically extracting key features from images; A32: Classification and judgment unit, including a classifier chip, a decision tree processor, and a probability calculation unit, used for intelligent classification of lesions based on feature data; A33: Report generation unit, including report generation software, display and printer interfaces, for automatically generating structured diagnostic reports.

[0011] Preferably, the diagnostic decision support module includes the following steps when it is in operation: S21: Retrieve the enhanced breast image from the data storage unit of the image acquisition and processing module, and perform standardized scaling and grayscale normalization on it; S22: Using the segmentation algorithm in the feature extraction unit, the outline of the suspected lesion area is automatically delineated on the preprocessed image, separating it from normal tissue; S23: The feature extraction algorithm processor calculates the morphological parameters of the segmented lesion region, including but not limited to the degree of edge burrs, aspect ratio, area, and perimeter; S24: The neural network accelerator performs texture analysis on the lesion region, extracts the contrast, correlation, and entropy values ​​of its internal gray-level co-occurrence matrix, and calculates the average density of the region; S25: Combine all calculated morphological, texture, and density feature parameters in a predetermined order to form a multidimensional feature vector data set; S26: The constructed multidimensional feature vector is input into the pre-trained classifier chip, and the classifier is calculated based on the built-in decision tree or support vector machine model; S27: The probability calculation unit outputs the probability values ​​of whether the lesion is benign or malignant based on the calculation results of the classifier; S28: The report generation software automatically fills in the preset report template based on the classification probability and feature quantification results, including lesion location, size, feature description and classification suggestions.

[0012] As a preferred option, the treatment planning and execution module includes: A41: Treatment planning unit, including a planning software processor, parameter optimizer, and user input interface, for developing personalized treatment plans; A42: Execution control unit, including robot arm controller, treatment tool, and force sensor, for driving the treatment tool to perform precise operations; A43: Safety monitoring unit, including safety relays, emergency stop buttons and limit switches, for real-time monitoring of the treatment process.

[0013] Preferably, the treatment planning and execution module includes the following steps when it is in operation: S31: The treatment planning unit receives the malignancy probability report and precise three-dimensional spatial coordinates of the lesion from the diagnostic decision support module and the lesion localization and navigation module, respectively; S32: The planning software processor accurately delineates the target volume and central target point that need to be removed or sampled on the 3D image based on the lesion coordinates and size; S33: Based on the treatment type, provide users with device selection and plan the optimal straight or curved path for the treatment tool from the skin entry point to the lesion target point; S34: The parameter optimizer automatically calculates the key parameters required for treatment based on the lesion characteristics and the planned path, including but not limited to the insertion depth of the biopsy needle, the power of the ablation energy, and the duration of action; S35: The safety monitoring unit performs collision detection between the planned path and the patient's anatomical structure model, and verifies whether the treatment parameters are within the preset safety threshold range; S36: The execution control unit sends a command to the robot arm controller, driving the robotic arm to move with the treatment tool to the planned skin entry point; S37: The robotic arm controller controls the robotic arm to insert the biopsy needle to the target depth according to the planned path and speed, or to attach the ablation probe to the lesion and apply the preset treatment energy; S39: The force sensor provides real-time feedback on the resistance encountered by the needle tip or probe, and the encoder continuously monitors the position of the robotic arm joints, stopping when the threshold is exceeded.

[0014] As a preferred option, the real-time monitoring and feedback module includes: A51: Physiological parameter monitoring unit, including an electrocardiogram sensor, a blood pressure monitor, and a blood oxygen saturation sensor, for real-time monitoring of the patient's vital signs; A52: Treatment efficacy assessment unit, including a real-time image processor, efficacy assessment algorithm, and feedback display, for assessing treatment progress through real-time imaging; A53: Adaptive adjustment unit, including adaptive controller, parameter adjuster and actuator, used to automatically adjust treatment parameters or pathways based on feedback data.

[0015] As a preferred option, the collaborative control center module includes: A61: Data fusion unit, including a data fusion processor, communication bus, and data buffer, is used to fuse multi-source data and provide a unified view; A62: Task scheduling unit, including a task scheduler, real-time operating system, and priority manager, is used to coordinate the execution order of tasks in various modules and optimize resource allocation; A63: User interaction unit, including touch screen, keyboard and voice recognition module, for providing an intuitive interface.

[0016] The beneficial effects of this invention are: 1. Existing integrated robotic systems for breast cancer diagnosis and treatment generally suffer from the problem that after the diagnostic module completes image acquisition and analysis, the data generated is usually sent to the treatment module in the form of a static report. This involves a step that requires manual interpretation and re-input, resulting in the loss of a large amount of background information that could improve treatment accuracy. The overall response speed and decision consistency of the system are highly dependent on the operator's experience and real-time judgment, which can easily lead to delays and the risk of human error. This solution combines the previously independent diagnostic decision support module with the treatment planning and execution module by constructing a collaborative control center. The system can automatically convert multiple features output by the diagnostic module directly into specific instructions, which greatly reduces the risk of information mistransmission or delay caused by human factors. This forms a data-driven closed loop that does not require human intervention, effectively ensuring that the diagnostic plan can be executed completely. 2. Existing systems, when performing treatment operations, mostly rely on preset, fixed treatment parameters and mechanical paths. They lack the power to make substantial adjustments to parameters or correct paths based on real-time feedback during treatment. Furthermore, monitoring functions are mostly limited to triggering emergency stops when safety thresholds are exceeded. When the treatment tool encounters unexpected changes in tissue density or when the lesion location shifts due to minor physiological movements of the patient, traditional systems cannot detect these changes or compensate for them; they simply continue executing the original instructions or simply terminate the procedure, making it difficult to guarantee treatment accuracy. This solution addresses this by embedding a real-time monitoring and feedback module into the system. Throughout the control loop, a dedicated algorithm analyzes real-time images to quantitatively assess key indicators such as whether the ablation area covers the target and whether the lesion morphology has changed as expected. Simultaneously, all real-time data is continuously fed to the adaptive adjustment unit. Once a slight deviation trend is detected, the system can proactively intervene. When the force sensor detects an abnormal increase in tissue resistance, the system will immediately fine-tune the feed force or speed of the robotic arm. When the real-time images show that the ablation zone is expanding in a biased manner, the system can dynamically adjust the energy release focus or action time of the ablation probe, significantly improving the intelligent fault tolerance and dynamic precision of the treatment process, making the entire treatment process more closely aligned with the patient's individual physiological dynamics. Attached Figure Description

[0017] Figure 1 The diagram shown is a schematic flowchart of the integrated robot collaborative control system for breast cancer diagnosis and treatment according to the present invention. Figure 2 The diagram shown is a schematic of the workflow of the lesion localization and navigation module of the integrated robot collaborative control system for breast cancer diagnosis and treatment according to the present invention. Figure 3 The diagram shown is a schematic of the workflow of the diagnostic decision support module of an integrated robotic collaborative control system for breast cancer diagnosis and treatment according to the present invention. Figure 4 The diagram shown is a schematic of the workflow of the treatment planning and execution module of the integrated robot collaborative control system for breast cancer diagnosis and treatment according to the present invention. Detailed Implementation

[0018] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0019] Please see Figure 1-4 This invention provides an embodiment of an integrated robotic collaborative control system for breast cancer diagnosis and treatment, comprising the following modules: Image acquisition and processing module: used to acquire breast medical images and perform image processing and quality enhancement; Lesion localization and navigation module: used to accurately calculate the location of lesions based on image data and plan the robot's movement path; Diagnostic decision support module: used to analyze image features, assist doctors in classifying lesions as benign or malignant, and generate diagnostic reports; Treatment planning and execution module: used to plan treatment parameters based on diagnostic results and control the robot to perform treatment operations; Real-time monitoring and feedback module: used to monitor the patient's physiological parameters and treatment effects, and provide feedback to dynamically adjust the treatment; Collaborative Control Center Module: Used to integrate all modules.

[0020] Preferably, the image acquisition and processing module includes: A11: Image acquisition unit, including an ultrasonic probe, an optical camera, and an infrared sensor, used to acquire raw image data using multiple sensors; A12: Image processing unit, including image processing chip, GPU and memory module, used for denoising, enhancing and segmenting acquired images; A13: Data storage unit, including hard disk drives, solid-state drives and cloud storage interfaces, for storing raw and processed image data.

[0021] As a preferred option, the lesion localization and navigation module includes: A21: Positioning unit, including a laser rangefinder, inertial measurement unit, and GPS receiver, used to calculate the three-dimensional coordinates of the lesion through multi-sensor fusion; A22: Path planning unit, including path planning processor, obstacle avoidance sensors, and map building software, used to generate safe and efficient movement paths to avoid obstacles; A23: Navigation execution unit, including motor drivers, wheels or robotic arms and encoders, for controlling the robot to move along a planned path.

[0022] Preferably, the lesion localization and navigation module includes the following steps when it is in operation: S11: The laser rangefinder performs self-calibration, the inertial measurement unit performs zero-position offset compensation, and the GPS receiver receives satellite signals to determine the robot's rough initial position; S12: A laser rangefinder scans the contours of the breast tissue, an inertial measurement unit monitors the robot's posture changes in real time, and an optical camera captures visual markers. The data from these three sources are then synchronized to the positioning unit processor. S13: Perform precise matrix transformation and registration on the lesion image coordinates provided by the image acquisition module, the robot's own coordinate system, and the patient's world coordinate system; S14: Based on the registered coordinate system, the precise coordinates of the lesion center point in three-dimensional space are calculated by fusing laser ranging and image depth data and using a triangulation algorithm; S15: The path planning processor plans a collision-free initial movement path in the environment model generated by the map building software based on the 3D coordinates of the lesion and the robot's current position; S16: As the robot moves along the initial path, the obstacle avoidance sensor continuously detects obstacles ahead. Once an obstacle is detected, path replanning is immediately triggered to generate a local detour path. S17: The navigation execution unit converts the final path command into pulse signals, which drive the motor driver to control the rotation speed and direction of the wheels or robotic arm, enabling the robot to move towards the target position; S18: The encoder provides real-time feedback on the actual rotation angle and displacement of the motor or robotic arm joint, compares it with the commanded position, and forms a closed-loop control to eliminate accumulated errors.

[0023] Preferably, the diagnostic decision support module includes: A31: Feature extraction unit, including a feature extraction algorithm processor, a neural network accelerator, and cache memory, for automatically extracting key features from images; A32: Classification and judgment unit, including a classifier chip, a decision tree processor, and a probability calculation unit, used for intelligent classification of lesions based on feature data; A33: Report generation unit, including report generation software, display and printer interfaces, for automatically generating structured diagnostic reports.

[0024] Preferably, the diagnostic decision support module includes the following steps when it is in operation: S21: Retrieve the enhanced breast image from the data storage unit of the image acquisition and processing module, and perform standardized scaling and grayscale normalization on it; S22: Using the segmentation algorithm in the feature extraction unit, the outline of the suspected lesion area is automatically delineated on the preprocessed image, separating it from normal tissue; S23: The feature extraction algorithm processor calculates the morphological parameters of the segmented lesion region, including but not limited to the degree of edge burrs, aspect ratio, area, and perimeter; S24: The neural network accelerator performs texture analysis on the lesion region, extracts the contrast, correlation, and entropy values ​​of its internal gray-level co-occurrence matrix, and calculates the average density of the region; S25: Combine all calculated morphological, texture, and density feature parameters in a predetermined order to form a multidimensional feature vector data set; S26: The constructed multidimensional feature vector is input into the pre-trained classifier chip, and the classifier is calculated based on the built-in decision tree or support vector machine model; S27: The probability calculation unit outputs the probability values ​​of whether the lesion is benign or malignant based on the calculation results of the classifier; S28: The report generation software automatically fills in the preset report template based on the classification probability and feature quantification results, including lesion location, size, feature description and classification suggestions.

[0025] As a preferred option, the treatment planning and execution module includes: A41: Treatment planning unit, including a planning software processor, parameter optimizer, and user input interface, for developing personalized treatment plans; A42: Execution control unit, including robot arm controller, treatment tool, and force sensor, for driving the treatment tool to perform precise operations; A43: Safety monitoring unit, including safety relays, emergency stop buttons and limit switches, for real-time monitoring of the treatment process.

[0026] Preferably, the treatment planning and execution module includes the following steps when it is in operation: S31: The treatment planning unit receives the malignancy probability report and precise three-dimensional spatial coordinates of the lesion from the diagnostic decision support module and the lesion localization and navigation module, respectively; S32: The planning software processor accurately delineates the target volume and central target point that need to be removed or sampled on the 3D image based on the lesion coordinates and size; S33: Based on the treatment type, provide users with device selection and plan the optimal straight or curved path for the treatment tool from the skin entry point to the lesion target point; S34: The parameter optimizer automatically calculates the key parameters required for treatment based on the lesion characteristics and the planned path, including but not limited to the insertion depth of the biopsy needle, the power of the ablation energy, and the duration of action; S35: The safety monitoring unit performs collision detection between the planned path and the patient's anatomical structure model, and verifies whether the treatment parameters are within the preset safety threshold range; S36: The execution control unit sends a command to the robot arm controller, driving the robotic arm to move with the treatment tool to the planned skin entry point; S37: The robotic arm controller controls the robotic arm to insert the biopsy needle to the target depth according to the planned path and speed, or to attach the ablation probe to the lesion and apply the preset treatment energy; S39: The force sensor provides real-time feedback on the resistance encountered by the needle tip or probe, and the encoder continuously monitors the position of the robotic arm joints, stopping when the threshold is exceeded.

[0027] As a preferred option, the real-time monitoring and feedback module includes: A51: Physiological parameter monitoring unit, including an electrocardiogram sensor, a blood pressure monitor, and a blood oxygen saturation sensor, for real-time monitoring of the patient's vital signs; A52: Treatment efficacy assessment unit, including a real-time image processor, efficacy assessment algorithm, and feedback display, for assessing treatment progress through real-time imaging; A53: Adaptive adjustment unit, including adaptive controller, parameter adjuster and actuator, used to automatically adjust treatment parameters or pathways based on feedback data.

[0028] As a preferred option, the collaborative control center module includes: A61: Data fusion unit, including a data fusion processor, communication bus, and data buffer, is used to fuse multi-source data and provide a unified view; A62: Task scheduling unit, including a task scheduler, real-time operating system, and priority manager, is used to coordinate the execution order of tasks in various modules and optimize resource allocation; A63: User interaction unit, including touch screen, keyboard and voice recognition module, for providing an intuitive interface.

[0029] Example 1 Background: In the breast surgery center of a cancer hospital, a patient was found to have a small, morphologically suspicious lesion in the upper outer quadrant of her left breast during routine screening. Due to the small size and deep location of the lesion, as well as the complex relationship with the surrounding tissues, traditional ultrasound-guided manual biopsy faced challenges such as inaccurate localization and uncertain sampling success rates. At the same time, if the rapid pathology results indicated malignancy, the patient hoped to have the lesion precisely removed in the same surgery as much as possible to avoid the physical and psychological burden and risks of secondary anesthesia and surgery. To address this challenging clinical need, the medical team decided to adopt the "Integrated Robotic Collaborative Control System for Breast Cancer Diagnosis and Treatment," which aims to achieve seamless integration from diagnosis to treatment through integrated and precise operation, providing patients with efficient, minimally invasive, and accurate diagnostic and treatment services.

[0030] Implementation steps: S41: The patient lies supine with the affected breast naturally flat and stabilized using a special support. The doctor attaches optical positioning markers to the patient's chest wall and around the breast, and activates the integrated robot system. The collaborative control center module initiates a self-test procedure: the ultrasound probe in the image acquisition and processing module performs frequency calibration, and the optical camera and infrared sensor adjust the focus; the laser rangefinder and inertial measurement unit in the lesion localization and navigation module perform zero-position calibration; the robot arm in the treatment planning and execution module performs a no-load positioning movement to confirm that there are no abnormalities in the range of motion of each joint. All modules report normal status through the task scheduling unit of the collaborative control center, and the system enters standby mode. S42: The doctor operates the robotic arm, causing the multimodal sensor of the image acquisition unit to scan the patient's breast. The ultrasound probe acquires a sequence of two-dimensional ultrasound images of the lesion area. The optical camera simultaneously captures the breast surface contour and the position of optical markers. The infrared sensor assists in constructing the three-dimensional shape of the surface. The image processing unit calls on GPU resources to denoise and enhance the ultrasound images, and uses a three-dimensional reconstruction algorithm to synthesize the two-dimensional sequence into three-dimensional volume data of the breast containing the lesion. All raw and processed data are stored in the data storage unit and managed uniformly by the data fusion unit of the collaborative control center. S43: The positioning unit begins operation. The laser rangefinder precisely measures the distance between the probe and the skin surface. The inertial measurement unit tracks the attitude changes of the sensor space. Combined with optical markers, the three-dimensional lesion model generated by the image acquisition module is precisely registered with the patient's actual position. Through multi-sensor data fusion and triangulation algorithms, the precise three-dimensional coordinates of the lesion in the real-world coordinate system are calculated. Subsequently, based on these target coordinates and combined with the patient's body surface and surrounding environment model generated by the map building software, the path planning unit plans a safe, collision-free approach path for any subsequent biopsy or treatment instruments. This path is then confirmed by the doctor on the touchscreen of the user interaction unit. S44: The diagnostic decision support module automatically retrieves enhanced 3D image data from the data storage unit. The feature extraction unit runs a segmentation algorithm to automatically outline the precise contours of suspicious lesions on the image, separating them from normal glandular tissue. Subsequently, the feature extraction algorithm processor and the neural network accelerator work together to quantify and calculate the morphological and textural features of the lesion area. The classification and judgment unit inputs the feature vector formed by combining the above features into a pre-trained classification model to calculate the probability value of the lesion being malignant. The report generation unit then automatically generates a structured diagnostic report, clearly listing the lesion's location, size, morphological description, and malignancy probability, and displays it on the screen for doctors to review, providing key evidence for treatment decisions. S45: Referring to the diagnostic report, the doctor decides to perform precise resection of the lesion with robot assistance. In the treatment planning unit interface, the doctor confirms the target lesion area. The planning software processor automatically delineates the target area to be resected based on the 3D model of the lesion. The parameter optimizer automatically recommends and calculates the cutting path, resection depth, and working parameters of the energy device based on tissue characteristics and preset safety margins. The safety monitoring unit simulates the entire planned resection path in a virtual environment, detects potential collision risks with surrounding important blood vessels and tissues, and verifies that all treatment parameters are within the absolute safety threshold. Finally, a complete and validated treatment plan is generated. S46: After the treatment plan is finalized by the doctor, the execution control unit begins operation. The robotic arm controller drives the robotic arm, carrying treatment tools such as a high-frequency electrosurgical unit, to precisely move to the planned skin entry point. During the robotic arm's movement, the encoder continuously provides feedback on joint angles to ensure positioning accuracy. After the resection begins, the force sensor monitors the force on the instrument's end in real time. The effect evaluation unit continuously acquires ultrasound images through a real-time image processor to dynamically assess the resection boundary and residual lesions. Simultaneously, the physiological parameter monitoring unit continuously collects the patient's vital signs data, such as ECG and blood oxygenation. All real-time data is integrated into the collaborative control center. S47: The adaptive adjustment unit of the real-time monitoring and feedback module continuously analyzes feedback data from force sensors, real-time images, and physiological parameters. When abnormally high tissue resistance is detected, which may affect the instrument's travel path, or when real-time ultrasound shows that the lesion has undergone slight displacement due to pressure, the adaptive controller will immediately send a fine-tuning command to the execution control unit according to a preset algorithm to dynamically adjust the propulsion force of the robotic arm or the spatial position of the instrument to ensure that the resection range is always consistent with the planned target area. Throughout the process, the safety monitoring unit is in the highest priority. Once any parameter touches the safety red line, the safety relay will be immediately triggered to stop all actions. S48: When the treatment effect evaluation unit determines through real-time imaging that the lesion has been completely removed and the resection boundary meets the preset requirements, the system sends an operation completion prompt to the doctor. After the doctor confirms, the system controls the robotic arm to withdraw the instrument and performs routine treatment on the surgical wound. The collaborative control center automatically archives all imaging data, positioning data, diagnostic reports, treatment planning parameters, and all key operation records during this diagnosis and treatment process, generating a complete digital surgical file for postoperative review and medical record management. Through integrated collaborative control, the system has successfully achieved an efficient and safe closed loop from accurate diagnosis to accurate treatment.

[0031] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A robotic collaborative control system for breast cancer diagnosis and treatment; characterized in that: It consists of the following modules: Image acquisition and processing module: used to acquire breast medical images and perform image processing and quality enhancement; Lesion localization and navigation module: used to accurately calculate the location of lesions based on image data and plan the robot's movement path; Diagnostic decision support module: used to analyze image features, assist doctors in classifying lesions as benign or malignant, and generate diagnostic reports; Treatment planning and execution module: used to plan treatment parameters based on diagnostic results and control the robot to perform treatment operations; Real-time monitoring and feedback module: used to monitor the patient's physiological parameters and treatment effects, and provide feedback to dynamically adjust the treatment; Collaborative Control Center Module: Used to integrate all modules.

2. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The image acquisition and processing module includes: A11: Image acquisition unit, including an ultrasonic probe, an optical camera, and an infrared sensor, used to acquire raw image data using multiple sensors; A12: Image processing unit, including image processing chip, GPU and memory module, used for denoising, enhancing and segmenting acquired images; A13: Data storage unit, including hard disk drives, solid-state drives and cloud storage interfaces, for storing raw and processed image data.

3. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The lesion localization and navigation module includes: A21: Positioning unit, including a laser rangefinder, inertial measurement unit, and GPS receiver, used to calculate the three-dimensional coordinates of the lesion through multi-sensor fusion; A22: Path planning unit, including path planning processor, obstacle avoidance sensors, and map building software, used to generate safe and efficient movement paths to avoid obstacles; A23: Navigation execution unit, including motor drivers, wheels or robotic arms and encoders, for controlling the robot to move along a planned path.

4. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 3, characterized in that: The lesion localization and navigation module operates by including the following steps: S11: The laser rangefinder performs self-calibration, the inertial measurement unit performs zero-position offset compensation, and the GPS receiver receives satellite signals to determine the robot's rough initial position; S12: A laser rangefinder scans the contours of the breast tissue, an inertial measurement unit monitors the robot's posture changes in real time, and an optical camera captures visual markers. The data from these three sources are then synchronized to the positioning unit processor. S13: Perform precise matrix transformation and registration on the lesion image coordinates provided by the image acquisition module, the robot's own coordinate system, and the patient's world coordinate system; S14: Based on the registered coordinate system, the precise coordinates of the lesion center point in three-dimensional space are calculated by fusing laser ranging and image depth data and using a triangulation algorithm; S15: The path planning processor plans a collision-free initial movement path in the environment model generated by the map building software based on the 3D coordinates of the lesion and the robot's current position; S16: As the robot moves along the initial path, the obstacle avoidance sensor continuously detects obstacles ahead. Once an obstacle is detected, path replanning is immediately triggered to generate a local detour path. S17: The navigation execution unit converts the final path command into pulse signals, which drive the motor driver to control the rotation speed and direction of the wheels or robotic arm, enabling the robot to move towards the target position; S18: The encoder provides real-time feedback on the actual rotation angle and displacement of the motor or robotic arm joint, compares it with the commanded position, and forms a closed-loop control to eliminate accumulated errors.

5. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The diagnostic decision support module includes: A31: Feature extraction unit, including a feature extraction algorithm processor, a neural network accelerator, and cache memory, for automatically extracting key features from images; A32: Classification and judgment unit, including a classifier chip, a decision tree processor, and a probability calculation unit, used for intelligent classification of lesions based on feature data; A33: Report generation unit, including report generation software, display and printer interfaces, for automatically generating structured diagnostic reports.

6. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 5, characterized in that: The diagnostic decision support module, when operating, includes the following steps: S21: Retrieve the enhanced breast image from the data storage unit of the image acquisition and processing module, and perform standardized scaling and grayscale normalization on it; S22: Using the segmentation algorithm in the feature extraction unit, the outline of the suspected lesion area is automatically delineated on the preprocessed image, separating it from normal tissue; S23: The feature extraction algorithm processor calculates the morphological parameters of the segmented lesion region, including but not limited to the degree of edge burrs, aspect ratio, area, and perimeter; S24: The neural network accelerator performs texture analysis on the lesion region, extracts the contrast, correlation, and entropy values ​​of its internal gray-level co-occurrence matrix, and calculates the average density of the region; S25: Combine all calculated morphological, texture, and density feature parameters in a predetermined order to form a multidimensional feature vector data set; S26: The constructed multidimensional feature vector is input into the pre-trained classifier chip, and the classifier is calculated based on the built-in decision tree or support vector machine model; S27: The probability calculation unit outputs the probability values ​​of whether the lesion is benign or malignant based on the calculation results of the classifier; S28: The report generation software automatically fills in the preset report template based on the classification probability and feature quantification results, including lesion location, size, feature description and classification suggestions.

7. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The treatment planning and execution module includes: A41: Treatment planning unit, including a planning software processor, parameter optimizer, and user input interface, for developing personalized treatment plans; A42: Execution control unit, including robot arm controller, treatment tool, and force sensor, for driving the treatment tool to perform precise operations; A43: Safety monitoring unit, including safety relays, emergency stop buttons and limit switches, for real-time monitoring of the treatment process.

8. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 7, characterized in that: The treatment planning and execution module, when in operation, includes the following steps: S31: The treatment planning unit receives the malignancy probability report and precise three-dimensional spatial coordinates of the lesion from the diagnostic decision support module and the lesion localization and navigation module, respectively; S32: The planning software processor accurately delineates the target volume and central target point that need to be removed or sampled on the 3D image based on the lesion coordinates and size; S33: Based on the treatment type, provide users with device selection and plan the optimal straight or curved path for the treatment tool from the skin entry point to the lesion target point; S34: The parameter optimizer automatically calculates the key parameters required for treatment based on the lesion characteristics and the planned path, including but not limited to the insertion depth of the biopsy needle, the power of the ablation energy, and the duration of action; S35: The safety monitoring unit performs collision detection between the planned path and the patient's anatomical structure model, and verifies whether the treatment parameters are within the preset safety threshold range; S36: The execution control unit sends a command to the robot arm controller, driving the robotic arm to move with the treatment tool to the planned skin entry point; S37: The robotic arm controller controls the robotic arm to insert the biopsy needle to the target depth according to the planned path and speed, or to attach the ablation probe to the lesion and apply the preset treatment energy; S39: The force sensor provides real-time feedback on the resistance encountered by the needle tip or probe, and the encoder continuously monitors the position of the robotic arm joints, stopping when the threshold is exceeded.

9. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The real-time monitoring and feedback module includes: A51: Physiological parameter monitoring unit, including an electrocardiogram sensor, a blood pressure monitor, and a blood oxygen saturation sensor, for real-time monitoring of the patient's vital signs; A52: Treatment efficacy assessment unit, including a real-time image processor, efficacy assessment algorithm, and feedback display, for assessing treatment progress through real-time imaging; A53: Adaptive adjustment unit, including adaptive controller, parameter adjuster and actuator, used to automatically adjust treatment parameters or pathways based on feedback data.

10. The integrated robot collaborative control system for breast cancer diagnosis and treatment according to claim 1, characterized in that: The collaborative control center module includes: A61: Data fusion unit, including a data fusion processor, communication bus, and data buffer, is used to fuse multi-source data and provide a unified view; A62: Task scheduling unit, including a task scheduler, real-time operating system, and priority manager, is used to coordinate the execution order of tasks in various modules and optimize resource allocation; A63: User interaction unit, including touch screen, keyboard and voice recognition module, for providing an intuitive interface.

Images